separate localization to extra package
* move localization Signed-off-by: Ke Xu <xuke@microsoft.com> * update reference Signed-off-by: Ke Xu <xuke@microsoft.com> * update reference Signed-off-by: Ke Xu <xuke@microsoft.com> * revert root dir Signed-off-by: Ke Xu <xuke@microsoft.com> * fix interpret Signed-off-by: Ke Xu <xuke@microsoft.com> * update fairlearn Signed-off-by: Ke Xu <xuke@microsoft.com> * add localization to publish Signed-off-by: Ke Xu <xuke@microsoft.com> * fix lint Signed-off-by: Ke Xu <xuke@microsoft.com> * update prettier ignore for rai static files Signed-off-by: Ke Xu <xuke@microsoft.com> * use english for not translated string Signed-off-by: Ke Xu <xuke@microsoft.com>
This commit is contained in:
xuke444
GitHub
Родитель
838fcd3179
Коммит
4b46a15e01
|
|
@ -53,7 +53,7 @@ jobs:
|
|||
strategy:
|
||||
matrix:
|
||||
node-version: [12.x]
|
||||
package: [core-ui, mlchartlib, fairness, interpret]
|
||||
package: [core-ui, mlchartlib, fairness, interpret, localization]
|
||||
|
||||
steps:
|
||||
- name: Use Node.js ${{ matrix.node-version }}
|
||||
|
|
|
|||
|
|
@ -3,4 +3,5 @@
|
|||
/dist
|
||||
/coverage
|
||||
_NOTICE.md
|
||||
raiwidgets/raiwidgets/templates/inlineDashboard.html
|
||||
raiwidgets/raiwidgets/templates/inlineDashboard.html
|
||||
raiwidgets/raiwidgets/static
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Třída {0}",
|
||||
"defaultFeatureNames": "Citlivý příznak {0}",
|
||||
"defaultSingleFeatureName": "Citlivý příznak",
|
||||
"defaultCustomMetricName": "Vlastní metrika {0}",
|
||||
"performanceTab": "Nestrannost ve výkonu",
|
||||
"opportunityTab": "Nestrannost v příležitosti",
|
||||
"modelComparisonTab": "Porovnání modelů",
|
||||
"tableTab": "Zobrazení podrobností",
|
||||
"dataSpecifications": "Statistiky dat",
|
||||
"attributes": "Atributy",
|
||||
"singleAttributeCount": "1 citlivý příznak",
|
||||
"attributesCount": "Počet citlivých příznaků: {0}",
|
||||
"instanceCount": "Počet instancí: {0}",
|
||||
"close": "Zavřít",
|
||||
"calculating": "Probíhá výpočet...",
|
||||
"performanceMetricLegacy": "Metrika výkonu",
|
||||
"errorOnInputs": "Chyba ve vstupu. Citlivé příznaky musí být v tuto chvíli kategorické hodnoty. Namapujte prosím hodnoty na intervalové kategorie a zkuste to znovu.",
|
||||
"Performance": {
|
||||
"header": "Jak chcete měřit výkon?",
|
||||
"modelMakes": "Značky modelů",
|
||||
"modelsMake": "Značka modelů",
|
||||
"body": "Vaše data obsahují několik popisků (celkem {0}) a predikce {2} {1}. Na základě této informace doporučujeme následující metriky. Vyberte prosím ze seznamu jednu metriku.",
|
||||
"binaryClassifier": "binární klasifikátor",
|
||||
"probabilisticRegressor": "Regresor probit",
|
||||
"regressor": "regresor",
|
||||
"binary": "binární",
|
||||
"continuous": "průběžné"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Nestrannost měřená ve smyslu nekonzistence",
|
||||
"body": "Metriky nekonzistence vyjadřují proměnlivost chování modelu mezi vybranými příznaky. Existují dva druhy metrik nekonzistence: další připravujeme..."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Dokumentace"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Zpět",
|
||||
"next": "Další"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Vítá vás",
|
||||
"fairnessDashboard": "Řídicí panel Fairness",
|
||||
"introBody": "Řídicí panel Fairness umožňuje posoudit kompromisy mezi výkonem a nestranností modelů.",
|
||||
"explanatoryStep": "Pokud chcete nastavit posouzení, musí se zadat citlivý příznak a metrika výkonu.",
|
||||
"getStarted": "Začínáme",
|
||||
"features": "Citlivé příznaky",
|
||||
"featuresInfo": "Citlivé příznaky se používají k rozdělení dat do skupin. Nestrannost modelu mezi těmito skupinami se měří jako metriky nekonzistence. Metriky nekonzistence vyjadřují, jak moc se chování modelu liší mezi jednotlivými skupinami.",
|
||||
"performance": "Metrika výkonu",
|
||||
"performanceInfo": "Metriky výkonu se používají k vyhodnocení celkové kvality modelu i kvality modelu v jednotlivých skupinách. Rozdíl mezi odlehlými hodnotami metriky výkonu ve skupinách se označuje za nekonzistenci ve výkonu."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Porovnání modelů",
|
||||
"howToRead": "Jak přečíst tento graf",
|
||||
"lower": "méně",
|
||||
"higher": "více",
|
||||
"howToReadText": "Tento graf reprezentuje každý z modelů {0} jako bod, který se dá vybrat. Osa x představuje {1}, kde {2} je lépe. Osa y představuje nekonzistenci a méně je lépe.",
|
||||
"insights": "Přehledy",
|
||||
"insightsText1": "Graf zobrazuje {0} a nekonzistenci modelů {1}.",
|
||||
"insightsText2": "{0} má rozsah od {1} do {2}. Rozsah nekonzistence je od {3} do {4}.",
|
||||
"insightsText3": "Nejpřesnější model dosahuje {0} z {1} a má nekonzistenci {2}.",
|
||||
"insightsText4": "Model s nejmenší nekonzistencí {2} dosahuje {0} z {1}.",
|
||||
"disparityInOutcomes": "Nekonzistence v predikcích",
|
||||
"disparityInPerformance": "Nekonzistence v {0}",
|
||||
"howToMeasureDisparity": "Jak se má nekonzistence měřit?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Model {0}",
|
||||
"title": "Nekonzistence ve výkonu",
|
||||
"globalPerformanceText": "Je celkové {0}",
|
||||
"performanceDisparityText": "Je nekonzistence v {0}",
|
||||
"editConfiguration": "Upravit konfiguraci",
|
||||
"backToComparisons": "Vícemodelové zobrazení",
|
||||
"outcomesTitle": "Nekonzistence v predikcích",
|
||||
"minTag": "Minimum",
|
||||
"maxTag": "Maximum",
|
||||
"groupLabel": "Podskupina",
|
||||
"underestimationError": "Podhodnocení",
|
||||
"underpredictionExplanation": "(predikováno = 0, skutečnost = 1)",
|
||||
"overpredictionExplanation": "(predikováno = 1, skutečnost = 0)",
|
||||
"overestimationError": "Nadhodnocení",
|
||||
"classificationOutcomesHowToRead": "Pruhový graf zobrazuje míru výběru v jednotlivých skupinách, tedy podíl bodů klasifikovaných jako 1.",
|
||||
"regressionOutcomesHowToRead": "Krabicové diagramy zobrazují distribuci chyb v jednotlivých skupinách. Jednotlivé datové body se zobrazují v popředí.",
|
||||
"classificationPerformanceHowToRead1": "Pruhový graf zobrazuje distribuci chyb v jednotlivých skupinách.",
|
||||
"classificationPerformanceHowToRead2": "Chyby jsou rozdělené na chyby nadhodnocení (predikuje se 1, když skutečný popisek je 0) a chyby podhodnocení (predikuje se 0, když skutečný popisek je 1).",
|
||||
"classificationPerformanceHowToRead3": "Uvedené míry se získají vydělením počtu chyb celkovou velikostí skupiny.",
|
||||
"probabilityPerformanceHowToRead1": "Pruhový graf zobrazuje střední absolutní chybu v jednotlivých skupinách rozdělenou na nadhodnocení a podhodnocení.",
|
||||
"probabilityPerformanceHowToRead2": "Pro každou ukázku změříme rozdíl mezi predikcí a popiskem. Pokud je kladný, nazveme ho nadhodnocením, a pokud je záporný, je to podhodnocení.",
|
||||
"probabilityPerformanceHowToRead3": "Uvedeme součet chyb nadhodnocení a součet chyb podhodnocení vydělené celkovou velikostí skupiny.",
|
||||
"regressionPerformanceHowToRead": "Chyba je rozdíl mezi predikcí a popiskem. Krabicové diagramy zobrazují distribuci chyb v jednotlivých skupinách. Jednotlivé datové body se zobrazují v popředí.",
|
||||
"distributionOfPredictions": "Distribuce predikcí",
|
||||
"distributionOfErrors": "Distribuce chyb",
|
||||
"tooltipPrediction": "Predikce: {0}",
|
||||
"tooltipError": "Chyba: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Se kterým příznakem byste chtěli vyhodnotit nestrannost modelu?",
|
||||
"body": "Nestrannost se vyhodnocuje ve smyslu nekonzistence chování modelů. Vaše data rozdělíme podle hodnot každého vybraného příznaku a vyhodnotíme, jak se mezi jednotlivými částmi liší metrika výkonu a predikce modelu.",
|
||||
"learnMore": "Další informace",
|
||||
"summaryCategoricalCount": "Tento příznak má následující počet jedinečných hodnot: {0}",
|
||||
"summaryNumericCount": "Tento číselný příznak má rozsah od {0} do {1} a je seskupený do {2} intervalů.",
|
||||
"showCategories": "Zobrazit vše",
|
||||
"hideCategories": "Sbalit",
|
||||
"categoriesOverflow": " a tento počet dalších kategorií: {0}",
|
||||
"editBinning": "Upravit skupiny",
|
||||
"subgroups": "Podskupiny"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Úspěšnost",
|
||||
"precisionScore": "Přesnost",
|
||||
"recallScore": "Úplnost",
|
||||
"zeroOneLoss": "Funkce zero-one loss",
|
||||
"specificityScore": "Skóre specifičnosti",
|
||||
"missRate": "Míra nesprávných predikcí",
|
||||
"falloutRate": "Míra vypuštění",
|
||||
"maxError": "Maximální chyba",
|
||||
"meanAbsoluteError": "Střední absolutní chyba",
|
||||
"meanSquaredError": " MSE (střední kvadratická chyba)",
|
||||
"meanSquaredLogError": "Střední kvadratická logaritmická chyba",
|
||||
"medianAbsoluteError": "Mediánová absolutní chyba",
|
||||
"average": "Průměrná predikce",
|
||||
"selectionRate": "Míra výběru",
|
||||
"overprediction": "Nadhodnocení",
|
||||
"underprediction": "Podhodnocení",
|
||||
"r2_score": "Skóre spolehlivosti R",
|
||||
"rms_error": "RMSE (odmocněná střední kvadratická chyba)",
|
||||
"auc": "Oblast pod křivkou ROC",
|
||||
"balancedRootMeanSquaredError": "Vyrovnané RMSE",
|
||||
"balancedAccuracy": "Vyrovnaná přesnost",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Část správně klasifikovaných datových bodů",
|
||||
"precisionDescription": "Část správně klasifikovaných datových bodů mezi body klasifikovanými jako 1",
|
||||
"recallDescription": "Část správně klasifikovaných datových bodů mezi body, jejichž skutečný popisek je 1. Alternativní názvy: míra pravdivě pozitivních predikcí, citlivost",
|
||||
"rmseDescription": "Druhá odmocnina průměru kvadratických chyb",
|
||||
"mseDescription": "Průměr kvadratických chyb",
|
||||
"meanAbsoluteErrorDescription": "Průměr absolutních hodnot chyb. Je robustnější vůči odlehlým hodnotám než MSE.",
|
||||
"r2Description": "Část odchylky v popiscích vysvětlené modelem",
|
||||
"aucDescription": "Kvalita predikcí při oddělování pozitivních příkladů od těch negativních, zobrazovaná jako skóre",
|
||||
"balancedRMSEDescription": "Váhy pozitivních a negativních příkladů se nastaví znovu, aby příklady měly stejnou celkovou váhu. Vhodné v situacích, kdy jsou základní data ve výrazné nerovnováze.",
|
||||
"balancedAccuracyDescription": "Váhy pozitivních a negativních příkladů se nastaví znovu, aby příklady měly stejnou celkovou váhu. Vhodné v situacích, kdy jsou základní data ve výrazné nerovnováze.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Nakonfigurovat intervaly",
|
||||
"makeCategorical": "Považovat za kategorické",
|
||||
"save": "Uložit",
|
||||
"cancel": "Zrušit",
|
||||
"numberOfBins": "Počet intervalů:",
|
||||
"categoryHeader": "Hodnoty intervalů:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Klasse \"{0}\"",
|
||||
"defaultFeatureNames": "Sensibles Feature {0}",
|
||||
"defaultSingleFeatureName": "Sensibles Feature",
|
||||
"defaultCustomMetricName": "Benutzerdefinierte Metrik {0}",
|
||||
"performanceTab": "Fairness bei der Leistung",
|
||||
"opportunityTab": "Fairness bei Chancen",
|
||||
"modelComparisonTab": "Modellvergleich",
|
||||
"tableTab": "Detailansicht",
|
||||
"dataSpecifications": "Datenstatistik",
|
||||
"attributes": "Attribute",
|
||||
"singleAttributeCount": "1 sensibles Feature",
|
||||
"attributesCount": "{0} sensible Features",
|
||||
"instanceCount": "{0} Instanzen",
|
||||
"close": "Schließen",
|
||||
"calculating": "Wird berechnet...",
|
||||
"performanceMetricLegacy": "Leistungsmetrik",
|
||||
"errorOnInputs": "Fehler bei der Eingabe. Sensible Features müssen zurzeit kategorische Werte sein. Ordnen Sie den Datengruppenkategorien Werte zu, und versuchen Sie es noch mal.",
|
||||
"Performance": {
|
||||
"header": "Wie möchten Sie die Leistung messen?",
|
||||
"modelMakes": "Ihrem Modell",
|
||||
"modelsMake": "Ihren Modellen",
|
||||
"body": "Ihre Daten enthalten {0} Beschriftungen, und Vorhersagen werden von {1} zur {2} erstellt. Basierend auf diesen Informationen empfehlen wir die folgenden Metriken. Wählen Sie eine Metrik aus der Liste aus.",
|
||||
"binaryClassifier": "binären Klassifizierung",
|
||||
"probabilisticRegressor": "Probit-Regression",
|
||||
"regressor": "Regression",
|
||||
"binary": "binäre",
|
||||
"continuous": "fortlaufende"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Fairness gemessen in Bezug auf die Abweichung",
|
||||
"body": "Abweichungsmetriken quantifizieren die Variation des Modellverhaltens für die ausgewählten Features. Es gibt zwei Arten von Abweichungsmetriken: Weitere Informationen folgen..."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Dokumentation"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Zurück",
|
||||
"next": "Weiter"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Willkommen beim",
|
||||
"fairnessDashboard": "Fairness-Dashboard",
|
||||
"introBody": "Mit dem Fairness-Dashboard können Sie Kompromisse zwischen Leistung und Fairness Ihrer Modelle bewerten.",
|
||||
"explanatoryStep": "Um die Bewertung einzurichten, müssen Sie ein sensibles Feature und eine Leistungsmetrik angeben.",
|
||||
"getStarted": "Erste Schritte",
|
||||
"features": "Sensible Features",
|
||||
"featuresInfo": "Anhand sensibler Features werden Ihre Daten in Gruppen unterteilt. Die Fairness Ihres Modells in diesen Gruppen wird mithilfe von Abweichungsmetriken gemessen. Diese drücken in Zahlen aus, inwieweit das Verhalten Ihres Modells in diesen Gruppen variiert.",
|
||||
"performance": "Leistungsmetrik",
|
||||
"performanceInfo": "Leistungsmetriken werden verwendet, um die Gesamtqualität Ihres Modells sowie die Qualität Ihres Modells in den einzelnen Gruppen auszuwerten. Der Unterschied zwischen den Extremwerten der Leistungsmetrik in den Gruppen wird als Leistungsabweichung gemeldet."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Modellvergleich",
|
||||
"howToRead": "Lesen dieses Diagramms",
|
||||
"lower": "ein geringerer Wert",
|
||||
"higher": "ein höherer Wert",
|
||||
"howToReadText": "Dieses Diagramm stellt jedes der {0} Modelle als auswählbaren Punkt dar. Die x-Achse stellt \"{1}\" dar, wobei {2} besser ist. Die y-Achse stellt die Abweichung dar, wobei ein geringerer Wert besser ist.",
|
||||
"insights": "Erkenntnisse",
|
||||
"insightsText1": "Das Diagramm zeigt \"{0}\" und die Abweichung bei {1} Modellen.",
|
||||
"insightsText2": "\"{0}\" liegt im Bereich zwischen {1} und {2}. Die Abweichung reicht von {3} bis {4}.",
|
||||
"insightsText3": "Das genaueste Modell erzielt für \"{0}\" einen Wert von {1} und eine Abweichung von {2}.",
|
||||
"insightsText4": "Das Modell mit der niedrigsten Abweichung erzielt für \"{0}\" einen Wert von {1} und eine Abweichung von {2}.",
|
||||
"disparityInOutcomes": "Abweichung in Vorhersagen",
|
||||
"disparityInPerformance": "Abweichung in {0}",
|
||||
"howToMeasureDisparity": "Wie soll die Abweichung gemessen werden?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Modell {0}",
|
||||
"title": "Leistungsabweichung",
|
||||
"globalPerformanceText": "Ist der Gesamtwert von \"{0}\"",
|
||||
"performanceDisparityText": "Ist die Abweichung in \"{0}\"",
|
||||
"editConfiguration": "Konfiguration bearbeiten",
|
||||
"backToComparisons": "Ansicht mit mehreren Modellen",
|
||||
"outcomesTitle": "Abweichung in Vorhersagen",
|
||||
"minTag": "Min.",
|
||||
"maxTag": "Max.",
|
||||
"groupLabel": "Untergruppe",
|
||||
"underestimationError": "Falsch negative Vorhersage",
|
||||
"underpredictionExplanation": "(vorhergesagt = 0, richtig = 1)",
|
||||
"overpredictionExplanation": "(vorhergesagt = 1, richtig = 0)",
|
||||
"overestimationError": "Falsch positive Vorhersage",
|
||||
"classificationOutcomesHowToRead": "Das Balkendiagramm zeigt die Auswahlrate der einzelnen Gruppen, d. h. den Anteil der als 1 klassifizierten Punkte.",
|
||||
"regressionOutcomesHowToRead": "Boxplots zeigen die Verteilung der Vorhersagen in den einzelnen Gruppen. Einzelne Datenpunkte werden darübergelegt.",
|
||||
"classificationPerformanceHowToRead1": "Das Balkendiagramm zeigt die Verteilung von Fehlern in den einzelnen Gruppen.",
|
||||
"classificationPerformanceHowToRead2": "Fehler werden in falsch positive Vorhersagen (Vorhersage von 1 bei einer Beschriftung von 0) und falsch negative Vorhersagen (Vorhersage von 0 bei einer Beschriftung von 1) unterteilt.",
|
||||
"classificationPerformanceHowToRead3": "Die gemeldeten Raten werden durch Teilen der Fehleranzahl durch die Gesamtgröße der Gruppe ermittelt.",
|
||||
"probabilityPerformanceHowToRead1": "Das Balkendiagramm zeigt die mittleren absoluten Fehler in den einzelnen Gruppen, aufgeteilt in falsch positive und falsch negative Vorhersagen.",
|
||||
"probabilityPerformanceHowToRead2": "In jedem Beispiel wird die Differenz zwischen der Vorhersage und der Beschriftung gemessen. Wenn diese positiv ist, wird dies als falsch positive Vorhersage bezeichnet. Ist sie negativ, handelt es sich um eine falsch negative Vorhersage.",
|
||||
"probabilityPerformanceHowToRead3": "Wir melden die Summe der Fehler durch falsch positive Vorhersagen und die Summe der Fehler durch falsch negative Vorhersagen, geteilt durch die Gesamtgröße der Gruppe.",
|
||||
"regressionPerformanceHowToRead": "Ein Fehler ist der Unterschied zwischen der Vorhersage und der Beschriftung. Boxplots zeigen die Verteilung von Fehlern in den einzelnen Gruppen. Einzelne Datenpunkte werden darübergelegt.",
|
||||
"distributionOfPredictions": "Verteilung der Vorhersagen",
|
||||
"distributionOfErrors": "Fehlerverteilung",
|
||||
"tooltipPrediction": "Vorhersage: {0}",
|
||||
"tooltipError": "Fehler: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Anhand welcher Features möchten Sie die Fairness Ihres Modells auswerten?",
|
||||
"body": "Fairness wird hinsichtlich der Abweichungen im Verhalten Ihres Modells ausgewertet. Wir teilen Ihre Daten entsprechend den Werten der einzelnen ausgewählten Features auf und bewerten, wie sich die Leistungsmetrik und die Vorhersagen Ihres Modells in den einzelnen Teilen unterscheiden.",
|
||||
"learnMore": "Weitere Informationen",
|
||||
"summaryCategoricalCount": "Dieses Feature weist {0} eindeutige Werte auf.",
|
||||
"summaryNumericCount": "Der Bereich dieses numerischen Features liegt zwischen {0} und {1} und ist in {2} Datengruppen unterteilt.",
|
||||
"showCategories": "Alle anzeigen",
|
||||
"hideCategories": "Reduzieren",
|
||||
"categoriesOverflow": " und {0} weitere Kategorien",
|
||||
"editBinning": "Gruppen bearbeiten",
|
||||
"subgroups": "Untergruppen"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Genauigkeit",
|
||||
"precisionScore": "Genauigkeit",
|
||||
"recallScore": "Wiedererkennung",
|
||||
"zeroOneLoss": "Null-Eins-Verlust",
|
||||
"specificityScore": "Spezifitätsscore",
|
||||
"missRate": "Fehlerrate",
|
||||
"falloutRate": "Ausfallrate",
|
||||
"maxError": "Maximaler Fehler",
|
||||
"meanAbsoluteError": "Mittlerer absoluter Fehler",
|
||||
"meanSquaredError": " MSE (mittlerer quadratischer Fehler)",
|
||||
"meanSquaredLogError": "Mittlerer quadratischer logarithmischer Fehler",
|
||||
"medianAbsoluteError": "Mittlere absolute Abweichung vom Median",
|
||||
"average": "Durchschnittliche Vorhersage",
|
||||
"selectionRate": "Auswahlrate",
|
||||
"overprediction": "Falsch positive Vorhersage",
|
||||
"underprediction": "Falsch negative Vorhersage",
|
||||
"r2_score": "R-quadratischer Score",
|
||||
"rms_error": "RMSE (Wurzel des mittleren quadratischen Fehlers)",
|
||||
"auc": "Fläche unter der ROC-Kurve",
|
||||
"balancedRootMeanSquaredError": "Ausgewogener RMSE",
|
||||
"balancedAccuracy": "Ausgewogene Genauigkeit",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Der Anteil der Datenpunkte, die korrekt klassifiziert werden.",
|
||||
"precisionDescription": "Der Anteil der Datenpunkte, die unter den als 1 klassifizierten Datenpunkten korrekt klassifiziert werden.",
|
||||
"recallDescription": "Der Anteil der Datenpunkte, die unter denjenigen korrekt klassifiziert werden, deren echte Beschriftung 1 lautet. Alternative Namen: True Positive-Rate, Sensitivität.",
|
||||
"rmseDescription": "Quadratwurzel des Durchschnitts von quadratischen Fehlern.",
|
||||
"mseDescription": "Der Durchschnitt aus quadratischen Fehlern.",
|
||||
"meanAbsoluteErrorDescription": "Der Durchschnitt der absoluten Fehlerwerte. Stabiler bei Ausreißern als MSE.",
|
||||
"r2Description": "Der Anteil der Varianz in den Beschriftungen, die vom Modell erklärt werden.",
|
||||
"aucDescription": "Die Qualität der Vorhersagen, dargestellt als Scores, beim Trennen positiver Beispiele von negativen Beispielen.",
|
||||
"balancedRMSEDescription": "Positive und negative Beispiele werden neu gewichtet, sodass sie insgesamt die gleiche Gewichtung aufweisen. Diese Option ist geeignet, wenn die zugrunde liegenden Daten hochgradig unausgewogen sind.",
|
||||
"balancedAccuracyDescription": "Positive und negative Beispiele werden neu gewichtet, sodass sie insgesamt die gleiche Gewichtung aufweisen. Diese Option ist geeignet, wenn die zugrunde liegenden Daten hochgradig unausgewogen sind.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Datengruppen konfigurieren",
|
||||
"makeCategorical": "Als kategorisch behandeln",
|
||||
"save": "Speichern",
|
||||
"cancel": "Abbrechen",
|
||||
"numberOfBins": "Datengruppenanzahl:",
|
||||
"categoryHeader": "Datengruppenwerte:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,213 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"_loremIpsum.comment": "DO NOT TRANSLATE. This is placeholder text the user will NOT see",
|
||||
"defaultClassNames": "Class {0}",
|
||||
"_defaultClassNames.comment": "models that output classes have this as the default class names when name are not given by the user",
|
||||
"defaultFeatureNames": "Sensitive feature {0}",
|
||||
"_defaultFeatureNames.comment": "models that output classes have this as the default class names when name are not given by the user",
|
||||
"defaultSingleFeatureName": "Sensitive feature",
|
||||
"defaultCustomMetricName": "Custom metric {0}",
|
||||
"_defaultCustomMetricName.comment": "prepend in front of the numerical index of the custom metric from the list of custom metrics",
|
||||
"performanceTab": "Fairness in Performance",
|
||||
"opportunityTab": "Fairness in Opportunity",
|
||||
"modelComparisonTab": "Model comparison",
|
||||
"tableTab": "Detail View",
|
||||
"dataSpecifications": "Data statistics",
|
||||
"attributes": "Attributes",
|
||||
"singleAttributeCount": "1 sensitive feature",
|
||||
"attributesCount": "{0} sensitive features",
|
||||
"_attributesCount.comment": "formatted string of the number of attributes",
|
||||
"instanceCount": "{0} instances",
|
||||
"_instanceCount.comment": "formatted string of the number of instances",
|
||||
"close": "Close",
|
||||
"done": "Done",
|
||||
"calculating": "Calculating...",
|
||||
"performanceMetricLegacy": "Performance metric",
|
||||
"sensitiveFeatures": "01 Sensitive features",
|
||||
"performanceMetric": "02 Performance metrics",
|
||||
"disparityMetric": "03 Disparity metrics",
|
||||
"errorOnInputs": "Error with input. Sensitive features must be categorical values at this time. Please map values to binned categories and retry.",
|
||||
"Performance": {
|
||||
"header": "How do you want to measure performance?",
|
||||
"modelMakes": "model makes",
|
||||
"modelsMake": "models make",
|
||||
"body": "Your data contains {0} labels and your {2} {1} predictions. Based on that information, we recommend the following metrics. Please select one metric from the list.",
|
||||
"_body.comment": "States whether labels are binary or continuous (0) and whether predictions are binary or continuous (1). (2) simply allows 'model(s) make' to be either singular or plural",
|
||||
"binaryClassifier": "binary classifier",
|
||||
"probabilisticRegressor": "probit regressor",
|
||||
"regressor": "regressor",
|
||||
"binary": "binary",
|
||||
"continuous": "continuous"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Fairness measured in terms of disparity",
|
||||
"bodyLegacy": "Disparity metrics quantify variation of your model's behavior across selected features. There are two kinds of disparity metrics: more to come....",
|
||||
"body": "Disparity metrics quantify variation of your model's behavior across selected features. There are four kinds of disparity metrics: more to come...."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Documentation"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Back",
|
||||
"next": "Next"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Welcome to the",
|
||||
"fairnessDashboard": "Fairness dashboard",
|
||||
"introBody": "The Fairness dashboard enables you to assess tradeoffs between performance and fairness of your models",
|
||||
"explanatoryStep": "To set up the assessment, you need to specify a sensitive feature and a performance metric.",
|
||||
"getStarted": "Get started",
|
||||
"features": "Sensitive features",
|
||||
"featuresInfo": "Sensitive features are used to split your data into groups. Fairness of your model across these groups is measured by disparity metrics. Disparity metrics quantify how much your model's behavior varies across these groups.",
|
||||
"performance": "Performance metric",
|
||||
"performanceInfo": "Performance metrics are used to evaluate the overall quality of your model as well as the quality of your model in each group. The difference between the extreme values of the performance metric across the groups is reported as the disparity in performance.",
|
||||
"parity": "Disparity metrics",
|
||||
"parityInfo": "Parity metrics are used to evaluate the overall quality of your model as well as the quality of your model in each group. The difference between the extreme values of accuracy is reported as the disparity in accuracy."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Model comparison",
|
||||
"howToRead": "How to read this chart",
|
||||
"lower": "lower",
|
||||
"higher": "higher",
|
||||
"howToReadText": "This chart represents each of the {0} models as a selectable point. The x-axis represents {1}, with {2} being better. The y-axis represents disparity, with lower being better.",
|
||||
"_howToReadText.comment": "Instructions for reading a chart. The number of models in the chart (0), the metric shown of the x-axix (1), and orientation for interpreting the x-axis",
|
||||
"insightsLegacy": "Insights",
|
||||
"insights": "Key Insights",
|
||||
"downloadReport": "Download report",
|
||||
"disparity": " The disparity ",
|
||||
"rangesFrom": " ranges from ",
|
||||
"to": " to ",
|
||||
"period": ". ",
|
||||
"introModalText": "Each model is a selectable point. Click or tap on model for it's full fairness assessment.",
|
||||
"helpModalText1": "The x-axis represents performance, with higher being better.",
|
||||
"helpModalText2": "The y-axis represents disparity, with lower being better.",
|
||||
"insightsText1": "The chart shows {0} and disparity of {1} models.",
|
||||
"_insightsText1.comment": "??? NOT FOUND IN SRC",
|
||||
"insightsText2": "{0} ranges from {1} to {2}. The disparity ranges from {3} to {4}.",
|
||||
"_insightsText2.comment": "The range (low to high) of the metric, and range (low to high) of the associated disparity for that metric",
|
||||
"insightsText3": "The most accurate model achieves {0} of {1} and a disparity of {2}.",
|
||||
"_insightsText3.comment": "For the most accurate model: The name of the performance measure (0), the formatted numerical value of the performance (1), and the formatted numerical value of the disparity (2)",
|
||||
"insightsText4": "The lowest-disparity model achieves {0} of {1} and a disparity of {2}.",
|
||||
"_insightsText4.comment": "For the lowest-disparity model: The name of the performance measure (0), the formatted numerical value of the performance (1), and the formatted numerical value of the disparity (2)",
|
||||
"disparityInOutcomes": "Disparity in predictions",
|
||||
"disparityInPerformance": "Disparity in {0}",
|
||||
"_disparityInPerformance.comment": "The name of the metric for which disparity is assessed",
|
||||
"howToMeasureDisparity": "How should disparity be measured?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Model {0}",
|
||||
"_modelName.comment": "The name of the model",
|
||||
"title": "Disparity in performance",
|
||||
"globalPerformanceText": "Is the overall {0}",
|
||||
"_globalPerformanceText.comment": "The title of the metric for which performance is being assessed",
|
||||
"performanceDisparityText": "Is the disparity in {0}",
|
||||
"_performanceDisparityText.comment": "The title of the metric for which performance is being assessed",
|
||||
"editConfiguration": "Edit configuration",
|
||||
"backToComparisonsLegacy": "Multimodel view",
|
||||
"backToComparisons": "Back to all models",
|
||||
"assessmentResults": "Assessment results for",
|
||||
"equalizedOddsDisparity": "Equalized odds disparity",
|
||||
"outcomesTitle": "Disparity in predictions",
|
||||
"expandSensitiveAttributes": "Expand sensitive attributes",
|
||||
"collapseSensitiveAttributes": "Collapse sensitive attributes",
|
||||
"minTag": "Min",
|
||||
"maxTag": "Max",
|
||||
"groupLabel": "Subgroup",
|
||||
"overallLabel": "Overall",
|
||||
"underestimationError": "Underprediction",
|
||||
"underpredictionExplanation": "(predicted = 0, true = 1)",
|
||||
"overpredictionExplanation": "(predicted = 1, true = 0)",
|
||||
"overestimationError": "Overprediction",
|
||||
"falseNegativeRate": "False negative rate",
|
||||
"falsePositiveRate": "False positive rate",
|
||||
"classificationOutcomesHowToRead": "The bar chart shows the selection rate in each group, meaning the fraction of points classified as 1.",
|
||||
"regressionOutcomesHowToRead": "Box plots show the distribution of predictions in each group. Individual data points are overlaid on top.",
|
||||
"classificationPerformanceHowToRead1": "The bar chart shows the distribution of errors in each group.",
|
||||
"classificationPerformanceHowToRead2": "Errors are split into overprediction errors (predicting 1 when the true label is 0), and underprediction errors (predicting 0 when the true label is 1).",
|
||||
"classificationPerformanceHowToRead3": "The reported rates are obtained by dividing the number of errors by the overall group size.",
|
||||
"probabilityPerformanceHowToRead1": "The bar chart shows mean absolute error in each group, split into overprediction and underprediction.",
|
||||
"probabilityPerformanceHowToRead2": "On each example, we measure the difference between the prediction and the label. If it is positive, we call it overprediction and if it is negative, we call it underprediction.",
|
||||
"probabilityPerformanceHowToRead3": "We report the sum of overprediction errors and the sum of underprediction errors divided by the overall group size.",
|
||||
"regressionPerformanceHowToRead": "Error is the difference between the prediction and the label. Box plots show the distribution of errors in each group. Individual data points are overlaid on top.",
|
||||
"distributionOfPredictions": "Distribution of predictions",
|
||||
"distributionOfErrors": "Distribution of errors",
|
||||
"tooltipPrediction": "Prediction: {0}",
|
||||
"_tooltipPrediction.comment": "Displays tooltip with the formatted numerical value of the prediction",
|
||||
"tooltipError": "Error: {0}",
|
||||
"_tooltipError.comment": "Displays tooltip with the formatted numerical value of the error"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Along which features would you like to evaluate your model's fairness?",
|
||||
"body": "Fairness is evaluated in terms of disparities in your model's behavior. We will split your data according to values of each selected feature, and evaluate how your model's performance metric and predictions differ across these splits.",
|
||||
"learnMore": "Learn more",
|
||||
"summaryCategoricalCount": "This feature has {0} unique values",
|
||||
"_summaryCategoricalCount.comment": "Number of unique values of the feature",
|
||||
"summaryNumericCount": "This numeric feature ranges from {0} to {1}, and is grouped into {2} bins.",
|
||||
"_summaryNumericalCount.comment": "The numerical range (low and high values) of the feature, and number of bin groups within that range",
|
||||
"showCategories": "Show all",
|
||||
"hideCategories": "Collapse",
|
||||
"categoriesOverflow": " and {0} additional categories",
|
||||
"_categoriesOverflow.comment": "??? NOT IN SRC - number of remaining additional categories",
|
||||
"editBinning": "Edit groups",
|
||||
"subgroups": "Subgroups"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Accuracy",
|
||||
"precisionScore": "Precision",
|
||||
"recallScore": "Recall",
|
||||
"zeroOneLoss": "Zero-one loss",
|
||||
"specificityScore": "Specificity score",
|
||||
"missRate": "Miss rate",
|
||||
"falloutRate": "Fallout rate",
|
||||
"maxError": "Max error",
|
||||
"meanAbsoluteError": "Mean absolute error",
|
||||
"meanSquaredError": " MSE (mean squared error)",
|
||||
"meanSquaredLogError": "Mean squared log error",
|
||||
"medianAbsoluteError": "Median absolute error",
|
||||
"average": "Average prediction",
|
||||
"selectionRate": "Selection rate",
|
||||
"overprediction": "Overprediction",
|
||||
"underprediction": "Underprediction",
|
||||
"falsePositiveRate": "False positive rate",
|
||||
"falseNegativeRate": "False negative rate",
|
||||
"r2_score": "R-squared score",
|
||||
"rms_error": "RMSE (root mean squared error)",
|
||||
"auc": "Area under ROC curve",
|
||||
"balancedRootMeanSquaredError": "Balanced RMSE",
|
||||
"balancedAccuracy": "Balanced accuracy",
|
||||
"f1Score": "F1-Score",
|
||||
"_f1Score.comment": "Data science terminology: https://en.wikipedia.org/wiki/F1_score",
|
||||
"logLoss": "Log Loss",
|
||||
"_logLoss.comment": "Data science terminology",
|
||||
"accuracyDescription": "The fraction of data points classified correctly.",
|
||||
"precisionDescription": "The fraction of data points classified correctly among those classified as 1.",
|
||||
"recallDescription": "The fraction of data points classified correctly among those whose true label is 1. Alternative names: true positive rate, sensitivity.",
|
||||
"rmseDescription": "Square root of the average of squared errors.",
|
||||
"mseDescription": "The average of squared errors.",
|
||||
"meanAbsoluteErrorDescription": "The average of absolute values of errors. More robust to outliers than MSE.",
|
||||
"r2Description": "The fraction of variance in the labels explained by the model.",
|
||||
"aucDescription": "The quality of the predictions, viewed as scores, in separating positive examples from negative examples.",
|
||||
"balancedRMSEDescription": "Positive and negative examples are reweighted to have equal total weight. Suitable if the underlying data is highly imbalanced.",
|
||||
"balancedAccuracyDescription": "Positive and negative examples are reweighted to have equal total weight. Suitable if the underlying data is highly imbalanced.",
|
||||
"falsePositiveRateDescription": "The fraction of data points classified incorrectly among those whose true label is 0.",
|
||||
"falseNegativeRateDescription": "The fraction of data points classified incorrectly among those whose true label is 1.",
|
||||
"parityDifference": "Parity difference",
|
||||
"parityDifferenceDescription": "Parity difference",
|
||||
"parityRatio": "Parity ratio",
|
||||
"parityRatioDescription": "Parity ratio",
|
||||
"errorRateDifference": "Error rate difference",
|
||||
"errorRateDifferenceDescription": "Error rate difference",
|
||||
"equalOpportunityDifference": "Equal opportunity difference",
|
||||
"equalOpportunityDifferenceDescription": "Equal opportunity difference",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Configure bins",
|
||||
"makeCategorical": "Treat as categorical",
|
||||
"save": "Save",
|
||||
"cancel": "Cancel",
|
||||
"numberOfBins": "Number of bins:",
|
||||
"categoryHeader": "Bin values:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Clase {0}",
|
||||
"defaultFeatureNames": "Característica confidencial {0}",
|
||||
"defaultSingleFeatureName": "Característica confidencial",
|
||||
"defaultCustomMetricName": "Métrica personalizada {0}",
|
||||
"performanceTab": "Equidad en rendimiento",
|
||||
"opportunityTab": "Equidad en oportunidades",
|
||||
"modelComparisonTab": "Comparación de modelos",
|
||||
"tableTab": "Vista detallada",
|
||||
"dataSpecifications": "Estadísticas de datos",
|
||||
"attributes": "Atributos",
|
||||
"singleAttributeCount": "1 característica confidencial",
|
||||
"attributesCount": "{0} características confidenciales",
|
||||
"instanceCount": "{0} instancias",
|
||||
"close": "Cerrar",
|
||||
"calculating": "Cálculo en curso...",
|
||||
"performanceMetricLegacy": "Métrica de rendimiento",
|
||||
"errorOnInputs": "Error con la entrada. Las características confidenciales deben ser valores de categorías en este momento. Asigne valores a las categorías con rangos y vuelva a intentarlo.",
|
||||
"Performancencence": {
|
||||
"header": "¿Cómo quiere medir el rendimiento?",
|
||||
"modelMakes": "modelo hace",
|
||||
"modelsMake": "modelos hacen",
|
||||
"body": "Los datos contienen etiquetas ({0}) y sus predicciones ({2} {1}). En función de esta información, se recomiendan las métricas siguientes. Seleccione una de la lista.",
|
||||
"binaryClassifier": "clasificador de elementos binarios",
|
||||
"probabilisticRegressor": "regresor probit",
|
||||
"regressor": "regresor",
|
||||
"binary": "binario",
|
||||
"continuous": "continuo"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Equidad medida en términos de disparidad",
|
||||
"body": "Las métricas de disparidad cuantifican la variación del comportamiento del modelo en las características seleccionadas. Hay dos tipos de métricas de disparidad, pero está previsto que haya más disponibles."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Documentación"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Atrás",
|
||||
"next": "Siguiente"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Le damos la bienvenida a",
|
||||
"fairnessDashboard": "Panel de Fairness",
|
||||
"introBody": "El panel de Fairness le permite evaluar las contrapartidas entre el rendimiento y la equidad de los modelos.",
|
||||
"explanatoryStep": "Para configurar la evaluación, debe especificar una característica confidencial y una métrica de rendimiento.",
|
||||
"getStarted": "Introducción",
|
||||
"features": "Características confidenciales",
|
||||
"featuresInfo": "Las características confidenciales se usan para dividir los datos en grupos. Las métricas de disparidad miden la equidad del modelo en estos grupos y cuantifican la variación del comportamiento del modelo en estos grupos.",
|
||||
"performance": "Métrica de rendimiento",
|
||||
"performanceInfo": "Las métricas de rendimiento se usan para evaluar la calidad general del modelo, así como la calidad del modelo en cada grupo. La diferencia entre los valores extremos de la métrica de rendimiento en los distintos grupos se notifica como una disparidad de rendimiento."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Comparación de modelos",
|
||||
"howToRead": "Cómo leer este gráfico",
|
||||
"lower": "menor",
|
||||
"higher": "mayor",
|
||||
"howToReadText": "Este gráfico representa cada uno de los {0} modelos como un punto seleccionable. El eje X representa {1}, donde un valor {2} es mejor; el eje Y representa la disparidad, donde un valor menor es mejor.",
|
||||
"insights": "Conclusiones",
|
||||
"insightsText1": "El gráfico muestra {0} y disparidad de los modelos de {1}.",
|
||||
"insightsText2": "{0} puede tener un valor de {1} a {2}. La disparidad puede tener un valor de {3} a {4}.",
|
||||
"insightsText3": "El modelo más preciso logra {0} de {1} y una disparidad de {2}.",
|
||||
"insightsText4": "El modelo de menor disparidad logra {0} de {1} y una disparidad de {2}.",
|
||||
"disparityInOutcomes": "Disparidad en predicciones",
|
||||
"disparityInPerformance": "Disparidad en {0}",
|
||||
"howToMeasureDisparity": "¿Cómo se deben medir las disparidades?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Modelo {0}",
|
||||
"title": "Disparidad en rendimiento",
|
||||
"globalPerformanceText": "Es el valor total de {0}",
|
||||
"performanceDisparityText": "Es la disparidad en {0}",
|
||||
"editConfiguration": "Editar configuración",
|
||||
"backToComparisons": "Vista multimodelo",
|
||||
"outcomesTitle": "Disparidad en predicciones",
|
||||
"minTag": "Mín.",
|
||||
"maxTag": "Máx.",
|
||||
"groupLabel": "Subgrupo",
|
||||
"underestimationError": "Infrapredicción",
|
||||
"underpredictionExplanation": "(predicción: 0, verdadero: 1)",
|
||||
"overpredictionExplanation": "(predicción: 1, verdadero: 0)",
|
||||
"overestimationError": "Sobrepredicción",
|
||||
"classificationOutcomesHowToRead": "El gráfico de barras muestra la probabilidad de selecciones de cada grupo, es decir, la fracción de puntos que se clasifica como 1.",
|
||||
"regressionOutcomesHowToRead": "Los diagramas de caja muestran la distribución de predicciones en cada grupo, mientras que los puntos de datos individuales están superpuestos.",
|
||||
"classificationPerformanceHowToRead1": "El gráfico de barras muestra la distribución de los errores en cada grupo.",
|
||||
"classificationPerformanceHowToRead2": "Los errores se dividen en errores de sobrepredicción (se predice 1 cuando la etiqueta verdadera es 0) y los errores de infrapredicción (se predice 0 cuando la etiqueta verdadera es 1).",
|
||||
"classificationPerformanceHowToRead3": "Las probabilidades indicadas se obtienen dividiendo el número de errores entre el tamaño total del grupo.",
|
||||
"probabilityPerformanceHowToRead1": "El gráfico de barras muestra un error absoluto medio en cada grupo, dividido en sobrepredicción e infrapredicción.",
|
||||
"probabilityPerformanceHowToRead2": "En cada ejemplo se mide la diferencia entre la predicción y la etiqueta. Si es positiva, se será una sobrepredicción y, si es negativa, será una infrapredicción.",
|
||||
"probabilityPerformanceHowToRead3": "Se informa de la suma de errores de sobrepredicción y de infrapredicción divididos entre el tamaño total del grupo.",
|
||||
"regressionPerformanceHowToRead": "El error es la diferencia entre la predicción y la etiqueta. Los diagramas de caja muestran la distribución de errores en cada grupo, mientras que los puntos de datos individuales están superpuestos.",
|
||||
"distributionOfPredictions": "Distribución de predicciones",
|
||||
"distributionOfErrors": "Distribución de errores",
|
||||
"tooltipPrediction": "Predicción: {0}",
|
||||
"tooltipError": "Error: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "¿Con respecto a qué características quiere evaluar la equidad de su modelo?",
|
||||
"body": "La equidad se evalúa en términos de disparidades en el comportamiento del modelo. Se dividirán los datos en función de los valores de las características seleccionadas y se evaluarán las diferencias entre las predicciones y las métricas de rendimiento del modelo de acuerdo con estas divisiones.",
|
||||
"learnMore": "Más información",
|
||||
"summaryCategoricalCount": "Esta característica tiene {0} valores únicos.",
|
||||
"summaryNumericCount": "Esta característica numérica abarca de {0} a {1} y se agrupa en {2} rangos.",
|
||||
"showCategories": "Mostrar todo",
|
||||
"hideCategories": "Contraer",
|
||||
"categoriesOverflow": " y {0} categorías adicionales",
|
||||
"editBinning": "Editar grupos",
|
||||
"subgroups": "Subgrupos"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Exactitud",
|
||||
"precisionScore": "Precisión",
|
||||
"recallScore": "Coincidencia",
|
||||
"zeroOneLoss": "Pérdida cero-uno",
|
||||
"specificityScore": "Puntuación de especificidad",
|
||||
"missRate": "Probabilidad de errores",
|
||||
"falloutRate": "Probabilidad de efectos colaterales",
|
||||
"maxError": "Error de máximo",
|
||||
"meanAbsoluteError": "Error absoluto de media",
|
||||
"meanSquaredError": " ECM (error cuadrático medio)",
|
||||
"meanSquaredLogError": "Error logarítmico de cuadrático medio",
|
||||
"medianAbsoluteError": "Error absoluto de mediana",
|
||||
"average": "Predicción promedio",
|
||||
"selectionRate": "Probabilidad de selecciones",
|
||||
"overprediction": "Sobrepredicción",
|
||||
"underprediction": "Infrapredicción",
|
||||
"r2_score": "Puntuación de R cuadrado",
|
||||
"rms_error": "RECM (raíz del error cuadrático medio)",
|
||||
"auc": "Área bajo la curva ROC",
|
||||
"balancedRootMeanSquaredError": "RECM equilibrado",
|
||||
"balancedAccuracy": "Exactitud equilibrada",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Fracción de puntos de datos clasificada correctamente.",
|
||||
"precisionDescription": "Fracción de puntos de datos clasificada correctamente entre los clasificados como 1.",
|
||||
"recallDescription": "Fracción de puntos de datos clasificada correctamente entre aquellos cuya etiqueta verdadera es 1. Nombres alternativos: índice de verdaderos positivos, confidencialidad.",
|
||||
"rmseDescription": "Raíz cuadrada del promedio de errores cuadráticos.",
|
||||
"mseDescription": "Promedio de errores cuadráticos.",
|
||||
"meanAbsoluteErrorDescription": "Promedio de valores absolutos de errores; más robusto a valores atípicos que ECM.",
|
||||
"r2Description": "Fracción de varianza de las etiquetas explicadas por el modelo.",
|
||||
"aucDescription": "Calidad de las predicciones, vistas como puntuaciones, en la separación de ejemplos positivos de ejemplos negativos.",
|
||||
"balancedRMSEDescription": "Los ejemplos positivos y negativos se vuelven a ponderar para que tengan la misma ponderación total. Esto es adecuado si los datos subyacentes están muy desequilibrados.",
|
||||
"balancedAccuracyDescription": "Los ejemplos positivos y negativos se vuelven a ponderar para que tengan la misma ponderación total. Esto es adecuado si los datos subyacentes están muy desequilibrados.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Configuración de rangos",
|
||||
"makeCategorical": "Tratar como valor categórico",
|
||||
"save": "Guardar",
|
||||
"cancel": "Cancelar",
|
||||
"numberOfBins": "Número de rangos:",
|
||||
"categoryHeader": "Valores de rangos:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Classe {0}",
|
||||
"defaultFeatureNames": "Caractéristique sensible {0}",
|
||||
"defaultSingleFeatureName": "Caractéristique sensible",
|
||||
"defaultCustomMetricName": "Métrique personnalisée {0}",
|
||||
"performanceTab": "Équité dans les performances",
|
||||
"opportunityTab": "Équité dans l'opportunité",
|
||||
"modelComparisonTab": "Comparaison de modèles",
|
||||
"tableTab": "Vue Détails",
|
||||
"dataSpecifications": "Statistiques des données",
|
||||
"attributes": "Attributs",
|
||||
"singleAttributeCount": "1 caractéristique sensible",
|
||||
"attributesCount": "{0} caractéristiques sensibles",
|
||||
"instanceCount": "{0} instances",
|
||||
"close": "Fermer",
|
||||
"calculating": "Calcul...",
|
||||
"performanceMetricLegacy": "Métrique de performances",
|
||||
"errorOnInputs": "Erreur d'entrée. Les caractéristiques sensibles doivent être des valeurs catégoriques pour l'instant. Mappez les valeurs à des catégories classées et réessayez.",
|
||||
"Performance": {
|
||||
"header": "Comment voulez-vous mesurer les performances ?",
|
||||
"modelMakes": "votre modèle fait",
|
||||
"modelsMake": "vos modèles font",
|
||||
"body": "Vos données contiennent des étiquettes de type « {0} » et {2} {1} prédictions. D'après ces informations, nous vous recommandons d'utiliser les métriques suivantes. Sélectionnez une métrique dans la liste.",
|
||||
"binaryClassifier": "classifieur binaire",
|
||||
"probabilisticRegressor": "régresseur probit",
|
||||
"regressor": "régresseur",
|
||||
"binary": "binaire",
|
||||
"continuous": "continu"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Équité mesurée en termes de disparité",
|
||||
"body": "Les métriques de disparité quantifient la variation du comportement de votre modèle sur les caractéristiques sélectionnées. Il existe deux types de métriques de disparité : d'autres sont à venir..."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Documentation"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Précédent",
|
||||
"next": "Suivant"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Bienvenue dans :",
|
||||
"fairnessDashboard": "Tableau de bord Fairness",
|
||||
"introBody": "Le tableau de bord Fairness vous permet d'évaluer les compromis entre les performances et l'équité de vos modèles",
|
||||
"explanatoryStep": "Pour configurer l'évaluation, vous devez spécifier une caractéristique sensible et une métrique de performances.",
|
||||
"getStarted": "Démarrer",
|
||||
"features": "Caractéristiques sensibles",
|
||||
"featuresInfo": "Les caractéristiques sensibles permettent de diviser vos données en groupes. L'équité de votre modèle dans ces groupes est évaluée par des métriques de disparité qui quantifient la variation du comportement de votre modèle dans ces groupes.",
|
||||
"performance": "Métrique de performances",
|
||||
"performanceInfo": "Les métriques de performances permettent d'évaluer la qualité globale de votre modèle ainsi que la qualité de votre modèle dans chaque groupe. La disparité des performances signalée est la différence entre les valeurs extrêmes de la métrique de performances dans les groupes."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Comparaison de modèles",
|
||||
"howToRead": "Comment lire ce graphique",
|
||||
"lower": "inférieure",
|
||||
"higher": "supérieure",
|
||||
"howToReadText": "Ce graphique représente chacun des modèles {0} sous la forme d'un point sélectionnable. L'axe X représente {1} (une valeur {2} étant meilleure). L'axe Y représente la disparité (une valeur inférieure étant meilleure).",
|
||||
"insights": "Insights",
|
||||
"insightsText1": "Le graphique montre {0} et la disparité de {1} modèles.",
|
||||
"insightsText2": "{0} est compris entre {1} et {2}. Les plages de disparité vont de {3} à {4}.",
|
||||
"insightsText3": "Le modèle le plus juste obtient {0} sur {1} et une disparité de {2}.",
|
||||
"insightsText4": "Le modèle de disparité le plus faible obtient {0} sur {1} et une disparité de {2}.",
|
||||
"disparityInOutcomes": "Disparité dans les prédictions",
|
||||
"disparityInPerformance": "Disparité dans {0}",
|
||||
"howToMeasureDisparity": "Comment la disparité doit-elle être mesurée ?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Modèle {0}",
|
||||
"title": "Disparité des performances",
|
||||
"globalPerformanceText": "Est le {0} d'ensemble",
|
||||
"performanceDisparityText": "Est la disparité dans {0}",
|
||||
"editConfiguration": "Modifier la configuration",
|
||||
"backToComparisons": "Vue multimodèle",
|
||||
"outcomesTitle": "Disparité dans les prédictions",
|
||||
"minTag": "Min",
|
||||
"maxTag": "Max",
|
||||
"groupLabel": "Sous-groupe",
|
||||
"underestimationError": "Sous-prédiction",
|
||||
"underpredictionExplanation": "(prédit = 0, true = 1)",
|
||||
"overpredictionExplanation": "(prédit = 1, true = 0)",
|
||||
"overestimationError": "Surprédiction",
|
||||
"classificationOutcomesHowToRead": "Le graphique à barres montre le taux de sélection dans chaque groupe, c'est-à-dire la fraction des points classifiés comme ayant la valeur 1.",
|
||||
"regressionOutcomesHowToRead": "Les diagrammes à surfaces montrent la répartition des prédictions dans chaque groupe. Les points de données individuels sont superposés.",
|
||||
"classificationPerformanceHowToRead1": "Le graphique à barres montre la répartition des erreurs dans chaque groupe.",
|
||||
"classificationPerformanceHowToRead2": "Les erreurs sont divisées en erreurs de surprédiction (prédire 1 quand l'étiquette true est 0) et en erreurs de sous-prédiction (prédire 0 quand l'étiquette true est 1).",
|
||||
"classificationPerformanceHowToRead3": "Les taux signalés sont obtenus en divisant le nombre d'erreurs par la taille totale du groupe.",
|
||||
"probabilityPerformanceHowToRead1": "Le graphique à barres montre l'erreur absolue moyenne dans chaque groupe, divisée en surprédiction et en sous-prédiction.",
|
||||
"probabilityPerformanceHowToRead2": "Dans chaque exemple, nous mesurons la différence entre la prédiction et l'étiquette. Si elle est positive, nous parlons de surprédiction ; si elle est négative, nous parlons de sous-prédiction.",
|
||||
"probabilityPerformanceHowToRead3": "Nous signalons la somme des erreurs de surprédiction et la somme des erreurs de sous-prédiction divisée par la taille totale du groupe.",
|
||||
"regressionPerformanceHowToRead": "L'erreur est la différence entre la prédiction et l'étiquette. Les diagrammes à surfaces montrent la répartition des erreurs dans chaque groupe. Les points de données individuels sont superposés.",
|
||||
"distributionOfPredictions": "Répartition des prédictions",
|
||||
"distributionOfErrors": "Distribution des erreurs",
|
||||
"tooltipPrediction": "Prédiction : {0}",
|
||||
"tooltipError": "Erreur : {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Selon quelles caractéristiques souhaitez-vous évaluer l'équité de votre modèle ?",
|
||||
"body": "L'équité est évaluée en termes de disparités dans le comportement de votre modèle. Nous allons diviser vos données en fonction des valeurs de chaque caractéristique sélectionnée, puis évaluer la manière dont la métrique et les prédictions de performance de votre modèle diffèrent dans ces divisions.",
|
||||
"learnMore": "En savoir plus",
|
||||
"summaryCategoricalCount": "Cette caractéristique a {0} valeurs uniques",
|
||||
"summaryNumericCount": "Cette caractéristique numérique, comprise entre {0} et {1}, est regroupée en {2} classes.",
|
||||
"showCategories": "Tout afficher",
|
||||
"hideCategories": "Réduire",
|
||||
"categoriesOverflow": " et {0} catégories supplémentaires",
|
||||
"editBinning": "Modifier les groupes",
|
||||
"subgroups": "Sous-groupes"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Justesse",
|
||||
"precisionScore": "Précision",
|
||||
"recallScore": "Rappeler",
|
||||
"zeroOneLoss": "Perte zéro-un",
|
||||
"specificityScore": "Score de spécificité",
|
||||
"missRate": "Taux d'échec",
|
||||
"falloutRate": "Taux de faux positifs",
|
||||
"maxError": "Erreur max.",
|
||||
"meanAbsoluteError": "Erreur absolue moyenne",
|
||||
"meanSquaredError": " MSE (erreur quadratique moyenne)",
|
||||
"meanSquaredLogError": "Erreur logarithmique quadratique moyenne",
|
||||
"medianAbsoluteError": "Erreur absolue médiane",
|
||||
"average": "Prédiction moyenne",
|
||||
"selectionRate": "Taux de sélection",
|
||||
"overprediction": "Surprédiction",
|
||||
"underprediction": "Sous-prédiction",
|
||||
"r2_score": "Score du coefficient de détermination",
|
||||
"rms_error": "RMSE (racine de l'erreur quadratique moyenne)",
|
||||
"auc": "Zone sous la courbe ROC",
|
||||
"balancedRootMeanSquaredError": "RMSE équilibrée",
|
||||
"balancedAccuracy": "Justesse équilibrée",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Fraction des points de données correctement classifiés.",
|
||||
"precisionDescription": "Fraction de points de données correctement classifiés parmi ceux classifiés comme ayant la valeur 1.",
|
||||
"recallDescription": "Fraction des points de données correctement classifiés parmi ceux dont l'étiquette true est 1. Autres noms : taux de vrais positifs, sensibilité.",
|
||||
"rmseDescription": "Racine carrée de la moyenne des erreurs quadratiques.",
|
||||
"mseDescription": "Moyenne des erreurs quadratiques.",
|
||||
"meanAbsoluteErrorDescription": "Moyenne des valeurs absolues des erreurs. Résiste mieux aux valeurs hors norme que MSE.",
|
||||
"r2Description": "Fraction de variance dans les étiquettes expliquée par le modèle.",
|
||||
"aucDescription": "Qualité des prédictions, présentées sous forme de scores, pour séparer les exemples positifs des exemples négatifs.",
|
||||
"balancedRMSEDescription": "Les exemples positifs et négatifs sont repondérés pour avoir un poids total égal. Convient si les données sous-jacentes sont fortement déséquilibrées.",
|
||||
"balancedAccuracyDescription": "Les exemples positifs et négatifs sont repondérés pour avoir un poids total égal. Convient si les données sous-jacentes sont fortement déséquilibrées.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Configurer les classes",
|
||||
"makeCategorical": "Considérer comme catégorique",
|
||||
"save": "Enregistrer",
|
||||
"cancel": "Annuler",
|
||||
"numberOfBins": "Nombre de classes :",
|
||||
"categoryHeader": "Valeurs de classe :"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "{0} osztály",
|
||||
"defaultFeatureNames": "Érzékeny jellemző: {0}",
|
||||
"defaultSingleFeatureName": "Érzékeny jellemző",
|
||||
"defaultCustomMetricName": "Egyéni metrika: {0}",
|
||||
"performanceTab": "Teljesítmény méltányossága",
|
||||
"opportunityTab": "Lehetőség méltányossága",
|
||||
"modelComparisonTab": "Modell-összehasonlítás",
|
||||
"tableTab": "Részletes nézet",
|
||||
"dataSpecifications": "Adatok statisztikái",
|
||||
"attributes": "Attribútumok",
|
||||
"singleAttributeCount": "1 érzékeny jellemző",
|
||||
"attributesCount": "{0} érzékeny jellemző",
|
||||
"instanceCount": "{0} példány",
|
||||
"close": "Bezárás",
|
||||
"calculating": "Számítás...",
|
||||
"performanceMetricLegacy": "Teljesítménymutató",
|
||||
"errorOnInputs": "Hiba történt a bemenettel. Az érzékeny jellemzőknek jelenleg kategorikus értékeknek kell lenniük. Rendelje hozzá az értékeket dobozolt kategóriákhoz, majd próbálja újra.",
|
||||
"Performance": {
|
||||
"header": "Hogyan szeretné mérni a teljesítményt?",
|
||||
"modelMakes": "modell végrehajt",
|
||||
"modelsMake": "modellek végrehajtanak",
|
||||
"body": "Az adatok {0} címkét és {2} {1} előrejelzést tartalmaznak. Ezen információ alapján a következő metrikákat javasoljuk. Válasszon ki egy metrikát a listából.",
|
||||
"binaryClassifier": "bináris osztályozó",
|
||||
"probabilisticRegressor": "probit magyarázó változója",
|
||||
"regressor": "magyarázó változó",
|
||||
"binary": "bináris",
|
||||
"continuous": "folyamatos"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Egyenlőtlenség alapján mért méltányosság",
|
||||
"body": "Az egyenlőtlenségi metrikák számszerűsítik a modell viselkedésének variációját a kiválasztott jellemzők tekintetében. Kétféle egyenlőtlenségi metrika létezik: hamarosan...."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Dokumentáció"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Vissza",
|
||||
"next": "Tovább"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Üdvözöli a",
|
||||
"fairnessDashboard": "Fairness irányítópult",
|
||||
"introBody": "A Fairness-irányítópult lehetővé teszi a modellek teljesítménye és a méltányossága közötti kompromisszumok értékelését",
|
||||
"explanatoryStep": "Az értékelés beállításához meg kell adnia egy érzékeny jellemzőt és egy teljesítménymutatót.",
|
||||
"getStarted": "Első lépések",
|
||||
"features": "Érzékeny jellemzők",
|
||||
"featuresInfo": "Az érzékeny jellemzők csoportokra osztják az adatokat. A modell csoportonkénti méltányossága az egyenlőtlenségi metrikák alapján van mérve. Az egyenlőtlenségi metrikák számszerűsítik, hogy a modell csoportonkénti viselkedése mennyire változatos.",
|
||||
"performance": "Teljesítménymutató",
|
||||
"performanceInfo": "A teljesítménymutató a modell összesített és csoportonkénti teljesítményének kiértékelésére szolgálnak. A teljesítménymutató szélsőséges értékei közötti csoportonkénti különbséget a rendszer teljesítménybeli egyenlőtlenségként fogja lejelenteni."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Modell-összehasonlítás",
|
||||
"howToRead": "A diagram értelmezése",
|
||||
"lower": "alacsonyabb",
|
||||
"higher": "magasabb",
|
||||
"howToReadText": "Ez a diagram a(z) {0} modellek mindegyikét kijelölhető pontként ábrázolja. Az x tengelyen {1} látható, ahol {2} jobb. Az Y tengely az egyenlőtlenséget mutatja, ahol az alacsonyabb érték jobb.",
|
||||
"insights": "Elemzések",
|
||||
"insightsText1": "A diagramon látható a(z) {0} és a(z) {1} modellek egyenlőtlensége.",
|
||||
"insightsText2": "A(z) {0} terjedelme: {1}–{2}. Az egyenlőtlenség terjedelme: {3}–{4}.",
|
||||
"insightsText3": "A legpontosabb modell {1}/{0} értéket és {2} egyenlőtlenséget ér el.",
|
||||
"insightsText4": "A legkisebb egyenlőtlenséges modell {1}/{0} értéket és {2} egyenlőtlenséget ér el.",
|
||||
"disparityInOutcomes": "Egyenlőtlenség az előrejelzések között",
|
||||
"disparityInPerformance": "Egyenlőtlenség itt: {0}",
|
||||
"howToMeasureDisparity": "Hogyan szeretné mérni az egyenlőtlenséget?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "{0} modell",
|
||||
"title": "Teljesítménybeli egyenlőtlenség",
|
||||
"globalPerformanceText": "A teljes {0}",
|
||||
"performanceDisparityText": "A(z) {0} egyenlőtlenségét jelenti",
|
||||
"editConfiguration": "Konfiguráció szerkesztése",
|
||||
"backToComparisons": "Többmodelles nézet",
|
||||
"outcomesTitle": "Egyenlőtlenség az előrejelzések között",
|
||||
"minTag": "Minimum",
|
||||
"maxTag": "Maximum",
|
||||
"groupLabel": "Alcsoport",
|
||||
"underestimationError": "Alábecslés",
|
||||
"underpredictionExplanation": "(előrejelzett = 0, igaz = 1)",
|
||||
"overpredictionExplanation": "(előrejelzett = 1, igaz = 0)",
|
||||
"overestimationError": "Túlbecslés",
|
||||
"classificationOutcomesHowToRead": "A sávdiagram az egyes csoportok kiválasztási arányát, vagyis az 1-es osztályozású ponthányadokat jeleníti meg.",
|
||||
"regressionOutcomesHowToRead": "A dobozdiagramok a csoportonként eloszló előrejelzéseket mutatják. Az egyéni adatpontok felül, átfedésben jelennek meg.",
|
||||
"classificationPerformanceHowToRead1": "A sávdiagram az eltérések eloszlását mutatja az egyes csoportokban.",
|
||||
"classificationPerformanceHowToRead2": "Az eltéréseket a rendszer túlbecslési eltérésekre (az előrejelzés 1, miközben a valós címke 0) és alábecslési eltérésekre (az előrejelzés 0, miközben a valós címke 1) bontja.",
|
||||
"classificationPerformanceHowToRead3": "A jelentett arányokat az eltérések számának a csoport teljes méretével való elosztása eredményezi.",
|
||||
"probabilityPerformanceHowToRead1": "A sávdiagram az egyes csoportokátlagos abszolút eltérését mutatja, túlbecslésre és alábecslésre osztva.",
|
||||
"probabilityPerformanceHowToRead2": "Minden példa esetében az előrejelzés és a címke közti különbséget mérjük. Ha a különbség pozitív, azt túlbecslésnek, ha negatív, azt alábecslésnek nevezzük.",
|
||||
"probabilityPerformanceHowToRead3": "A jelentésben a túlbecslések és az alábecslések eltérései összegének és a csoport teljes méretének hányadosa szerepel.",
|
||||
"regressionPerformanceHowToRead": "Az eltérés az előrejelzés és a címke közötti különbséget jelöli. A dobozdiagramok a csoportonkénti hibaeloszlást mutatják. Az egyéni adatpontok felül, átfedésben jelennek meg.",
|
||||
"distributionOfPredictions": "Előrejelzések eloszlása",
|
||||
"distributionOfErrors": "Eltérések eloszlása",
|
||||
"tooltipPrediction": "Előrejelzés: {0}",
|
||||
"tooltipError": "Eltérés: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Mely jellemzők mellett szeretné kiértékelni a modell méltányosságát?",
|
||||
"body": "A méltányosságot a rendszer a modell viselkedésében mutatkozó egyenlőtlenségek alapján értékeli ki. Az adatokat az egyes kiválasztott jellemzők értékei szerint osztjuk fel, és kiértékeljük, hogy a modell teljesítménymutatója és előrejelzései felosztásonként hogyan változik.",
|
||||
"learnMore": "További információ",
|
||||
"summaryCategoricalCount": "Ez a jellemző {0} egyedi értékeket tartalmaz",
|
||||
"summaryNumericCount": "Ez a numerikus jellemző {0} és {1} között mozog, és {2} dobozba van csoportosítva.",
|
||||
"showCategories": "Összes megjelenítése",
|
||||
"hideCategories": "Összecsukás",
|
||||
"categoriesOverflow": " és további {0} kategória",
|
||||
"editBinning": "Csoportok szerkesztése",
|
||||
"subgroups": "Alcsoportok"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Pontosság",
|
||||
"precisionScore": "Pontosság",
|
||||
"recallScore": "Felidézés",
|
||||
"zeroOneLoss": "Nulla–egy veszteség",
|
||||
"specificityScore": "Specifikussági pontszám",
|
||||
"missRate": "Kihagyási arány",
|
||||
"falloutRate": "Kiesési valószínűség",
|
||||
"maxError": "Maximális eltérés",
|
||||
"meanAbsoluteError": "Átlagos abszolút eltérés",
|
||||
"meanSquaredError": " MSE (átlagos négyzetes eltérés)",
|
||||
"meanSquaredLogError": "Átlagos négyzetes naplóeltérés",
|
||||
"medianAbsoluteError": "Medián abszolút eltérés",
|
||||
"average": "Átlagos előrejelzés",
|
||||
"selectionRate": "Kiválasztási arány",
|
||||
"overprediction": "Túlbecslés",
|
||||
"underprediction": "Alábecslés",
|
||||
"r2_score": "Négyzetgyök pontszám",
|
||||
"rms_error": "RMSE (gyökér-átlagos négyzetes eltérés)",
|
||||
"auc": "ROC-görbe alatti terület",
|
||||
"balancedRootMeanSquaredError": "Kiegyensúlyozott RMSE",
|
||||
"balancedAccuracy": "Kiegyensúlyozott pontosság",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "A megfelelően osztályozott adatpontok hányada.",
|
||||
"precisionDescription": "Az 1-es osztályozású adatpontok hányada, amelyek helyesen vannak osztályozva.",
|
||||
"recallDescription": "Az 1-es valós címkével rendelkező adatpontok hányada, amelyek helyesen vannak osztályozva. Alternatív megnevezés: valós pozitív arány, érzékenység.",
|
||||
"rmseDescription": "A négyzetes eltérések átlagának négyzetgyöke.",
|
||||
"mseDescription": "A négyzetes eltérések átlaga.",
|
||||
"meanAbsoluteErrorDescription": "Az eltérések abszolút értékeinek átlaga. Ellenállóbbak a kiugró értékekkel szemben, mint az MSE.",
|
||||
"r2Description": "A címkék modell által kifejtett varianciahányada.",
|
||||
"aucDescription": "Az előrejelzések pontszámként ábrázolt minősége a pozitív és a negatív példák elkülönítése terén.",
|
||||
"balancedRMSEDescription": "A rendszer újrasúlyozza a pozitív és negatív példákat, hogy a teljes súly egyenlő legyen. Akkor ajánlott, ha a mögöttes adatok nagy mértékben kiegyensúlyozatlanok.",
|
||||
"balancedAccuracyDescription": "A rendszer újrasúlyozza a pozitív és negatív példákat, hogy a teljes súly egyenlő legyen. Akkor ajánlott, ha a mögöttes adatok nagy mértékben kiegyensúlyozatlanok.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Dobozok konfigurálása",
|
||||
"makeCategorical": "Kezelés kategorikusként",
|
||||
"save": "Mentés",
|
||||
"cancel": "Mégse",
|
||||
"numberOfBins": "Dobozok száma:",
|
||||
"categoryHeader": "Dobozértékek:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Classe {0}",
|
||||
"defaultFeatureNames": "Caratteristica sensibile {0}",
|
||||
"defaultSingleFeatureName": "Caratteristica sensibile",
|
||||
"defaultCustomMetricName": "Metrica personalizzata {0}",
|
||||
"performanceTab": "Equità nelle prestazioni",
|
||||
"opportunityTab": "Equità nelle opportunità",
|
||||
"modelComparisonTab": "Confronto tra modelli",
|
||||
"tableTab": "Visualizzazione dettagli",
|
||||
"dataSpecifications": "Statistiche dati",
|
||||
"attributes": "Attributi",
|
||||
"singleAttributeCount": "1 caratteristica sensibile",
|
||||
"attributesCount": "{0} caratteristiche sensibili",
|
||||
"instanceCount": "{0} istanze",
|
||||
"close": "Chiudi",
|
||||
"calculating": "Calcolo...",
|
||||
"performanceMetricLegacy": "Metrica delle prestazioni",
|
||||
"errorOnInputs": "Errore con input. Le caratteristiche sensibili devono essere al momento valori categorici. Eseguire il mapping dei valori alle categorie di cui è stato effettuato il binning e riprovare.",
|
||||
"Performancencencence": {
|
||||
"header": "Come si vogliono misurare le prestazioni?",
|
||||
"modelMakes": "il modello crea",
|
||||
"modelsMake": "i modelli creano",
|
||||
"body": "I dati contengono etichette di tipo {0} e {2} stime con {1}. In base a queste informazioni, è consigliabile usare le metriche seguenti. Selezionare una metrica dall'elenco.",
|
||||
"binaryClassifier": "classificatore binario",
|
||||
"probabilisticRegressor": "regressore probit",
|
||||
"regressor": "regressore",
|
||||
"binary": "binario",
|
||||
"continuous": "continuo"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Equità misurata in termini di disparità",
|
||||
"body": "Le metriche di disparità quantificano la variazione del comportamento del modello in tutte le funzionalità selezionate. Esistono due tipi di metriche di disparità: altro da aggiungere...."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Documentazione"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Indietro",
|
||||
"next": "Avanti"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Benvenuti in",
|
||||
"fairnessDashboard": "Dashboard Fairness",
|
||||
"introBody": "Il dashboard Fairness consente di valutare i compromessi tra prestazioni ed equità dei modelli",
|
||||
"explanatoryStep": "Per configurare la valutazione, è necessario specificare una caratteristica sensibile e una metrica delle prestazioni.",
|
||||
"getStarted": "Attività iniziali",
|
||||
"features": "Caratteristiche sensibili",
|
||||
"featuresInfo": "Le caratteristiche sensibili vengono usate per suddividere i dati in gruppi. L'equità del modello in questi gruppi viene misurata con le metriche di disparità. Le metriche di disparità quantificano quanto varia il comportamento del modello in questi gruppi.",
|
||||
"performance": "Metrica delle prestazioni",
|
||||
"performanceInfo": "Le metriche delle prestazioni vengono usate per valutare la qualità complessiva del modello e la qualità del modello in ogni gruppo. La differenza tra i valori estremi della metrica delle prestazioni nei gruppi viene indicata come disparità nelle prestazioni."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Confronto tra modelli",
|
||||
"howToRead": "Come leggere questo grafico",
|
||||
"lower": "più basso",
|
||||
"higher": "più alto",
|
||||
"howToReadText": "Questo grafico rappresenta ciascuno dei {0} modelli come punto selezionabile. L'asse x rappresenta {1}, con un valore {2} come valore migliore. L'asse y rappresenta la disparità, con un valore più basso come valore migliore.",
|
||||
"insights": "Informazioni dettagliate",
|
||||
"insightsText1": "Il grafico mostra {0} e la disparità di {1} modelli.",
|
||||
"insightsText2": "{0} varia da {1} a {2}. La disparità varia da {3} a {4}.",
|
||||
"insightsText3": "Il modello più accurato raggiunge {0} di {1} e una disparità di {2}.",
|
||||
"insightsText4": "Il modello con disparità più bassa raggiunge {0} di {1} e una disparità di {2}.",
|
||||
"disparityInOutcomes": "Disparità nelle previsioni",
|
||||
"disparityInPerformance": "Disparità in {0}",
|
||||
"howToMeasureDisparity": "Come misurare la disparità?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Modello {0}",
|
||||
"title": "Disparità nelle prestazioni",
|
||||
"globalPerformanceText": "{0} in generale",
|
||||
"performanceDisparityText": "Disparità in {0}",
|
||||
"editConfiguration": "Modifica configurazione",
|
||||
"backToComparisons": "Vista multimodello",
|
||||
"outcomesTitle": "Disparità nelle previsioni",
|
||||
"minTag": "Min",
|
||||
"maxTag": "Max",
|
||||
"groupLabel": "Sottogruppo",
|
||||
"underestimationError": "Sottostima",
|
||||
"underpredictionExplanation": "(previsto = 0, vero = 1)",
|
||||
"overpredictionExplanation": "(previsto = 1, vero = 0)",
|
||||
"overestimationError": "Sovrastima",
|
||||
"classificationOutcomesHowToRead": "Il grafico a barre mostra il tasso di selezione in ogni gruppo, ovvero la frazione dei punti classificati come 1.",
|
||||
"regressionOutcomesHowToRead": "I box plot mostrano la distribuzione delle previsioni in ogni gruppo. I singoli punti dati sono sovrapposti in alto.",
|
||||
"classificationPerformanceHowToRead1": "Il grafico a barre mostra la distribuzione degli errori in ogni gruppo.",
|
||||
"classificationPerformanceHowToRead2": "Gli errori vengono suddivisi in errori di sovrastima (previsione di 1 quando l'etichetta vero è 0) ed errori di sottostima (previsione di 0 quando l'etichetta vero è 1).",
|
||||
"classificationPerformanceHowToRead3": "Le percentuali indicate si ottengono dividendo il numero di errori per le dimensioni complessive del gruppo.",
|
||||
"probabilityPerformanceHowToRead1": "Il grafico a barre mostra l'errore assoluto medio in ogni gruppo, suddiviso in sovrastima e sottostima.",
|
||||
"probabilityPerformanceHowToRead2": "In ogni esempio, viene misurata la differenza tra la previsione e l'etichetta. Se è positiva, viene definita sovrastima e se è negativa, viene definita sottostima.",
|
||||
"probabilityPerformanceHowToRead3": "Viene riportata la somma degli errori di sovrastima e la somma degli errori di sottostima divisi per le dimensioni complessive del gruppo.",
|
||||
"regressionPerformanceHowToRead": "Errore indica la differenza tra la previsione e l'etichetta. I box plot mostrano la distribuzione degli errori in ogni gruppo. I singoli punti dati sono sovrapposti in alto.",
|
||||
"distributionOfPredictions": "Distribuzione delle previsioni",
|
||||
"distributionOfErrors": "Distribuzione degli errori",
|
||||
"tooltipPrediction": "Previsione: {0}",
|
||||
"tooltipError": "Errore: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Con quali funzionalità si vuole valutare l'equità del modello?",
|
||||
"body": "L'equità viene valutata in termini di disparità nel comportamento del modello. I dati verranno suddivisi in base ai valori di ogni funzionalità selezionata e verrà valutato il modo in cui le previsioni e la metrica delle prestazioni del modello differiscono tra queste suddivisioni.",
|
||||
"learnMore": "Altre informazioni",
|
||||
"summaryCategoricalCount": "Questa funzionalità ha {0} valori univoci",
|
||||
"summaryNumericCount": "Questa funzionalità numerica varia da {0} a {1} ed è raggruppata in {2} bin.",
|
||||
"showCategories": "Mostra tutto",
|
||||
"hideCategories": "Comprimi",
|
||||
"categoriesOverflow": " e {0} categorie aggiuntive",
|
||||
"editBinning": "Modifica gruppi",
|
||||
"subgroups": "Sottogruppi"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Accuratezza",
|
||||
"precisionScore": "Precisione",
|
||||
"recallScore": "Richiama",
|
||||
"zeroOneLoss": "Perdita zero-uno",
|
||||
"specificityScore": "Punteggio di specificità",
|
||||
"missRate": "Frequenza di fallimento",
|
||||
"falloutRate": "Tasso di ricaduta",
|
||||
"maxError": "Errore massimo",
|
||||
"meanAbsoluteError": "Errore assoluto medio",
|
||||
"meanSquaredError": " MSE (errore quadratico medio)",
|
||||
"meanSquaredLogError": "Errore logaritmico quadratico medio",
|
||||
"medianAbsoluteError": "Errore assoluto mediano",
|
||||
"average": "Previsione media",
|
||||
"selectionRate": "Tasso di selezione",
|
||||
"overprediction": "Sovrastima",
|
||||
"underprediction": "Sottostima",
|
||||
"r2_score": "Punteggio R quadrato",
|
||||
"rms_error": "RMSE (radice dell'errore quadratico medio)",
|
||||
"auc": "Area sotto la curva ROC",
|
||||
"balancedRootMeanSquaredError": "RMSE bilanciato",
|
||||
"balancedAccuracy": "Accuratezza bilanciata",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Frazione dei punti dati classificati correttamente.",
|
||||
"precisionDescription": "Frazione dei punti dati classificati correttamente tra quelli classificati come 1.",
|
||||
"recallDescription": "Frazione dei punti dati classificati correttamente tra quelli la cui etichetta vero è 1. Nomi alternativi: percentuale di veri positivi, sensibilità.",
|
||||
"rmseDescription": "Radice quadrata della media degli errori quadratici.",
|
||||
"mseDescription": "Media degli errori quadratici.",
|
||||
"meanAbsoluteErrorDescription": "Media dei valori assoluti degli errori. Più affidabile per gli outlier rispetto a MSE.",
|
||||
"r2Description": "Frazione della varianza nelle etichette spiegata dal modello.",
|
||||
"aucDescription": "Qualità delle previsioni, visualizzate come punteggi, nel separare esempi positivi da esempi negativi.",
|
||||
"balancedRMSEDescription": "Gli esempi positivi e negativi vengono riponderati in modo da avere un peso totale uguale. Adatto se i dati sottostanti sono altamente sbilanciati.",
|
||||
"balancedAccuracyDescription": "Gli esempi positivi e negativi vengono riponderati in modo da avere un peso totale uguale. Adatto se i dati sottostanti sono altamente sbilanciati.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Configura bin",
|
||||
"makeCategorical": "Gestisci come categorie",
|
||||
"save": "Salva",
|
||||
"cancel": "Annulla",
|
||||
"numberOfBins": "Numero di bin:",
|
||||
"categoryHeader": "Valori bin:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "クラス {0}",
|
||||
"defaultFeatureNames": "重要な特徴量 {0}",
|
||||
"defaultSingleFeatureName": "重要な特徴量",
|
||||
"defaultCustomMetricName": "カスタム メトリック {0}",
|
||||
"performanceTab": "パフォーマンスの公平性",
|
||||
"opportunityTab": "機会の公平性",
|
||||
"modelComparisonTab": "モデルの比較",
|
||||
"tableTab": "詳細ビュー",
|
||||
"dataSpecifications": "データの統計情報",
|
||||
"attributes": "属性",
|
||||
"singleAttributeCount": "1 個の重要な特徴量",
|
||||
"attributesCount": "{0} 個の重要な特徴量",
|
||||
"instanceCount": "{0} 個のインスタンス",
|
||||
"close": "閉じる",
|
||||
"calculating": "計算しています...",
|
||||
"performanceMetricLegacy": "パフォーマンス メトリック",
|
||||
"errorOnInputs": "入力でエラーが発生しました。現在、重要な特徴量はカテゴリ値である必要があります。値をビン分割されたカテゴリにマップしてから、もう一度お試しください。",
|
||||
"Performancencencence": {
|
||||
"header": "パフォーマンスの測定方法",
|
||||
"modelMakes": "モデルの結果",
|
||||
"modelsMake": "モデルの結果",
|
||||
"body": "お客様のデータには、{0} ラベルと {2} {1} 予測が含まれています。この情報に基づいて、次のメトリックをお勧めします。リストからメトリックをいずれか 1 つ選択してください。",
|
||||
"binaryClassifier": "2 項分類子",
|
||||
"probabilisticRegressor": "プロビット リグレッサー",
|
||||
"regressor": "リグレッサー",
|
||||
"binary": "2 項",
|
||||
"continuous": "継続的"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "公平性を不均衡の観点から測定",
|
||||
"body": "不均衡メトリックは、選択した特徴量間のモデルの動作のバリエーションを定量化します。次の 2 種類の不均衡メトリックがあり、さらに増える予定です。"
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "ドキュメント"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "戻る",
|
||||
"next": "次へ"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "ようこそ:",
|
||||
"fairnessDashboard": "Fairness ダッシュボード",
|
||||
"introBody": "Fairness ダッシュボードを使用すると、モデルのパフォーマンスと公平性との間のトレードオフを評価できます",
|
||||
"explanatoryStep": "評価を設定するには、重要な特徴量とパフォーマンス メトリックを指定する必要があります。",
|
||||
"getStarted": "はじめに",
|
||||
"features": "重要な特徴量",
|
||||
"featuresInfo": "重要な特徴量は、データをグループに分割するために使用されます。これらのグループ間のモデルの公平性は、不均衡メトリックによって測定されます。不均衡メトリックは、これらのグループ間でモデルの動作がどれだけ変わるかを定量化します。",
|
||||
"performance": "パフォーマンス メトリック",
|
||||
"performanceInfo": "パフォーマンス メトリックは、モデルの全体的な品質と、各グループのモデルの品質を評価するために使用されます。グループ間のパフォーマンス メトリックの極値の差は、パフォーマンスの不均衡として報告されます。"
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "モデルの比較",
|
||||
"howToRead": "このグラフの読み取り方法",
|
||||
"lower": "値が小さい",
|
||||
"higher": "値が大きい",
|
||||
"howToReadText": "このグラフでは、各 {0} モデルを選択可能なポイントとして表します。X 軸は {1} を表し、{2} ほど優れています。Y 軸は、不均衡を表し、値が小さいほど優れています。",
|
||||
"insights": "分析情報",
|
||||
"insightsText1": "このグラフには、{1} モデルの {0} と不均衡が示されています。",
|
||||
"insightsText2": "{0} の範囲は {1} から {2} までです。不均衡の範囲は {3} から {4} までです。",
|
||||
"insightsText3": "最も正確なモデルは {1} の {0} と {2} の不均衡を実現します。",
|
||||
"insightsText4": "最小の不均衡モデルは、{1} の {0} と {2} の不均衡を実現します。",
|
||||
"disparityInOutcomes": "予測の不均衡",
|
||||
"disparityInPerformance": "{0} の不均衡",
|
||||
"howToMeasureDisparity": "不均衡の測定方法"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "モデル {0}",
|
||||
"title": "パフォーマンスの不均衡",
|
||||
"globalPerformanceText": "全体的な {0}",
|
||||
"performanceDisparityText": "{0} の不均衡",
|
||||
"editConfiguration": "構成の編集",
|
||||
"backToComparisons": "マルチモデル ビュー",
|
||||
"outcomesTitle": "予測の不均衡",
|
||||
"minTag": "最小値",
|
||||
"maxTag": "最大値",
|
||||
"groupLabel": "サブグループ",
|
||||
"underestimationError": "過小予測",
|
||||
"underpredictionExplanation": "(予測 = 0、true = 1)",
|
||||
"overpredictionExplanation": "(予測 = 1、true = 0)",
|
||||
"overestimationError": "過剰予測",
|
||||
"classificationOutcomesHowToRead": "横棒グラフは、各グループの選択率を示しており、ポイントの割合が 1 として分類されます。",
|
||||
"regressionOutcomesHowToRead": "ボックス プロットは、各グループの予測の分布を示します。個々のデータ ポイントは上に重ねて表示されています。",
|
||||
"classificationPerformanceHowToRead1": "横棒グラフは、各グループの誤差の分布を示します。",
|
||||
"classificationPerformanceHowToRead2": "誤差は過剰予測による誤差 (true ラベルが 0 の場合に 1 を予測) と過少予測による誤差 (true ラベルが 1 の場合に 0 を予測) に分割されます。",
|
||||
"classificationPerformanceHowToRead3": "報告率は、誤差の数をグループ全体のサイズで除算して取得します。",
|
||||
"probabilityPerformanceHowToRead1": "横棒グラフは、各グループの平均絶対誤差を示しており、過剰予測と過少予測に分割されます。",
|
||||
"probabilityPerformanceHowToRead2": "それぞれの例では、予測とラベルの差を測定します。正の場合は過剰予測と呼び、負の場合は過少予測と呼びます。",
|
||||
"probabilityPerformanceHowToRead3": "過剰予測の誤差の合計と過小予測の誤差の合計をグループ全体のサイズで除算したものが報告されます。",
|
||||
"regressionPerformanceHowToRead": "誤差は、予測とラベルの差です。ボックス プロットは、各グループの誤差の分布を示します。個々のデータ ポイントが上に重ねて表示されています。",
|
||||
"distributionOfPredictions": "予測の分布",
|
||||
"distributionOfErrors": "誤差の分布",
|
||||
"tooltipPrediction": "予測: {0}",
|
||||
"tooltipError": "誤差: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "モデルの公平性を評価する特徴量",
|
||||
"body": "公平性は、モデルの動作における不均衡の観点から評価されます。選択された各特徴量の値に従ってデータが分割され、モデルのパフォーマンス メトリックと予測がこれらの分割間でどのように異なるかが評価されます。",
|
||||
"learnMore": "詳細情報",
|
||||
"summaryCategoricalCount": "この特徴量には {0} 個の一意の値があります",
|
||||
"summaryNumericCount": "この数値の特徴量は {0} から {1} までの範囲に及び、{2} 個のビンにグループ化されています。",
|
||||
"showCategories": "すべて表示",
|
||||
"hideCategories": "折りたたむ",
|
||||
"categoriesOverflow": "と {0} 個の追加カテゴリ",
|
||||
"editBinning": "グループの編集",
|
||||
"subgroups": "サブグループ"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "精度",
|
||||
"precisionScore": "精度",
|
||||
"recallScore": "再現率",
|
||||
"zeroOneLoss": "0-1 損失",
|
||||
"specificityScore": "特異度スコア",
|
||||
"missRate": "見逃し率",
|
||||
"falloutRate": "フォールアウト率",
|
||||
"maxError": "誤差の最大値",
|
||||
"meanAbsoluteError": "平均絶対誤差",
|
||||
"meanSquaredError": " MSE (平均二乗誤差)",
|
||||
"meanSquaredLogError": "平均二乗対数誤差",
|
||||
"medianAbsoluteError": "中央絶対誤差",
|
||||
"average": "平均予測",
|
||||
"selectionRate": "選択率",
|
||||
"overprediction": "過剰予測",
|
||||
"underprediction": "過小予測",
|
||||
"r2_score": "決定係数スコア",
|
||||
"rms_error": "RMSE (平均平方二乗誤差)",
|
||||
"auc": "ROC 曲線下の面積",
|
||||
"balancedRootMeanSquaredError": "バランスの取れた RMSE",
|
||||
"balancedAccuracy": "バランス精度",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "正しく分類されたデータ ポイントの割合。",
|
||||
"precisionDescription": "1 に分類されたデータ ポイントの中で正しく分類されたデータ ポイントの割合。",
|
||||
"recallDescription": "true ラベルが 1 のデータ ポイントの中で正しく分類されたデータ ポイントの割合。代替名: 真陽性率、感度。",
|
||||
"rmseDescription": "平均二乗誤差の平方根。",
|
||||
"mseDescription": "二乗誤差の平均値。",
|
||||
"meanAbsoluteErrorDescription": "誤差の絶対値の平均値。MSE よりも外れ値に対して頑健です。",
|
||||
"r2Description": "モデルによって説明されているラベルの差異の割合。",
|
||||
"aucDescription": "負の例から正の例を分ける際の予測の品質をスコアとして見たものです。",
|
||||
"balancedRMSEDescription": "正および負の例は、総重量が等しくなるようにもう一度加重されます。基になるデータが極めて不均衡である場合に適しています。",
|
||||
"balancedAccuracyDescription": "正および負の例は、総重量が等しくなるようにもう一度加重されます。基になるデータが極めて不均衡である場合に適しています。",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "ビンの構成",
|
||||
"makeCategorical": "カテゴリ別として扱う",
|
||||
"save": "保存",
|
||||
"cancel": "キャンセル",
|
||||
"numberOfBins": "ビンの数:",
|
||||
"categoryHeader": "ビンの値:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "클래스 {0}",
|
||||
"defaultFeatureNames": "중요한 기능 {0}",
|
||||
"defaultSingleFeatureName": "중요한 기능",
|
||||
"defaultCustomMetricName": "사용자 지정 메트릭 {0}",
|
||||
"performanceTab": "성능의 공정성",
|
||||
"opportunityTab": "기회의 공정성",
|
||||
"modelComparisonTab": "모델 비교",
|
||||
"tableTab": "자세히 보기",
|
||||
"dataSpecifications": "데이터 통계",
|
||||
"attributes": "특성",
|
||||
"singleAttributeCount": "중요한 기능 1개",
|
||||
"attributesCount": "중요한 기능 {0}개",
|
||||
"instanceCount": "인스턴스 {0}개",
|
||||
"close": "닫기",
|
||||
"calculating": "계산하는 중...",
|
||||
"performanceMetricLegacy": "성능 메트릭",
|
||||
"errorOnInputs": "입력에 오류가 있습니다. 지금은 중요한 기능이 범주 값이어야 합니다. 값을 계급 구간으로 나뉜 범주에 매핑한 후 다시 시도하세요.",
|
||||
"Performancencencencence": {
|
||||
"header": "성능을 어떻게 측정하시겠습니까?",
|
||||
"modelMakes": "모델 생성",
|
||||
"modelsMake": "모델 생성",
|
||||
"body": "데이터에 {0} 레이블과 {2} {1} 예측이 포함되어 있습니다. 해당 정보를 기반으로, 다음과 같은 메트릭이 권장됩니다. 목록에서 메트릭 하나를 선택하세요.",
|
||||
"binaryClassifier": "이진 분류자",
|
||||
"probabilisticRegressor": "프로빗 회귀 변수",
|
||||
"regressor": "회귀 변수",
|
||||
"binary": "이진",
|
||||
"continuous": "연속"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "차이 측면에서 측정된 공정성",
|
||||
"body": "차이 메트릭은 선택한 기능에서 모델 동작의 변형을 정량화합니다. 차이 메트릭에는 두 가지 종류가 있으며 더 많은 종류가 제공될 예정입니다."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "설명서"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "뒤로",
|
||||
"next": "다음"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "환영",
|
||||
"fairnessDashboard": "Fairness 대시보드",
|
||||
"introBody": "Fairness 대시보드에서 모델의 성능과 공정성 간의 상쇄 관계를 평가할 수 있습니다.",
|
||||
"explanatoryStep": "평가를 설정하려면 중요한 기능과 성능 메트릭을 지정해야 합니다.",
|
||||
"getStarted": "시작",
|
||||
"features": "중요한 기능",
|
||||
"featuresInfo": "중요한 기능은 데이터를 그룹으로 분할하는 데 사용됩니다. 해당 그룹에서 모델의 공정성은 차이 메트릭으로 측정됩니다. 차이 메트릭은 모델의 동작이 해당 그룹에서 얼마나 다른지를 정량화합니다.",
|
||||
"performance": "성능 메트릭",
|
||||
"performanceInfo": "성능 메트릭은 모델의 전체 품질과 각 그룹의 모델 품질을 평가하는 데 사용됩니다. 그룹에서 성능 메트릭의 극단적인 값 간 차이는 성능의 차이로 보고됩니다."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "모델 비교",
|
||||
"howToRead": "이 차트를 읽는 방법",
|
||||
"lower": "더 낮은",
|
||||
"higher": "더 높은",
|
||||
"howToReadText": "이 차트에서는 {0}개 모델을 각각 선택 가능한 요소로 나타냅니다. x축은 {1}을(를) 나타내고 {2} 경우가 더 좋습니다. y축은 차이를 나타내고 더 낮은 경우가 더 좋습니다.",
|
||||
"insights": "인사이트",
|
||||
"insightsText1": "차트에는 {1}개 모델의 {0} 및 차이가 표시됩니다.",
|
||||
"insightsText2": "{0}의 범위는 {1}~{2}입니다. 차이 범위는 {3}~{4}입니다.",
|
||||
"insightsText3": "가장 정확한 모델은 {1}의 {0} 및 {2}의 차이를 실현합니다.",
|
||||
"insightsText4": "차이가 최저인 모델은 {1}의 {0} 및 {2}의 차이를 실현합니다.",
|
||||
"disparityInOutcomes": "예측의 차이",
|
||||
"disparityInPerformance": "{0}의 차이",
|
||||
"howToMeasureDisparity": "차이를 측정하는 방법은 무엇입니까?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "모델 {0}",
|
||||
"title": "성능의 차이",
|
||||
"globalPerformanceText": "전체 {0}인지 여부",
|
||||
"performanceDisparityText": "{0}의 차이인지 여부",
|
||||
"editConfiguration": "구성 편집",
|
||||
"backToComparisons": "다중 모델 보기",
|
||||
"outcomesTitle": "예측의 차이",
|
||||
"minTag": "최솟값",
|
||||
"maxTag": "최댓값",
|
||||
"groupLabel": "하위 그룹",
|
||||
"underestimationError": "과소 예측",
|
||||
"underpredictionExplanation": "(예측 = 0, 실제 = 1)",
|
||||
"overpredictionExplanation": "(예측 = 1, 실제 = 0)",
|
||||
"overestimationError": "과대 예측",
|
||||
"classificationOutcomesHowToRead": "가로 막대형 차트는 각 그룹의 선택 비율을 보여 주며, 1로 분류된 요소의 비율을 의미합니다.",
|
||||
"regressionOutcomesHowToRead": "상자 그림은 각 그룹의 예측 분포를 보여 줍니다. 개별 데이터 요소가 위에 오버레이되어 있습니다.",
|
||||
"classificationPerformanceHowToRead1": "가로 막대형 차트는 각 그룹의 오차 분포를 보여 줍니다.",
|
||||
"classificationPerformanceHowToRead2": "오차는 과대 예측 오차(실제 레이블이 0인 경우 1을 예측)와 과소 예측 오차(실제 레이블이 1인 경우 0을 예측)로 나뉩니다.",
|
||||
"classificationPerformanceHowToRead3": "보고된 비율은 오차 수를 전체 그룹 크기로 나눠 얻을 수 있습니다.",
|
||||
"probabilityPerformanceHowToRead1": "가로 막대형 차트는 각 그룹의 절대 평균 오차를 의미하며, 과대 예측과 과소 예측으로 나누어져 있습니다.",
|
||||
"probabilityPerformanceHowToRead2": "각 예제에서는 예측과 레이블 간의 차이를 측정합니다. 양수이면 과대 예측이라고 하고, 음수이면 과소 예측이라고 합니다.",
|
||||
"probabilityPerformanceHowToRead3": "과대 예측 오차의 합계와 과소 예측 오차의 합계를 전체 그룹 크기로 나눈 값을 보고합니다.",
|
||||
"regressionPerformanceHowToRead": "오차는 예측과 레이블 간의 차이입니다. 상자 그림은 각 그룹의 오차 분포를 보여 줍니다. 개별 데이터 요소가 위에 오버레이되어 있습니다.",
|
||||
"distributionOfPredictions": "예측 분포",
|
||||
"distributionOfErrors": "오차 분포",
|
||||
"tooltipPrediction": "예측: {0}",
|
||||
"tooltipError": "오차: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "어떤 기능에 따라 모델의 공정성을 평가하시겠습니까?",
|
||||
"body": "공정성은 모델 동작의 차이 측면에서 평가됩니다. 선택한 각 기능의 값에 따라 데이터를 분할하고 해당 분할에서 모델의 성능 메트릭 및 예측이 어떻게 다른지를 평가합니다.",
|
||||
"learnMore": "자세한 정보",
|
||||
"summaryCategoricalCount": "이 기능에는 고유 값이 {0}개 있습니다.",
|
||||
"summaryNumericCount": "이 숫자 기능은 범위가 {0}~{1}이며, {2}개 bin으로 그룹화되었습니다.",
|
||||
"showCategories": "모두 표시",
|
||||
"hideCategories": "축소",
|
||||
"categoriesOverflow": " 및 {0}개 추가 범주",
|
||||
"editBinning": "그룹 편집",
|
||||
"subgroups": "하위 그룹"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "정확도",
|
||||
"precisionScore": "정밀도",
|
||||
"recallScore": "회수",
|
||||
"zeroOneLoss": "제로-원 손실",
|
||||
"specificityScore": "특정성 점수",
|
||||
"missRate": "실패율",
|
||||
"falloutRate": "좋지 못한 결과 비율",
|
||||
"maxError": "최대 오차",
|
||||
"meanAbsoluteError": "절대 평균 오차",
|
||||
"meanSquaredError": " MSE(평균 제곱 오차)",
|
||||
"meanSquaredLogError": "평균 제곱 로그 오차",
|
||||
"medianAbsoluteError": "중앙값 절대 오차",
|
||||
"average": "평균 예측",
|
||||
"selectionRate": "선택 비율",
|
||||
"overprediction": "과대 예측",
|
||||
"underprediction": "과소 예측",
|
||||
"r2_score": "R-제곱 점수",
|
||||
"rms_error": "RMSE(제곱 평균 오차)",
|
||||
"auc": "ROC 곡선 아래의 영역",
|
||||
"balancedRootMeanSquaredError": "균형 잡힌 RMSE",
|
||||
"balancedAccuracy": "균형 정확도",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "제대로 분류된 데이터 요소의 비율입니다.",
|
||||
"precisionDescription": "1로 분류된 데이터 요소 중에서 제대로 분류된 데이터 요소의 비율입니다.",
|
||||
"recallDescription": "실제 레이블이 1인 데이터 요소 중 제대로 분류된 데이터 요소의 비율입니다. 대체 이름: 진양성 비율, 민감도.",
|
||||
"rmseDescription": "제곱 오차의 평균에 대한 제곱근입니다.",
|
||||
"mseDescription": "제곱 오차의 평균입니다.",
|
||||
"meanAbsoluteErrorDescription": "오차의 절대값에 대한 평균입니다. MSE보다 이상값에 영향을 덜 받습니다.",
|
||||
"r2Description": "모델에서 설명하는 레이블의 분산 비율입니다.",
|
||||
"aucDescription": "올바른 예제와 잘못된 예제를 구분하여 점수로 표시한 예측의 품질입니다.",
|
||||
"balancedRMSEDescription": "올바른 예제와 잘못된 예제는 같은 총 가중치를 갖도록 가중치가 다시 지정됩니다. 기본 데이터가 매우 불균형한 경우 적합합니다.",
|
||||
"balancedAccuracyDescription": "올바른 예제와 잘못된 예제는 같은 총 가중치를 갖도록 가중치가 다시 지정됩니다. 기본 데이터가 매우 불균형한 경우 적합합니다.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "bin 구성",
|
||||
"makeCategorical": "범주로 처리",
|
||||
"save": "저장",
|
||||
"cancel": "취소",
|
||||
"numberOfBins": "bin 수:",
|
||||
"categoryHeader": "bin 값:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Klasse {0}",
|
||||
"defaultFeatureNames": "Gevoelige functie {0}",
|
||||
"defaultSingleFeatureName": "Gevoelige functie",
|
||||
"defaultCustomMetricName": "Aangepaste metrische gegevens {0}",
|
||||
"performanceTab": "Verdeling in prestaties",
|
||||
"opportunityTab": "Verdeling in verkoopkans",
|
||||
"modelComparisonTab": "Modelvergelijking",
|
||||
"tableTab": "Detailweergave",
|
||||
"dataSpecifications": "Gegevensstatistieken",
|
||||
"attributes": "Kenmerken",
|
||||
"singleAttributeCount": "1 gevoelige functie",
|
||||
"attributesCount": "{0} gevoelige functies",
|
||||
"instanceCount": "{0} exemplaren",
|
||||
"close": "Sluiten",
|
||||
"calculating": "Berekenen...",
|
||||
"performanceMetricLegacy": "Metrische prestatiegegevens",
|
||||
"errorOnInputs": "Fout bij invoer. Gevoelige functies moeten op dit moment categorisch zijn. Wijs waarden toe aan categorieën in bins en probeer het opnieuw.",
|
||||
"Performancencencencence": {
|
||||
"header": "Hoe wilt u de prestaties meten?",
|
||||
"modelMakes": "model maakt",
|
||||
"modelsMake": "modellen maken",
|
||||
"body": "Uw gegevens bevatten {0} labels en uw {2} {1} voorspellingen. Op basis van deze informatie raden we de volgende metrische gegevens aan. Selecteer één waarde in de lijst.",
|
||||
"binaryClassifier": "binaire classificatie",
|
||||
"probabilisticRegressor": "probit regressor",
|
||||
"regressor": "regressor",
|
||||
"binary": "binair",
|
||||
"continuous": "doorlopend"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Verdeling gemeten in termen van verschil",
|
||||
"body": "Met de verschilwaarden wordt de variatie van het gedrag van uw model op geselecteerde functies gekwantificeerd. Er zijn twee soorten verschilwaarden: wordt vervolgd...."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Documentatie"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Vorige",
|
||||
"next": "Volgende"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Welkom bij",
|
||||
"fairnessDashboard": "Fairness-dashboard",
|
||||
"introBody": "Met het Fairness-dashboard kunt u de compromissen tussen prestaties en verdeling van uw modellen evalueren",
|
||||
"explanatoryStep": "Als u de evaluatie wilt instellen, moet u een gevoelige functie en metrische prestatiegegevens opgeven.",
|
||||
"getStarted": "Aan de slag",
|
||||
"features": "Gevoelige functies",
|
||||
"featuresInfo": "Gevoelige functies worden gebruikt om uw gegevens te splitsen in groepen. De verdeling van uw model over deze groepen wordt gemeten door de verschilwaarden te meten. Met de verschilwaarden wordt bepaald hoeveel het gedrag van uw model tussen deze groepen kan variëren.",
|
||||
"performance": "Metrische prestatiegegevens",
|
||||
"performanceInfo": "Metrische prestatiegegevens worden gebruikt om de algehele kwaliteit van uw model en de kwaliteit van uw model in elke groep te evalueren. Het verschil tussen de uitersten van de metrische prestatiegegevens in de groepen wordt gerapporteerd als de prestatieverhouding."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Modelvergelijking",
|
||||
"howToRead": "Hoe u deze grafiek moet lezen",
|
||||
"lower": "lager",
|
||||
"higher": "hoger",
|
||||
"howToReadText": "Deze grafiek toont elk van de {0} modellen als een selecteerbaar punt. De x-as toont {1}, waarbij {2} beter is. De y-as toont het verschil, waarbij lager beter is.",
|
||||
"insights": "Inzichten",
|
||||
"insightsText1": "De grafiek toont {0} en het verschil tussen {1} modellen.",
|
||||
"insightsText2": "{0} loopt van {1} tot {2}. Het verschil loopt van {3} tot {4}.",
|
||||
"insightsText3": "Het meest nauwkeurige model bereikt {0} van {1} en een verschil van {2}.",
|
||||
"insightsText4": "Het model met het kleinste verschil bereikt {0} van {1} en een verschil van {2}.",
|
||||
"disparityInOutcomes": "Verschil in voorspellingen",
|
||||
"disparityInPerformance": "Verschil in {0}",
|
||||
"howToMeasureDisparity": "Hoe moet het verschil worden gemeten?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Model {0}",
|
||||
"title": "Verschil in prestaties",
|
||||
"globalPerformanceText": "Is de algemene {0}",
|
||||
"performanceDisparityText": "Is het verschil in {0}",
|
||||
"editConfiguration": "Configuratie bewerken",
|
||||
"backToComparisons": "Weergave met meerdere modellen",
|
||||
"outcomesTitle": "Verschil in voorspellingen",
|
||||
"minTag": "Minimum",
|
||||
"maxTag": "Maximum",
|
||||
"groupLabel": "Subgroep",
|
||||
"underestimationError": "Ondervoorspelling",
|
||||
"underpredictionExplanation": "(voorspeld = 0, werkelijk = 1)",
|
||||
"overpredictionExplanation": "(voorspeld = 1, werkelijk = 0)",
|
||||
"overestimationError": "Overvoorspelling",
|
||||
"classificationOutcomesHowToRead": "Het staafdiagram toont het selectiepercentage in elke groep, dat wil zeggen het aantal punten dat wordt geclassificeerd als 1.",
|
||||
"regressionOutcomesHowToRead": "In boxplots wordt de distributie van voorspellingen in elke groep weergegeven. Afzonderlijke gegevenspunten worden hier overheen gelegd.",
|
||||
"classificationPerformanceHowToRead1": "In het staafdiagram wordt de distributie van fouten in elke groep weergegeven.",
|
||||
"classificationPerformanceHowToRead2": "Fouten zijn opgesplitst in fouten met overvoorspelling (voorspelling 1 wanneer het werkelijke label 0 is) en fouten met ondervoorspelling (voorspelling 0 wanneer het werkelijke label 1 is).",
|
||||
"classificationPerformanceHowToRead3": "De gerapporteerde aantallen worden verkregen door het aantal fouten te delen door de totale groepsgrootte.",
|
||||
"probabilityPerformanceHowToRead1": "Het staafdiagram laat de gemiddelde absolute fout in elke groep zien, gesplitst naar overvoorspelling en ondervoorspellinig.",
|
||||
"probabilityPerformanceHowToRead2": "In elk voorbeeld wordt het verschil gemeten tussen de voorspelling en het label. Als het positief is, wordt het overvoorspelling genoemd en als het negatief is, wordt het ondervoorspelling genoemd.",
|
||||
"probabilityPerformanceHowToRead3": "We rapporteren de som van de fouten met overvoorspelling en de som van de fouten met ondervoorspelling, gedeeld door de totale groepsgrootte.",
|
||||
"regressionPerformanceHowToRead": "De fout is het verschil tussen de voorspelling en het label. In boxplots wordt de distributie van fouten in elke groep weergegeven. Afzonderlijke gegevenspunten worden hier overheen gelegd.",
|
||||
"distributionOfPredictions": "Distributie van voorspellingen",
|
||||
"distributionOfErrors": "Distributie van fouten",
|
||||
"tooltipPrediction": "Voorspelling: {0}",
|
||||
"tooltipError": "Fout: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Voor welke functies wilt u de verdeling van uw model evalueren?",
|
||||
"body": "Verdeling wordt geëvalueerd in termen van verschillen in het gedrag van uw model. Uw gegevens worden gesplitst op basis van de waarden van elke geselecteerde functie en er wordt geëvalueerd hoe deze splitsingen verschillen in de prestatiegegevens en voorspellingen van uw model.",
|
||||
"learnMore": "Meer informatie",
|
||||
"summaryCategoricalCount": "Deze functie heeft {0} unieke waarden",
|
||||
"summaryNumericCount": "Deze numerieke functie heeft een bereik van {0} tot {1} en is gegroepeerd in {2} bins.",
|
||||
"showCategories": "Alles weergeven",
|
||||
"hideCategories": "Samenvouwen",
|
||||
"categoriesOverflow": " en {0} aanvullende categorieën",
|
||||
"editBinning": "Groepen bewerken",
|
||||
"subgroups": "Subgroepen"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Nauwkeurigheid",
|
||||
"precisionScore": "Precisie",
|
||||
"recallScore": "Terughalen",
|
||||
"zeroOneLoss": "Nul-één verlies",
|
||||
"specificityScore": "Score voor specificiteit",
|
||||
"missRate": "Aantal missers",
|
||||
"falloutRate": "Uitvalpercentage",
|
||||
"maxError": "Maximumfout",
|
||||
"meanAbsoluteError": "Gemiddelde absolute fout",
|
||||
"meanSquaredError": " MSE (gemiddelde fout in het kwadraat)",
|
||||
"meanSquaredLogError": "Gemiddelde logboekfout in kwadraat",
|
||||
"medianAbsoluteError": "Mediaan absolute fout",
|
||||
"average": "Gemiddelde voorspelling",
|
||||
"selectionRate": "Selectiesnelheid",
|
||||
"overprediction": "Overvoorspelling",
|
||||
"underprediction": "Ondervoorspelling",
|
||||
"r2_score": "R-kwadraatscore",
|
||||
"rms_error": "RMSE (wortel gemiddelde kwadratische fout)",
|
||||
"auc": "Gebied onder ROC-curve",
|
||||
"balancedRootMeanSquaredError": "Verdeelde RMSE",
|
||||
"balancedAccuracy": "Verdeelde nauwkeurigheid",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Het deel van de gegevenspunten dat op de juiste wijze is ingedeeld.",
|
||||
"precisionDescription": "Het deel van de gegevenspunten dat correct is geclassificeerd van alle gegevenspunten die als 1 zijn geclassificeerd.",
|
||||
"recallDescription": "Het deel van de gegevenspunten dat correct is geclassificeerd van alle gegevenspunten waarvan het werkelijke label 1 is. Alternatieve benamingen: frequentie terecht-positieven, gevoeligheid.",
|
||||
"rmseDescription": "Vierkantswortel van het gemiddelde van de kwadratische fouten.",
|
||||
"mseDescription": "Het gemiddelde van de kwadratische fouten.",
|
||||
"meanAbsoluteErrorDescription": "Het gemiddelde van de absolute waarden van fouten. Robuuster voor uitschieters dan MSE.",
|
||||
"r2Description": "Het deel afwijking in de labels die worden uitgelegd door het model.",
|
||||
"aucDescription": "De kwaliteit van de voorspellingen, die worden weergegeven als scores, in het scheiden van positieve voorbeelden van negatieve voorbeelden.",
|
||||
"balancedRMSEDescription": "Positieve en negatieve voorbeelden worden opnieuw gewogen zodat ze even zwaar wegen. Geschikt als de onderliggende gegevens zeer ongelijk zijn verdeeld.",
|
||||
"balancedAccuracyDescription": "Positieve en negatieve voorbeelden worden opnieuw gewogen zodat ze even zwaar wegen. Geschikt als de onderliggende gegevens zeer ongelijk zijn verdeeld.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Bins configureren",
|
||||
"makeCategorical": "Beschouwen als categorisch",
|
||||
"save": "Opslaan",
|
||||
"cancel": "Annuleren",
|
||||
"numberOfBins": "Aantal bins:",
|
||||
"categoryHeader": "Binwaarden:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum.",
|
||||
"defaultClassNames": "Klasa {0}",
|
||||
"defaultFeatureNames": "Cecha wrażliwa {0}",
|
||||
"defaultSingleFeatureName": "Cecha wrażliwa",
|
||||
"defaultCustomMetricName": "Metryka niestandardowa {0}",
|
||||
"performanceTab": "Atrakcyjność wydajności",
|
||||
"opportunityTab": "Atrakcyjność okazji",
|
||||
"modelComparisonTab": "Porównanie modeli",
|
||||
"tableTab": "Widok szczegółów",
|
||||
"dataSpecifications": "Statystyki danych",
|
||||
"attributes": "Atrybuty",
|
||||
"singleAttributeCount": "1 cecha wrażliwa",
|
||||
"attributesCount": "Cechy wrażliwe w liczbie {0}",
|
||||
"instanceCount": "Wystąpienia: {0}",
|
||||
"close": "Zamknij",
|
||||
"calculating": "Trwa obliczanie...",
|
||||
"performanceMetricLegacy": "Metryka wydajności",
|
||||
"errorOnInputs": "Błąd danych wejściowych. Cechy wrażliwe muszą być teraz wartościami kategorialnymi. Zamapuj wartości do kategorii w przedziałach i ponów próbę.",
|
||||
"Performancence": {
|
||||
"header": "Jak chcesz mierzyć wydajność?",
|
||||
"modelMakes": "model tworzy",
|
||||
"modelsMake": "modele tworzą",
|
||||
"body": "Dane zawierają etykiety w liczbie {0} oraz przewidywania ({2} {1}). Na podstawie tych informacji zalecamy następujące metryki. Wybierz jedną metrykę z listy.",
|
||||
"binaryClassifier": "klasyfikator binarny",
|
||||
"probabilisticRegressor": "regresor probitów",
|
||||
"regressor": "regresor",
|
||||
"binary": "binarny",
|
||||
"continuous": "ciągły"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Atrakcyjność mierzona za pomocą rozbieżności",
|
||||
"body": "Metryki rozbieżności określają wartość odchylenia dla zachowania modelu w ramach wybranych cech. Istnieją dwa rodzaje metryk rozbieżności: informacje zostaną uzupełnione...."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Dokumentacja"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Wstecz",
|
||||
"next": "Dalej"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Witamy w",
|
||||
"fairnessDashboard": "Pulpit nawigacyjny Fairness",
|
||||
"introBody": "Pulpit nawigacyjny Fairness umożliwia ocenę kompromisów między wydajnością i atrakcyjnością modeli",
|
||||
"explanatoryStep": "Aby skonfigurować ocenę, należy określić cechę wrażliwą i metrykę wydajności.",
|
||||
"getStarted": "Rozpocznij",
|
||||
"features": "Cechy wrażliwe",
|
||||
"featuresInfo": "Cechy wrażliwe służą do dzielenia danych na grupy. Atrakcyjność modelu w ramach tych grup jest mierzona za pomocą metryk rozbieżności. Metryki rozbieżności określają wielkość różnic w zachowaniu modelu względem grup.",
|
||||
"performance": "Metryka wydajności",
|
||||
"performanceInfo": "Metryki wydajności służą do oceny ogólnej jakości modelu oraz jakości modelu w każdej grupie. Różnica między skrajnymi wartościami metryki wydajności w grupach jest zgłaszana jako rozbieżność wydajności."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Porównanie modeli",
|
||||
"howToRead": "Jak czytać ten wykres",
|
||||
"lower": "mniejsza",
|
||||
"higher": "większa",
|
||||
"howToReadText": "Ten wykres reprezentuje każdy z {0} modeli jako punkt, który można wybrać. Oś X reprezentuje {1}, przy czym wartość {2} jest lepsza. Oś Y reprezentuje rozbieżność, przy czym wartość mniejsza jest lepsza.",
|
||||
"insights": "Analizy",
|
||||
"insightsText1": "Wykres pokazuje {0} i rozbieżność {1} modeli.",
|
||||
"insightsText2": "Zakresy w liczbie {0} od {1} do {2}. Zakresy rozbieżności od {3} do {4}.",
|
||||
"insightsText3": "Najbardziej dokładny model osiąga {0} z {1} i rozbieżność {2}.",
|
||||
"insightsText4": "Model najniższej rozbieżności osiąga {0} z {1} i rozbieżność {2}.",
|
||||
"disparityInOutcomes": "Rozbieżność w przewidywaniach",
|
||||
"disparityInPerformance": "Rozbieżność w {0}",
|
||||
"howToMeasureDisparity": "Jak należy mierzyć rozbieżności?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Model {0}",
|
||||
"title": "Rozbieżność wydajności",
|
||||
"globalPerformanceText": "To ogólnie {0}",
|
||||
"performanceDisparityText": "To rozbieżność w {0}",
|
||||
"editConfiguration": "Edytuj konfigurację",
|
||||
"backToComparisons": "Widok wielu modeli",
|
||||
"outcomesTitle": "Rozbieżność w przewidywaniach",
|
||||
"minTag": "Min.",
|
||||
"maxTag": "Maks.",
|
||||
"groupLabel": "Podgrupa",
|
||||
"underestimationError": "Niedoszacowanie",
|
||||
"underpredictionExplanation": "(przewidywane = 0, rzeczywiste = 1)",
|
||||
"overpredictionExplanation": "(przewidywane = 1, rzeczywiste = 0)",
|
||||
"overestimationError": "Przeszacowanie",
|
||||
"classificationOutcomesHowToRead": "Wykres słupkowy pokazuje współczynnik wyboru w każdej grupie, co oznacza ułamek punktów sklasyfikowanych jako 1.",
|
||||
"regressionOutcomesHowToRead": "Wykresy skrzynkowe pokazują rozkład przewidywań w każdej grupie. Poszczególne punkty danych są nałożone na górze.",
|
||||
"classificationPerformanceHowToRead1": "Wykres słupkowy pokazuje rozkład błędów w każdej grupie.",
|
||||
"classificationPerformanceHowToRead2": "Błędy są podzielone na błędy przeszacowania (przewidywanie 1, gdy rzeczywista etykieta to 0) i niedoszacowania (przewidywanie 0, gdy rzeczywista etykieta to 1).",
|
||||
"classificationPerformanceHowToRead3": "Zgłoszone współczynniki uzyskuje się przez podzielenie liczby błędów przez łączny rozmiar grupy.",
|
||||
"probabilityPerformanceHowToRead1": "Wykres słupkowy pokazuje średni błąd bezwzględny w każdej grupie podzielony na przeszacowania i niedoszacowania.",
|
||||
"probabilityPerformanceHowToRead2": "Dla każdej próbki mierzymy różnicę między przewidywaniem i etykietą. Jeśli jest dodatnia, oznacza to przeszacowanie, a jeśli ujemna — niedoszacowanie.",
|
||||
"probabilityPerformanceHowToRead3": "Zgłaszamy sumę błędów przeszacowania oraz sumę błędów niedoszacowania podzieloną przez łączny rozmiar grupy.",
|
||||
"regressionPerformanceHowToRead": "Błąd to różnica między przewidywaniem i etykietą. Wykresy skrzynkowe pokazują rozkład błędów w każdej grupie. Poszczególne punkty danych są nałożone na górze.",
|
||||
"distributionOfPredictions": "Rozkład przewidywań",
|
||||
"distributionOfErrors": "Rozkład błędów",
|
||||
"tooltipPrediction": "Przewidywanie: {0}",
|
||||
"tooltipError": "Błąd: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Względem jakich cech chcesz ocenić atrakcyjność modelu?",
|
||||
"body": "Atrakcyjność jest oceniana pod względem rozbieżności w zachowaniu modelu. Dane zostaną podzielone według wartości każdej wybranej cechy, a następnie zostaną ocenione różnice między metryką wydajności i przewidywaniami modelu w ramach tych podziałów.",
|
||||
"learnMore": "Dowiedz się więcej",
|
||||
"summaryCategoricalCount": "Liczba unikatowych wartości tej cechy to {0}",
|
||||
"summaryNumericCount": "Zakres tej cechy liczbowej to od {0} do {1} w {2} przedziałach.",
|
||||
"showCategories": "Pokaż wszystko",
|
||||
"hideCategories": "Zwiń",
|
||||
"categoriesOverflow": " i dodatkowe kategorie w liczbie {0}",
|
||||
"editBinning": "Edytuj grupy",
|
||||
"subgroups": "Podgrupy"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Dokładność",
|
||||
"precisionScore": "Dokładność",
|
||||
"recallScore": "Unieważnij",
|
||||
"zeroOneLoss": "Utrata zerojedynkowa",
|
||||
"specificityScore": "Wynik swoistości",
|
||||
"missRate": "Współczynnik chybień",
|
||||
"falloutRate": "Współczynnik odpadów",
|
||||
"maxError": "Maksymalny błąd",
|
||||
"meanAbsoluteError": "Średni błąd bezwzględny",
|
||||
"meanSquaredError": " MSE (średni błąd kwadratowy)",
|
||||
"meanSquaredLogError": "Błąd dziennika średniego kwadratowego",
|
||||
"medianAbsoluteError": "Mediana błędu bezwzględnego",
|
||||
"average": "Średnie przewidywanie",
|
||||
"selectionRate": "Współczynnik wyboru",
|
||||
"overprediction": "Przeszacowanie",
|
||||
"underprediction": "Niedoszacowanie",
|
||||
"r2_score": "Wynik współczynnika R do kwadratu",
|
||||
"rms_error": "RMSE (średni błąd kwadratowy)",
|
||||
"auc": "Obszar pod krzywą ROC",
|
||||
"balancedRootMeanSquaredError": "Zrównoważone RMSE",
|
||||
"balancedAccuracy": "Zrównoważona dokładność",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Ułamek punktów danych sklasyfikowanych poprawnie.",
|
||||
"precisionDescription": "Ułamek punktów danych sklasyfikowanych poprawnie wśród tych, które sklasyfikowano jako 1.",
|
||||
"recallDescription": "Ułamek punktów danych sklasyfikowanych poprawnie wśród tych, których rzeczywista etykieta to 1. Alternatywne nazwy: rzeczywisty współczynnik poprawności, wrażliwość.",
|
||||
"rmseDescription": "Pierwiastek kwadratowy ze średniej błędów do kwadratu.",
|
||||
"mseDescription": "Średnia kwadratów błędów.",
|
||||
"meanAbsoluteErrorDescription": "Średnia wartości bezwzględnych błędów. Bardziej niezawodna dla wartości odstających niż wartość MSE.",
|
||||
"r2Description": "Ułamek odchylenia w etykietach objaśnionych przez model.",
|
||||
"aucDescription": "Jakość przewidywań dla oddzielania próbek pozytywnych od negatywnych wyświetlana w postaci wyników.",
|
||||
"balancedRMSEDescription": "Próbki pozytywne i negatywne są ważone ponownie, tak aby miały równą wagę łączną. Odpowiednie, jeśli dane podstawowe są bardzo niezrównoważone.",
|
||||
"balancedAccuracyDescription": "Próbki pozytywne i negatywne są ważone ponownie, tak aby miały równą wagę łączną. Odpowiednie, jeśli dane podstawowe są bardzo niezrównoważone.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Konfiguruj przedziały",
|
||||
"makeCategorical": "Traktuj jako kategorialne",
|
||||
"save": "Zapisz",
|
||||
"cancel": "Anuluj",
|
||||
"numberOfBins": "Liczba przedziałów:",
|
||||
"categoryHeader": "Wartości przedziału:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Classe {0}",
|
||||
"defaultFeatureNames": "Recurso confidencial {0}",
|
||||
"defaultSingleFeatureName": "Recurso confidencial",
|
||||
"defaultCustomMetricName": "Métrica personalizada {0}",
|
||||
"performanceTab": "Equidade no Desempenho",
|
||||
"opportunityTab": "Equidade na Oportunidade",
|
||||
"modelComparisonTab": "Comparação de modelo",
|
||||
"tableTab": "Exibição de Detalhes",
|
||||
"dataSpecifications": "Estatísticas de dados",
|
||||
"attributes": "Atributos",
|
||||
"singleAttributeCount": "1 recurso confidencial",
|
||||
"attributesCount": "{0} recursos confidenciais",
|
||||
"instanceCount": "{0} instâncias",
|
||||
"close": "Fechar",
|
||||
"calculating": "Calculando...",
|
||||
"performanceMetricLegacy": "Métrica de desempenho",
|
||||
"errorOnInputs": "Erro com a entrada. Recursos confidenciais precisam ser valores categóricos neste momento. Mapeie valores para as categorias compartimentalizadas e tente novamente.",
|
||||
"Performance": {
|
||||
"header": "Como deseja medir o desempenho?",
|
||||
"modelMakes": "realizado por modelo",
|
||||
"modelsMake": "realizado por modelos",
|
||||
"body": "Os seus dados contêm {0} rótulos e as suas previsões de {2} {1}. Com base nessas informações, recomendamos as métricas a seguir. Selecione uma métrica na lista.",
|
||||
"binaryClassifier": "classificador binário",
|
||||
"probabilisticRegressor": "regressor de probit",
|
||||
"regressor": "regressor",
|
||||
"binary": "binário",
|
||||
"continuous": "contínuo"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Equidade medida em termos de disparidade",
|
||||
"body": "As métricas de disparidade quantificam a variação do comportamento do seu modelo nos recursos selecionados. Há dois tipos de métricas de disparidade: mais em breve...."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Documentação"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Voltar",
|
||||
"next": "Avançar"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Bem-vindo(a) a",
|
||||
"fairnessDashboard": "Painel do Fairness",
|
||||
"introBody": "O painel do Fairness permite que você avalie as compensações entre o desempenho e a equidade dos seus modelos",
|
||||
"explanatoryStep": "Para configurar a avaliação, é necessário especificar um recurso confidencial e uma métrica de desempenho.",
|
||||
"getStarted": "Introdução",
|
||||
"features": "Recursos confidenciais",
|
||||
"featuresInfo": "Recursos confidenciais são usados para dividir os seus dados em grupos. A equidade do seu modelo nesses grupos é medida por métricas de disparidade. As métricas de disparidade quantificam o quanto o comportamento do modelo varia nesses grupos.",
|
||||
"performance": "Métrica de desempenho",
|
||||
"performanceInfo": "As métricas de desempenho são usadas para avaliar a qualidade geral do seu modelo, bem como a qualidade do seu modelo em cada grupo. A diferença entre os valores extremos da métrica de desempenho nos grupos é relatada como a disparidade no desempenho."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Comparação de modelo",
|
||||
"howToRead": "Como ler este gráfico",
|
||||
"lower": "menor",
|
||||
"higher": "maior",
|
||||
"howToReadText": "Este gráfico representa cada um dos modelos de {0} como um ponto selecionável. O eixo x representa {1}, no qual {2} é melhor. O eixo y representa a disparidade, na qual um valor menor é melhor.",
|
||||
"insights": "Insights",
|
||||
"insightsText1": "O gráfico mostra {0} e a disparidade de modelos de {1}.",
|
||||
"insightsText2": "{0} tem um intervalo de {1} a {2}. A disparidade tem um intervalo de {3} a {4}.",
|
||||
"insightsText3": "O modelo mais preciso atinge {0} de {1} e uma disparidade de {2}.",
|
||||
"insightsText4": "O modelo com menor disparidade atinge {0} de {1} e uma disparidade de {2}.",
|
||||
"disparityInOutcomes": "Disparidade nas previsões",
|
||||
"disparityInPerformance": "A disparidade no {0}",
|
||||
"howToMeasureDisparity": "Como deve ser medida a disparidade?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Modelo {0}",
|
||||
"title": "Disparidade no desempenho",
|
||||
"globalPerformanceText": "É o {0} geral",
|
||||
"performanceDisparityText": "É a disparidade em {0}",
|
||||
"editConfiguration": "Editar a configuração",
|
||||
"backToComparisons": "Exibição multimodelo",
|
||||
"outcomesTitle": "Disparidade nas previsões",
|
||||
"minTag": "Mín.",
|
||||
"maxTag": "Máx.",
|
||||
"groupLabel": "Subgrupo",
|
||||
"underestimationError": "Subprevisão",
|
||||
"underpredictionExplanation": "(previsto = 0, verdadeiro = 1)",
|
||||
"overpredictionExplanation": "(previsto = 1, verdadeiro = 0)",
|
||||
"overestimationError": "Sobreprevisão",
|
||||
"classificationOutcomesHowToRead": "O gráfico de barras mostra a taxa de seleção em cada grupo, o que significa a fração de pontos classificados como 1.",
|
||||
"regressionOutcomesHowToRead": "Os gráficos de caixa mostram a distribuição de previsões em cada grupo. Os pontos de dados individuais estão sobrepostos na parte superior.",
|
||||
"classificationPerformanceHowToRead1": "O gráfico de barras mostra a distribuição de erros em cada grupo.",
|
||||
"classificationPerformanceHowToRead2": "Os erros são divididos em erros de sobreprevisão (prever 1 quando o rótulo verdadeiro é 0) e erros de subprevisão (prever 0 quando o rótulo verdadeiro é 1).",
|
||||
"classificationPerformanceHowToRead3": "As taxas relatadas são obtidas ao dividir o número de erros pelo tamanho geral do grupo.",
|
||||
"probabilityPerformanceHowToRead1": "O gráfico de barras mostra um erro absoluto médio em cada grupo, dividido em sobreprevisão e subprevisão.",
|
||||
"probabilityPerformanceHowToRead2": "Em cada exemplo, medimos a diferença entre a previsão e o rótulo. Se for positiva, nós a chamaremos de sobreprevisão, e se for negativa, nós a chamaremos de subprevisão.",
|
||||
"probabilityPerformanceHowToRead3": "Relatamos a soma de erros de sobreprevisão e a soma de erros de subprevisão divididas pelo tamanho do grupo geral.",
|
||||
"regressionPerformanceHowToRead": "O erro é a diferença entre a previsão e o rótulo. Os gráficos de caixa mostram a distribuição de erros em cada grupo. Os pontos de dados individuais estão sobrepostos na parte superior.",
|
||||
"distributionOfPredictions": "Distribuição de previsões",
|
||||
"distributionOfErrors": "Distribuição de erros",
|
||||
"tooltipPrediction": "Previsão: {0}",
|
||||
"tooltipError": "Erro: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Junto com quais recursos você deseja avaliar a equidade do seu modelo?",
|
||||
"body": "A equidade é avaliada em termos de disparidades no comportamento do modelo. Vamos dividir os seus dados de acordo com os valores de cada recurso selecionado e avaliar como a métrica de desempenho e as previsões do modelo diferem nessas divisões.",
|
||||
"learnMore": "Saiba mais",
|
||||
"summaryCategoricalCount": "Este recurso tem {0} valores exclusivos",
|
||||
"summaryNumericCount": "Este recurso numérico varia de {0} a {1} e é agrupado em {2} compartimentos.",
|
||||
"showCategories": "Mostrar tudo",
|
||||
"hideCategories": "Recolher",
|
||||
"categoriesOverflow": " e {0} categorias adicionais",
|
||||
"editBinning": "Editar grupos",
|
||||
"subgroups": "Subgrupos"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Precisão",
|
||||
"precisionScore": "Precisão",
|
||||
"recallScore": "Recall",
|
||||
"zeroOneLoss": "Perda zero-um",
|
||||
"specificityScore": "Pontuação de especificidade",
|
||||
"missRate": "Taxa de erros",
|
||||
"falloutRate": "Taxa de fallout",
|
||||
"maxError": "Máximo de erros",
|
||||
"meanAbsoluteError": "Erro absoluto médio",
|
||||
"meanSquaredError": " MSE (erro quadrático médio)",
|
||||
"meanSquaredLogError": "Erro logarítmico quadrático médio",
|
||||
"medianAbsoluteError": "Erro absoluto mediano",
|
||||
"average": "Previsão média",
|
||||
"selectionRate": "Taxa de seleção",
|
||||
"overprediction": "Sobreprevisão",
|
||||
"underprediction": "Subprevisão",
|
||||
"r2_score": "Pontuação de R-quadrado",
|
||||
"rms_error": "RMSE (raiz do erro quadrático médio)",
|
||||
"auc": "Área abaixo da curva do ROC",
|
||||
"balancedRootMeanSquaredError": "RMSE equilibrado",
|
||||
"balancedAccuracy": "Precisão balanceada",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "A fração de pontos de dados classificados corretamente.",
|
||||
"precisionDescription": "A fração de pontos de dados classificados corretamente entre aqueles classificados como 1.",
|
||||
"recallDescription": "A fração de pontos de dados classificados corretamente entre aqueles cujo rótulo verdadeiro é 1. Nomes alternativos: taxa de verdadeiro positivo, confidencialidade.",
|
||||
"rmseDescription": "Raiz quadrada da média de erros quadráticos.",
|
||||
"mseDescription": "A média de erros quadráticos.",
|
||||
"meanAbsoluteErrorDescription": "A média de valores absolutos de erros. Mais robusto para exceções do que para MSE.",
|
||||
"r2Description": "A fração de variância nos rótulos explicada pelo modelo.",
|
||||
"aucDescription": "A qualidade das previsões, exibidas como pontuações, ao separar exemplos positivos de exemplos negativos.",
|
||||
"balancedRMSEDescription": "Exemplos positivos e negativos são ponderados novamente para que tenham o peso total igual. Adequado se os dados subjacentes forem altamente desequilibrados.",
|
||||
"balancedAccuracyDescription": "Exemplos positivos e negativos são ponderados novamente para que tenham o peso total igual. Adequado se os dados subjacentes forem altamente desequilibrados.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Configurar compartimentos",
|
||||
"makeCategorical": "Tratar como categórico",
|
||||
"save": "Salvar",
|
||||
"cancel": "Cancelar",
|
||||
"numberOfBins": "Número de compartimentos:",
|
||||
"categoryHeader": "Valores de compartimento:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Classe {0}",
|
||||
"defaultFeatureNames": "Funcionalidade sensível {0}",
|
||||
"defaultSingleFeatureName": "Funcionalidade sensível",
|
||||
"defaultCustomMetricName": "Métrica personalizada {0}",
|
||||
"performanceTab": "Equidade no Desempenho",
|
||||
"opportunityTab": "Equidade na Oportunidade",
|
||||
"modelComparisonTab": "Comparação de modelos",
|
||||
"tableTab": "Vista de Detalhes",
|
||||
"dataSpecifications": "Estatísticas de dados",
|
||||
"attributes": "Atributos",
|
||||
"singleAttributeCount": "1 funcionalidade sensível",
|
||||
"attributesCount": "{0} funcionalidades sensíveis",
|
||||
"instanceCount": "{0} instâncias",
|
||||
"close": "Fechar",
|
||||
"calculating": "A calcular...",
|
||||
"performanceMetricLegacy": "Métrica de desempenho",
|
||||
"errorOnInputs": "Erro na entrada. As funcionalidades sensíveis têm de ser valores categóricos neste momento. Mapeie os valores para as categorias discretizadas e tente novamente.",
|
||||
"Performance": {
|
||||
"header": "Como quer medir o desempenho?",
|
||||
"modelMakes": "o modelo faz",
|
||||
"modelsMake": "os modelos fazem",
|
||||
"body": "Os dados contêm {0} etiquetas e {2} {1} predições. Com base nessa informação, recomendamos as seguintes métricas. Selecione uma métrica da lista.",
|
||||
"binaryClassifier": "classificador binário",
|
||||
"probabilisticRegressor": "regressor probit",
|
||||
"regressor": "regressor",
|
||||
"binary": "binário",
|
||||
"continuous": "contínuo"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Equidade medida em termos de disparidade",
|
||||
"body": "As métricas de disparidade quantificam a variação do comportamento do seu modelo nas funcionalidades selecionadas. Há dois tipos de métricas de disparidade: mais novidades em breve...."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Documentação"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Anterior",
|
||||
"next": "Seguinte"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Bem-vindo ao",
|
||||
"fairnessDashboard": "Dashboard do Fairness",
|
||||
"introBody": "O dashboard do Fairness permite avaliar compromissos entre desempenho e equidade dos seus modelos",
|
||||
"explanatoryStep": "Para configurar a avaliação, é necessário especificar uma funcionalidade sensível e uma métrica de desempenho.",
|
||||
"getStarted": "Introdução",
|
||||
"features": "Funcionalidades sensíveis",
|
||||
"featuresInfo": "As funcionalidades sensíveis servem para dividir os seus dados em grupos. A equidade do seu modelo entre estes grupos é medida por métricas de disparidade. As métricas de disparidade quantificam a variação do comportamento do seu modelo entre estes grupos.",
|
||||
"performance": "Métrica de desempenho",
|
||||
"performanceInfo": "As métricas de desempenho servem para avaliar a qualidade global do seu modelo, bem como a qualidade do seu modelo em cada grupo. A diferença entre os valores extremos da métrica de desempenho entre os grupos é reportada como a disparidade no desempenho."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Comparação de modelos",
|
||||
"howToRead": "Como ler este gráfico",
|
||||
"lower": "mais baixa",
|
||||
"higher": "mais alta",
|
||||
"howToReadText": "Este gráfico representa cada um dos {0} modelos como um ponto selecionável. O eixo x representa {1}, sendo que {2} é melhor. O eixo y representa a disparidade, sendo que mais baixa é melhor.",
|
||||
"insights": "Informações",
|
||||
"insightsText1": "O gráfico mostra {0} e disparidade de {1} modelos.",
|
||||
"insightsText2": "{0} varia de {1} a {2}. A disparidade varia de {3} a {4}.",
|
||||
"insightsText3": "O modelo mais preciso consegue {0} de {1} e uma disparidade de {2}.",
|
||||
"insightsText4": "O modelo com menor disparidade consegue {0} de {1} e uma disparidade de {2}.",
|
||||
"disparityInOutcomes": "Disparidade nas predições",
|
||||
"disparityInPerformance": "Disparidade em {0}",
|
||||
"howToMeasureDisparity": "Como se deve medir a disparidade?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Modelo {0}",
|
||||
"title": "Disparidade no desempenho",
|
||||
"globalPerformanceText": "É o {0} global",
|
||||
"performanceDisparityText": "É a disparidade em {0}",
|
||||
"editConfiguration": "Editar configuração",
|
||||
"backToComparisons": "Vista de vários modelos",
|
||||
"outcomesTitle": "Disparidade nas predições",
|
||||
"minTag": "Mín.",
|
||||
"maxTag": "Máx.",
|
||||
"groupLabel": "Subgrupo",
|
||||
"underestimationError": "Subpredição",
|
||||
"underpredictionExplanation": "(previsto = 0, verdadeiro = 1)",
|
||||
"overpredictionExplanation": "(previsto = 1, verdadeiro = 0)",
|
||||
"overestimationError": "Sobrepredição",
|
||||
"classificationOutcomesHowToRead": "O gráfico de barras mostra a taxa de seleção em cada grupo, o que significa a fração de pontos classificados como 1.",
|
||||
"regressionOutcomesHowToRead": "Os gráficos de caixas mostram a distribuição das predições em cada grupo. Os pontos de dados individuais estão sobrepostos em cima.",
|
||||
"classificationPerformanceHowToRead1": "O gráfico de barras mostra a distribuição de erros em cada grupo.",
|
||||
"classificationPerformanceHowToRead2": "Os erros dividem-se em erros de sobrepredição (prever 1 quando a etiqueta verdadeira é 0) e erros de subpredição (prever 0 quando a etiqueta verdadeira é 1).",
|
||||
"classificationPerformanceHowToRead3": "As taxas reportadas são obtidas dividindo o número de erros pelo tamanho do grupo global.",
|
||||
"probabilityPerformanceHowToRead1": "O gráfico de barras mostra o erro absoluto médio em cada grupo, dividido em sobrepredição e subpredição.",
|
||||
"probabilityPerformanceHowToRead2": "Em cada exemplo, medimos a diferença entre a predição e a etiqueta. Se for positivo, chamamos sobrepredição e, se for negativo, chamamos subpredição.",
|
||||
"probabilityPerformanceHowToRead3": "Reportamos a soma dos erros de sobrepredição e a soma dos erros de subpredição divididos pelo tamanho do grupo global.",
|
||||
"regressionPerformanceHowToRead": "Erro é a diferença entre a predição e a etiqueta. Os gráficos de caixas mostram a distribuição de erros em cada grupo. Os pontos de dados individuais estão sobrepostos em cima.",
|
||||
"distributionOfPredictions": "Distribuição de predições",
|
||||
"distributionOfErrors": "Distribuição de erros",
|
||||
"tooltipPrediction": "Predição: {0}",
|
||||
"tooltipError": "Erro: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Em que funcionalidades gostaria de avaliar a equidade do seu modelo?",
|
||||
"body": "A equidade é avaliada em termos de disparidades no comportamento do seu modelo. Dividiremos os seus dados de acordo com os valores de cada funcionalidade selecionada e avaliaremos de que forma a métrica de desempenho do seu modelo e as predições diferem entre estas divisões.",
|
||||
"learnMore": "Saiba mais",
|
||||
"summaryCategoricalCount": "Esta funcionalidade tem {0} valores exclusivos",
|
||||
"summaryNumericCount": "Esta funcionalidade numérica varia de {0} a {1}, e está agrupada em {2} discretizações.",
|
||||
"showCategories": "Mostrar tudo",
|
||||
"hideCategories": "Fechar",
|
||||
"categoriesOverflow": " e {0} categorias adicionais",
|
||||
"editBinning": "Editar grupos",
|
||||
"subgroups": "Subgrupos"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Precisão",
|
||||
"precisionScore": "Precisão",
|
||||
"recallScore": "Revocação",
|
||||
"zeroOneLoss": "Perda zero-um",
|
||||
"specificityScore": "Classificação de especificidade",
|
||||
"missRate": "Taxa de erros",
|
||||
"falloutRate": "Taxa de dispersão",
|
||||
"maxError": "Erro máximo",
|
||||
"meanAbsoluteError": "Erro absoluto médio",
|
||||
"meanSquaredError": " MSE (erro quadrático médio)",
|
||||
"meanSquaredLogError": "Erro de registo quadrático médio",
|
||||
"medianAbsoluteError": "Erro mediano absoluto",
|
||||
"average": "Predição média",
|
||||
"selectionRate": "Taxa de seleção",
|
||||
"overprediction": "Sobrepredição",
|
||||
"underprediction": "Subpredição",
|
||||
"r2_score": "Classificação de R ao quadrado",
|
||||
"rms_error": "RMSE (raiz do erro quadrático médio)",
|
||||
"auc": "Área sob a curva ROC",
|
||||
"balancedRootMeanSquaredError": "RMSE equilibrado",
|
||||
"balancedAccuracy": "Precisão equilibrada",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "A fração de pontos de dados classificados corretamente.",
|
||||
"precisionDescription": "A fração de pontos de dados classificados corretamente entre os classificados como 1.",
|
||||
"recallDescription": "A fração de pontos de dados classificados corretamente entre aqueles cuja etiqueta verdadeira é 1. Nomes alternativos: taxa de verdadeiros positivos, sensibilidade.",
|
||||
"rmseDescription": "A raiz quadrada da média de erros quadráticos.",
|
||||
"mseDescription": "A média de erros quadráticos.",
|
||||
"meanAbsoluteErrorDescription": "A média dos valores absolutos dos erros. Mais robusto para valores atípicos do que o MSE.",
|
||||
"r2Description": "A fração de desvio nas etiquetas explicadas pelo modelo.",
|
||||
"aucDescription": "A qualidade das predições, visualizada como classificações, na separação de exemplos positivos de exemplos negativos.",
|
||||
"balancedRMSEDescription": "Os exemplos positivos e negativos são reponderados para terem uma ponderação total igual. Adequado se existir um grande desequilíbrio nos dados subjacentes.",
|
||||
"balancedAccuracyDescription": "Os exemplos positivos e negativos são reponderados para terem uma ponderação total igual. Adequado se existir um grande desequilíbrio nos dados subjacentes.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Configurar discretizações",
|
||||
"makeCategorical": "Tratar como categórico",
|
||||
"save": "Guardar",
|
||||
"cancel": "Cancelar",
|
||||
"numberOfBins": "Número de discretizações:",
|
||||
"categoryHeader": "Valores de discretização:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Класс {0}",
|
||||
"defaultFeatureNames": "Чувствительный признак {0}",
|
||||
"defaultSingleFeatureName": "Чувствительный признак",
|
||||
"defaultCustomMetricName": "Настраиваемая метрика {0}",
|
||||
"performanceTab": "Достоверность по производительности",
|
||||
"opportunityTab": "Достоверность по возможностям",
|
||||
"modelComparisonTab": "Сравнение моделей",
|
||||
"tableTab": "Подробное представление",
|
||||
"dataSpecifications": "Статистика данных",
|
||||
"attributes": "Атрибуты",
|
||||
"singleAttributeCount": "1 чувствительный признак",
|
||||
"attributesCount": "Чувствительные признаки: {0}",
|
||||
"instanceCount": "Экземпляров: {0}",
|
||||
"close": "Закрыть",
|
||||
"calculating": "Вычисление…",
|
||||
"performanceMetricLegacy": "Метрика производительности",
|
||||
"errorOnInputs": "Ошибка ввода. Чувствительные признаки сейчас должны быть категориальными значениями. Назначьте значения категориям интервалов и повторите попытку.",
|
||||
"Performance": {
|
||||
"header": "Как вы хотите измерять производительность?",
|
||||
"modelMakes": "модель делает",
|
||||
"modelsMake": "модели делают",
|
||||
"body": "Ваши данные содержат {0} метки, а {1} прогнозы ({2}). На основе этих сведений мы рекомендуем использовать следующие метрики. Выберите одну метрику из списка.",
|
||||
"binaryClassifier": "двоичный классификатор",
|
||||
"probabilisticRegressor": "пробит-регрессор",
|
||||
"regressor": "регрессор",
|
||||
"binary": "двоичные",
|
||||
"continuous": "непрерывные"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Измерение достоверности через несоответствие",
|
||||
"body": "Метрики несоответствия количественно характеризуют отличия в поведении модели по выбранным признакам. Доступны два вида метрик, но будут добавлены и другие…"
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Документация"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Назад",
|
||||
"next": "Далее"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Вас приветствует",
|
||||
"fairnessDashboard": "Панель мониторинга Fairness",
|
||||
"introBody": "Панель мониторинга Fairness позволяет оценивать компромисс между производительностью и достоверностью моделей.",
|
||||
"explanatoryStep": "Чтобы настроить оценку, необходимо указать чувствительный признак и метрику производительности.",
|
||||
"getStarted": "Начало работы",
|
||||
"features": "Чувствительные признаки",
|
||||
"featuresInfo": "Чувствительные признаки используются для разделения данных на группы. Достоверность вашей модели в этих группах измеряется с помощью метрик несоответствия, которые количественно показывают различия в ее поведении в разных группах.",
|
||||
"performance": "Метрика производительности",
|
||||
"performanceInfo": "Метрики производительности используются для оценки общего качества модели, а также качества модели в каждой группе. Различие между крайними значениями метрики в группах представляется как несоответствие в работе."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Сравнение моделей",
|
||||
"howToRead": "Как читать эту диаграмму",
|
||||
"lower": "ниже",
|
||||
"higher": "выше",
|
||||
"howToReadText": "На этой диаграмме каждая из моделей ({0}) представляет собой точку для выбора. Ось X представляет функцию \"{1}\" (чем {2}, тем лучше), а ось Y характеризует несоответствие (чем ниже, тем лучше).",
|
||||
"insights": "Аналитика",
|
||||
"insightsText1": "Диаграмма показывает функцию \"{0}\" и несоответствие моделей ({1}).",
|
||||
"insightsText2": "Функция \"{0}\" имеет диапазон от {1} до {2}, а несоответствие — от {3} до {4}.",
|
||||
"insightsText3": "Самая правильная модель обеспечивает для функции \"{0}\" значение {1} и имеет несоответствие {2}.",
|
||||
"insightsText4": "Модель с наименьшим несоответствием обеспечивает для функции \"{0}\" значение {1} и имеет несоответствие {2}.",
|
||||
"disparityInOutcomes": "Несоответствие прогнозов",
|
||||
"disparityInPerformance": "Несоответствие в {0}",
|
||||
"howToMeasureDisparity": "Как измерять несоответствие?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Модель {0}",
|
||||
"title": "Несоответствие производительности",
|
||||
"globalPerformanceText": "Общее значение для \"{0}\"",
|
||||
"performanceDisparityText": "Является несоответствием в функции \"{0}\"",
|
||||
"editConfiguration": "Изменить конфигурацию",
|
||||
"backToComparisons": "Многомодельное представление",
|
||||
"outcomesTitle": "Несоответствие прогнозов",
|
||||
"minTag": "Мин.",
|
||||
"maxTag": "Макс.",
|
||||
"groupLabel": "Подгруппа",
|
||||
"underestimationError": "Преуменьшение прогноза",
|
||||
"underpredictionExplanation": "(прогноз = 0, истина = 1)",
|
||||
"overpredictionExplanation": "(прогноз = 1, истина = 0)",
|
||||
"overestimationError": "Преувеличение прогноза",
|
||||
"classificationOutcomesHowToRead": "Линейчатая диаграмма показывает степень отбора в каждой группе, то есть долю точек с классификацией 1.",
|
||||
"regressionOutcomesHowToRead": "Блочные диаграммы показывают распределение прогнозов в каждой группе. Сверху наложены отдельные точки данных.",
|
||||
"classificationPerformanceHowToRead1": "Линейчатая диаграмма показывает распределение ошибок в каждой группе.",
|
||||
"classificationPerformanceHowToRead2": "Ошибки разделяются на ошибки преувеличения прогноза (прогнозирование значения 1, когда истинная метка — 0) и преуменьшения прогноза (прогнозирование 0, когда истинная метка — 1).",
|
||||
"classificationPerformanceHowToRead3": "Выдаваемые показатели получаются делением числа ошибок на общий размер группы.",
|
||||
"probabilityPerformanceHowToRead1": "Линейчатая диаграмма показывает среднюю абсолютную ошибку в каждой группе с подразделением на преуменьшение и преуменьшение прогноза.",
|
||||
"probabilityPerformanceHowToRead2": "В каждом примере мы измеряем разницу между прогнозом и меткой. Если разница положительная, это называется преувеличением прогноза, а если отрицательная — преуменьшением.",
|
||||
"probabilityPerformanceHowToRead3": "Мы сообщаем сумму ошибок преувеличения прогноза и сумму ошибок преуменьшения, деленные на общий размер группы.",
|
||||
"regressionPerformanceHowToRead": "Ошибка представляет собой разницу между прогнозом и меткой. Блочные диаграммы показывают распределение ошибок в каждой группе. Сверху наложены отдельные точки данных.",
|
||||
"distributionOfPredictions": "Распределение прогнозов",
|
||||
"distributionOfErrors": "Распределение ошибок",
|
||||
"tooltipPrediction": "Прогноз: {0}",
|
||||
"tooltipError": "Ошибка: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Для каких признаков вы хотите оценить достоверность модели?",
|
||||
"body": "Достоверность оценивается с точки зрения несоответствий в поведении модели. Мы разделим ваши данные по значениям каждого выбранного признака и оценим, насколько метрики производительности и прогнозы вашей модели различаются в этих группах.",
|
||||
"learnMore": "Дополнительные сведения",
|
||||
"summaryCategoricalCount": "Уникальных значений признака: {0}",
|
||||
"summaryNumericCount": "Этот числовой признак имеет диапазон от {0} до {1} и группируется в следующее число интервалов: {2}.",
|
||||
"showCategories": "Показать все",
|
||||
"hideCategories": "Свернуть",
|
||||
"categoriesOverflow": " и другие категории ({0})",
|
||||
"editBinning": "Изменить группы",
|
||||
"subgroups": "Подгруппы"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Правильность",
|
||||
"precisionScore": "Точность",
|
||||
"recallScore": "Полнота",
|
||||
"zeroOneLoss": "Двоичная функция потерь",
|
||||
"specificityScore": "Оценка специфичности",
|
||||
"missRate": "Доля промахов",
|
||||
"falloutRate": "Доля ложных заключений",
|
||||
"maxError": "Максимальная ошибка",
|
||||
"meanAbsoluteError": "Средняя абсолютная ошибка",
|
||||
"meanSquaredError": " Среднеквадратичная ошибка",
|
||||
"meanSquaredLogError": "Среднеквадратичная ошибка журнала",
|
||||
"medianAbsoluteError": "Медианная абсолютная ошибка",
|
||||
"average": "Средний прогноз",
|
||||
"selectionRate": "Степень отбора",
|
||||
"overprediction": "Преувеличение прогноза",
|
||||
"underprediction": "Преуменьшение прогноза",
|
||||
"r2_score": "Оценка \"R-квадрат\"",
|
||||
"rms_error": "Среднеквадратичная ошибка (корень)",
|
||||
"auc": "Площадь под ROC-кривой",
|
||||
"balancedRootMeanSquaredError": "Сбалансированная среднеквадратичная ошибка",
|
||||
"balancedAccuracy": "Сбалансированная правильность",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Доля точек данных с корректной классификацией.",
|
||||
"precisionDescription": "Доля точек данных с корректной классификацией среди тех, что классифицированы как 1.",
|
||||
"recallDescription": "Доля точек данных с корректной классификацией среди тех, чья истинная метка — 1. Другие названия: доля истинноположительных результатов, чувствительность.",
|
||||
"rmseDescription": "Квадратный корень среднего значения из квадратов ошибки.",
|
||||
"mseDescription": "Среднее значение из квадратов ошибки.",
|
||||
"meanAbsoluteErrorDescription": "Среднее число абсолютных величин ошибки. Больше подходит для крайних значений, чем для среднеквадратичной ошибки.",
|
||||
"r2Description": "Доля изменчивости в метках, объясняемая моделью.",
|
||||
"aucDescription": "Качество (числовые оценки) прогнозов на основе отделения положительных примеров от отрицательных.",
|
||||
"balancedRMSEDescription": "Переназначение весов для положительных и отрицательных примеров так, чтобы общий вес совпадал. Рекомендуется для сильно несбалансированных данных.",
|
||||
"balancedAccuracyDescription": "Переназначение весов для положительных и отрицательных примеров так, чтобы общий вес совпадал. Рекомендуется для сильно несбалансированных данных.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Настройка интервалов",
|
||||
"makeCategorical": "Рассматривать как категориальные",
|
||||
"save": "Сохранить",
|
||||
"cancel": "Отмена",
|
||||
"numberOfBins": "Число интервалов:",
|
||||
"categoryHeader": "Значения интервалов:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "Klass {0}",
|
||||
"defaultFeatureNames": "Känslig funktion {0}",
|
||||
"defaultSingleFeatureName": "Känslig funktion",
|
||||
"defaultCustomMetricName": "Anpassat mått {0}",
|
||||
"performanceTab": "Rättvisa i prestanda",
|
||||
"opportunityTab": "Rättvisa i affärsmöjlighet",
|
||||
"modelComparisonTab": "Modelljämförelse",
|
||||
"tableTab": "Detaljvy",
|
||||
"dataSpecifications": "Datastatistik",
|
||||
"attributes": "Attribut",
|
||||
"singleAttributeCount": "1 känslig funktion",
|
||||
"attributesCount": "{0} känsliga funktioner",
|
||||
"instanceCount": "{0} instanser",
|
||||
"close": "Stäng",
|
||||
"calculating": "Beräknar...",
|
||||
"performanceMetricLegacy": "Prestandamått",
|
||||
"errorOnInputs": "Fel med indata. De känsliga funktionerna måste vara kategoriska värden just nu. Mappa värden till diskretiserade kategorier och försök igen.",
|
||||
"Performance": {
|
||||
"header": "Hur vill du mäta prestanda?",
|
||||
"modelMakes": "modellen gör",
|
||||
"modelsMake": "modellerna gör",
|
||||
"body": "Dina data innehåller {0} etiketter och {2} {1} förutsägelser. Baserat på den informationen rekommenderar vi följande mått. Välj ett mått från listan.",
|
||||
"binaryClassifier": "binär klassificerare",
|
||||
"probabilisticRegressor": "probitregressor",
|
||||
"regressor": "regressor",
|
||||
"binary": "binär",
|
||||
"continuous": "kontinuerlig"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Rättvisa mätt avseende diskrepans",
|
||||
"body": "Diskrepansmått kvantifierar variationer i dina modellers beteende för valda funktioner. Det finns två olika typer av diskrepansmått: mer kommer..."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Dokumentation"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Tillbaka",
|
||||
"next": "Nästa"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Välkommen till",
|
||||
"fairnessDashboard": "Fairness-instrumentpanel",
|
||||
"introBody": "Med Fairness-instrumentpanelen kan du utvärdera kompromisser mellan prestanda och rättvisa för dina modeller",
|
||||
"explanatoryStep": "Om du vill konfigurera utvärderingen måste du ange en känslig funktion och ett prestandamått.",
|
||||
"getStarted": "Kom igång",
|
||||
"features": "Känsliga funktioner",
|
||||
"featuresInfo": "Känsliga funktioner används för att dela upp dina data i grupper. Din modells rättvisa för dessa grupper mäts av diskrepansmått. Diskrepansmått anger hur mycket av modellens beteende som varierar för de här grupperna.",
|
||||
"performance": "Prestandamått",
|
||||
"performanceInfo": "Prestandamått används för att utvärdera den totala kvaliteten på din modell och kvaliteten på din modell i varje grupp. Skillnaden mellan de extrema värdena på prestandamåttet i grupperna rapporteras som en prestandadiskrepans."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Modelljämförelse",
|
||||
"howToRead": "Så här läser du det här diagrammet",
|
||||
"lower": "lägre",
|
||||
"higher": "högre",
|
||||
"howToReadText": "Det här diagrammet representerar var och en av {0}-modellerna som en valbar punkt. X-axeln representerar {1}, där {2} är bättre. Y-axeln representerar diskrepans, där lägre är bättre.",
|
||||
"insights": "Insikter",
|
||||
"insightsText1": "Diagrammet visar {0} och diskrepanser för {1} modeller.",
|
||||
"insightsText2": "{0} intervall från {1} till {2}. Diskrepansen är från {3} till {4}.",
|
||||
"insightsText3": "Den mest exakta modellen uppnår {0} av {1} och en diskrepans på {2}.",
|
||||
"insightsText4": "Modellen med lägsta diskrepans uppnår {0} av {1} och en diskrepans på {2}.",
|
||||
"disparityInOutcomes": "Diskrepans i förutsägelser",
|
||||
"disparityInPerformance": "Diskrepans i {0}",
|
||||
"howToMeasureDisparity": "Hur ska diskrepanser mätas?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "Modell {0}",
|
||||
"title": "Prestandadiskrepans",
|
||||
"globalPerformanceText": "Är den totala {0}",
|
||||
"performanceDisparityText": "Är diskrepansen i {0}",
|
||||
"editConfiguration": "Redigera konfigurationen",
|
||||
"backToComparisons": "Multimodellvy",
|
||||
"outcomesTitle": "Diskrepans i förutsägelser",
|
||||
"minTag": "Min",
|
||||
"maxTag": "Max",
|
||||
"groupLabel": "Undergrupp",
|
||||
"underestimationError": "Underförutsägelse",
|
||||
"underpredictionExplanation": "(uppskattat = 0, sant = 1)",
|
||||
"overpredictionExplanation": "(uppskattat = 1, sant = 0)",
|
||||
"overestimationError": "Överförutsägelse",
|
||||
"classificationOutcomesHowToRead": "I stapeldiagrammet visas markeringshastigheten i varje grupp, vilket innebär fraktionen punkter som klassificeras som 1.",
|
||||
"regressionOutcomesHowToRead": "I låddiagram visas fördelningen av förutsägelser i varje grupp. Enskilda datapunkter läggs ovanpå.",
|
||||
"classificationPerformanceHowToRead1": "Stapeldiagrammet visar distributionen av fel i varje grupp.",
|
||||
"classificationPerformanceHowToRead2": "Fel delas upp i överförutsägelsefel (förutsäger 1 när den sanna etiketten är 0) och underförutsägelsefel (förutsäger 0 när den sanna etiketten är 1).",
|
||||
"classificationPerformanceHowToRead3": "De rapporterade frekvenserna erhålls genom att dela antalet fel med den totala gruppstorleken.",
|
||||
"probabilityPerformanceHowToRead1": "Stapeldiagrammet visar ett absolut fel i varje grupp, uppdelat på överförutsägelse och underförutsägelse.",
|
||||
"probabilityPerformanceHowToRead2": "I varje exempel mäter vi skillnaden mellan förutsägelse och etikett. Om det är positivt, så kallar vi det överförutsägelse och om det är negativt så kallar vi det underförutsägelse.",
|
||||
"probabilityPerformanceHowToRead3": "Vi rapporterar summan av antalet överförutsägelser och summan av antalet underförutsägelser dividerat med den totala gruppstorleken.",
|
||||
"regressionPerformanceHowToRead": "Fel är skillnaden mellan förutsägelsen och etiketten. I låddiagram visas fördelningen av fel i varje grupp. Enskilda datapunkter läggs ovanpå.",
|
||||
"distributionOfPredictions": "Distribution av förutsägelser",
|
||||
"distributionOfErrors": "Distribution av fel",
|
||||
"tooltipPrediction": "Förutsägelse: {0}",
|
||||
"tooltipError": "Fel: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Efter vilka funktioner vill du utvärdera din modells rättvisa?",
|
||||
"body": "Rättvisa utvärderas i förhållande till diskrepanser i modellens beteende. Vi delar upp dina data enligt värdena för varje vald funktion och utvärderar hur modellens prestandamått och förutsägelser skiljer sig åt i dessa delningar.",
|
||||
"learnMore": "Läs mer",
|
||||
"summaryCategoricalCount": "Den här funktionen har {0} unika värden",
|
||||
"summaryNumericCount": "Den här numeriska funktionen sträcker sig från {0} till {1} och grupperas i {2} diskretiseringar.",
|
||||
"showCategories": "Visa alla",
|
||||
"hideCategories": "Minimera",
|
||||
"categoriesOverflow": " och {0} ytterligare kategorier",
|
||||
"editBinning": "Redigera grupper",
|
||||
"subgroups": "Undergrupper"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Noggrannhet",
|
||||
"precisionScore": "Precision",
|
||||
"recallScore": "Träffsäkerhet",
|
||||
"zeroOneLoss": "Zero-one-förlust",
|
||||
"specificityScore": "Specificitetspoäng",
|
||||
"missRate": "Missfrekvens",
|
||||
"falloutRate": "Utfallshastighet",
|
||||
"maxError": "Max fel",
|
||||
"meanAbsoluteError": "Medelvärde för absoluta fel",
|
||||
"meanSquaredError": " MSE (medelkvadratfel)",
|
||||
"meanSquaredLogError": "Medelkvadratvärde för loggfel",
|
||||
"medianAbsoluteError": "Medianvärde av absoluta fel",
|
||||
"average": "Genomsnittlig förutsägelse",
|
||||
"selectionRate": "Markeringshastighet",
|
||||
"overprediction": "Överförutsägelse",
|
||||
"underprediction": "Underförutsägelse",
|
||||
"r2_score": "R-kvadratvärde",
|
||||
"rms_error": "RMSE (rot-medelkvadratfel)",
|
||||
"auc": "Område under ROC-kurvan",
|
||||
"balancedRootMeanSquaredError": "Balanserad RMSE",
|
||||
"balancedAccuracy": "Balanserad noggrannhet",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Datapunktsfraktionen klassificeras korrekt.",
|
||||
"precisionDescription": "Fraktionen av datapunkter som klassificeras korrekt bland de som klassificerats som 1.",
|
||||
"recallDescription": "Fraktionen av datapunkter som klassificeras korrekt bland de vars sanna etikett är 1. Alternativa namn: sant positiv hastighet, känslighet.",
|
||||
"rmseDescription": "Kvadratroten för genomsnittet av kvadratfel.",
|
||||
"mseDescription": "Medelvärdet för kvadratfel.",
|
||||
"meanAbsoluteErrorDescription": "Medelvärdet av absoluta värden för fel. Mer robust för extremvärden än MSE.",
|
||||
"r2Description": "Fraktionen av varians i etiketterna som beskrivs av modellen.",
|
||||
"aucDescription": "Kvaliteten på förutsägelserna, visade som poäng, vid separering av positiva exempel från negativa exempel.",
|
||||
"balancedRMSEDescription": "Positiva och negativa exempel viktas om för att ha samma total viktning. Lämpligt om underliggande data är högt obalanserade.",
|
||||
"balancedAccuracyDescription": "Positiva och negativa exempel viktas om för att ha samma total viktning. Lämpligt om underliggande data är högt obalanserade.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Konfigurera diskretiseringar",
|
||||
"makeCategorical": "Hantera som kategoriskt",
|
||||
"save": "Spara",
|
||||
"cancel": "Avbryt",
|
||||
"numberOfBins": "Antal diskretiseringar:",
|
||||
"categoryHeader": "Diskretiserade värden:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "{0} sınıfı",
|
||||
"defaultFeatureNames": "{0} hassas özelliği",
|
||||
"defaultSingleFeatureName": "Hassas özellik",
|
||||
"defaultCustomMetricName": "Özel {0} ölçümü",
|
||||
"performanceTab": "Performans Eşitliği",
|
||||
"opportunityTab": "Fırsat Eşitliği",
|
||||
"modelComparisonTab": "Model karşılaştırması",
|
||||
"tableTab": "Ayrıntı Görünümü",
|
||||
"dataSpecifications": "Veri istatistikleri",
|
||||
"attributes": "Öznitelikler",
|
||||
"singleAttributeCount": "1 hassas özellik",
|
||||
"attributesCount": "{0} hassas özellik",
|
||||
"instanceCount": "{0} örnek",
|
||||
"close": "Kapat",
|
||||
"calculating": "Hesaplanıyor...",
|
||||
"performanceMetricLegacy": "Performans ölçümü",
|
||||
"errorOnInputs": "Giriş hatası. Hassas özellikler şu anda kategorik değerler olmalıdır. Lütfen değerleri bölmelenmiş kategorilere eşleyip yeniden deneyin.",
|
||||
"Performance": {
|
||||
"header": "Performansı nasıl ölçmek istiyorsunuz?",
|
||||
"modelMakes": "model şunları yapar:",
|
||||
"modelsMake": "modeller şunları yapar:",
|
||||
"body": "Verileriniz {0} etiket ve {2} {1} tahmininizi içeriyor. Bu bilgilere dayanarak, aşağıdaki ölçümleri kullanmanızı öneririz. Lütfen listeden bir ölçüm seçin.",
|
||||
"binaryClassifier": "ikili sınıflandırıcı",
|
||||
"probabilisticRegressor": "probit regresörü",
|
||||
"regressor": "regresör",
|
||||
"binary": "ikili",
|
||||
"continuous": "sürekli"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "Eşitlik, farklar temel alınarak ölçülür",
|
||||
"body": "Fark ölçümleri, modelinizin seçilen özelliklerde gösterdiği davranışın varyasyon miktarını ölçer. İki tür fark ölçümü vardır. Daha fazla ölçüm kullanıma sunulacak..."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "Belgeler"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "Geri",
|
||||
"next": "Sonraki"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "Hoş geldiniz",
|
||||
"fairnessDashboard": "Fairness panosu",
|
||||
"introBody": "Fairness panosu, modellerinizin performansı ve eşitliği arasındaki karşılaştırmaları değerlendirmenizi sağlar",
|
||||
"explanatoryStep": "Değerlendirmeyi ayarlamak için hassas bir özellik ve performans ölçümü belirtmeniz gerekir.",
|
||||
"getStarted": "Kullanmaya başlayın",
|
||||
"features": "Hassas özellikler",
|
||||
"featuresInfo": "Hassas özellikler, verilerinizi gruplara bölmek için kullanılır. Modelinizin bu gruplar arasındaki eşitliği, fark ölçümleriyle ölçülür. Fark ölçümleri, modelinizin bu gruplar arasında ne kadar farklı davrandığını ölçer.",
|
||||
"performance": "Performans ölçümü",
|
||||
"performanceInfo": "Performans ölçümleri, modelinizin genel ve her gruptaki kalitesini değerlendirmek için kullanılır. Gruplar arasındaki performans ölçümünün uç değerleri arasındaki fark, performans farkı olarak bildirilir."
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "Model karşılaştırması",
|
||||
"howToRead": "Bu grafiği okuma",
|
||||
"lower": "daha düşük",
|
||||
"higher": "daha yüksek",
|
||||
"howToReadText": "Bu grafik, {0} modelin her birini seçilebilir bir nokta olarak temsil eder. X ekseni {1} temsil eder, {2} değerler daha iyidir. Y ekseni farkı temsil eder, düşük değerler daha iyidir.",
|
||||
"insights": "İçgörüler",
|
||||
"insightsText1": "Grafik, {1} modelin {0} ve farkını gösterir.",
|
||||
"insightsText2": "{0}, {1} ile {2} aralığındayken, fark {3} ile {4} aralığındadır.",
|
||||
"insightsText3": "En doğru model, {0}/{1} ve {2} farkını alır.",
|
||||
"insightsText4": "En düşük farka sahip model {0}/{1} ve {2} farkını alır.",
|
||||
"disparityInOutcomes": "Tahmin farkı",
|
||||
"disparityInPerformance": "{0} farkı",
|
||||
"howToMeasureDisparity": "Fark nasıl ölçülür?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "{0} modeli",
|
||||
"title": "Performans farkı",
|
||||
"globalPerformanceText": "Genel {0}",
|
||||
"performanceDisparityText": "{0} farkıdır",
|
||||
"editConfiguration": "Yapılandırmayı düzenle",
|
||||
"backToComparisons": "Çoklu model görünümü",
|
||||
"outcomesTitle": "Tahmin farkı",
|
||||
"minTag": "Minimum",
|
||||
"maxTag": "Maksimum",
|
||||
"groupLabel": "Alt grup",
|
||||
"underestimationError": "Düşük tahmin",
|
||||
"underpredictionExplanation": "(tahmini = 0, true = 1)",
|
||||
"overpredictionExplanation": "(tahmini = 1, true = 0)",
|
||||
"overestimationError": "Yüksek tahmin",
|
||||
"classificationOutcomesHowToRead": "Çubuk grafik, her gruptaki seçim oranını gösterir. Bu, 1 olarak sınıflandırılan noktaların kesir değerini ifade eder.",
|
||||
"regressionOutcomesHowToRead": "Kutu grafikleri her gruptaki tahminlerin dağılımını gösterir. Veri noktaları ayrı ayrı olacak şekilde en üstte yer alır.",
|
||||
"classificationPerformanceHowToRead1": "Çubuk grafik, her gruptaki hataların dağılımını gösterir.",
|
||||
"classificationPerformanceHowToRead2": "Hatalar, yüksek tahmin hataları (1 tahmin edilirken gerçek etiketin 0 olması) ve düşük tahmin hataları (0 tahmin edilirken gerçek etiketin 1 olması) olarak ayrılır.",
|
||||
"classificationPerformanceHowToRead3": "Bildirilen oranlar, hata sayısı genel grup boyutuna bölünerek bulunur.",
|
||||
"probabilityPerformanceHowToRead1": "Çubuk grafik, her gruptaki ortalama mutlak hataları yüksek tahmin ve düşük tahmin olarak ayrılmış şekilde gösterir.",
|
||||
"probabilityPerformanceHowToRead2": "Her örnekte, tahmin ile etiket arasındaki farkı ölçeriz. Fark pozitifse yüksek tahmin, negatifse düşük tahmin olarak adlandırırız.",
|
||||
"probabilityPerformanceHowToRead3": "Yüksek tahmin hataları ile düşük tahmin hatalarının toplamını toplam grup boyutuna bölerek bildiririz.",
|
||||
"regressionPerformanceHowToRead": "Hata, tahmin ile etiket arasındaki farktır. Kutu grafikleri, her gruptaki hataların dağılımını gösterir. Veri noktaları ayrı ayrı olacak şekilde en üste yer alır.",
|
||||
"distributionOfPredictions": "Tahmin dağılımı",
|
||||
"distributionOfErrors": "Hata dağılımı",
|
||||
"tooltipPrediction": "Tahmin: {0}",
|
||||
"tooltipError": "Hata: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "Modelinizin eşitliğini hangi özelliklerle birlikte değerlendirmek istiyorsunuz?",
|
||||
"body": "Eşitlik, modelinizin davranışındaki farklara yönelik olarak değerlendirilir. Verilerinizi seçilen her bir özelliğin değerlerine göre böleriz ve modelinizin performans ölçümü ile tahminlerinin bu bölmeler arasında nasıl farklılık gösterdiğini değerlendiririz.",
|
||||
"learnMore": "Daha fazla bilgi",
|
||||
"summaryCategoricalCount": "Bu özellik, {0} benzersiz değer içeriyor",
|
||||
"summaryNumericCount": "Bu sayısal özellik {0} ile {1} aralığındadır ve {2} bölmeye gruplandırılmıştır.",
|
||||
"showCategories": "Tümünü göster",
|
||||
"hideCategories": "Daralt",
|
||||
"categoriesOverflow": " ve {0} ek kategori",
|
||||
"editBinning": "Grupları düzenle",
|
||||
"subgroups": "Alt gruplar"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "Doğruluk",
|
||||
"precisionScore": "Duyarlık",
|
||||
"recallScore": "Geri çağırma",
|
||||
"zeroOneLoss": "0-1 kayıp",
|
||||
"specificityScore": "Belirginlik puanı",
|
||||
"missRate": "İsabetsizlik oranı",
|
||||
"falloutRate": "Hata oranı",
|
||||
"maxError": "Maksimum hata sayısı",
|
||||
"meanAbsoluteError": "Ortalama mutlak hata sayısı",
|
||||
"meanSquaredError": " MSE (ortalama hata karesi)",
|
||||
"meanSquaredLogError": "Ortalama günlük hatası karesi",
|
||||
"medianAbsoluteError": "Ortanca mutlak hata",
|
||||
"average": "Ortalama tahmin",
|
||||
"selectionRate": "Seçim oranı",
|
||||
"overprediction": "Yüksek tahmin",
|
||||
"underprediction": "Düşük tahmin",
|
||||
"r2_score": "R kare puanı",
|
||||
"rms_error": "RMSE (kök ortalama hata karesi)",
|
||||
"auc": "ROC eğrisinin altındaki alan",
|
||||
"balancedRootMeanSquaredError": "Dengeli RMSE",
|
||||
"balancedAccuracy": "Dengeli doğruluk",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "Doğru olarak sınıflandırılmış veri noktalarının kesir değeri.",
|
||||
"precisionDescription": "1 olarak sınıflandırılanların arasında doğru şekilde sınıflandırılmış veri noktalarının kesir değeri.",
|
||||
"recallDescription": "Gerçek etiketi 1 olanların arasında doğru bir şekilde sınıflandırılmış veri noktalarının kesir değeri. Alternatif adlar: gerçek pozitif oran, hassasiyet.",
|
||||
"rmseDescription": "Hata karelerinin ortalamasının karekökü.",
|
||||
"mseDescription": "Hata karelerinin ortalaması.",
|
||||
"meanAbsoluteErrorDescription": "Hata mutlak değerlerinin ortalaması. Aykırı değerlere karşı MSE'den daha dayanıklıdır.",
|
||||
"r2Description": "Model tarafından açıklanan etiketlerdeki varyansın kesir değeri.",
|
||||
"aucDescription": "Olumlu örnekleri olumsuz örneklerden ayırırken puan olarak görüntülenen tahminlerin kalitesi.",
|
||||
"balancedRMSEDescription": "Pozitif ve negatif örnekler eşit toplam ağırlığa sahip olacak şekilde yeniden ağırlıklandırılır. Bu, temel alınan verilerin fazla dengesiz olması durumunda uygundur.",
|
||||
"balancedAccuracyDescription": "Pozitif ve negatif örnekler eşit toplam ağırlığa sahip olacak şekilde yeniden ağırlıklandırılır. Bu, temel alınan verilerin fazla dengesiz olması durumunda uygundur.",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "Bölmeleri yapılandır",
|
||||
"makeCategorical": "Kategorik olarak değerlendir",
|
||||
"save": "Kaydet",
|
||||
"cancel": "İptal",
|
||||
"numberOfBins": "Bölme sayısı:",
|
||||
"categoryHeader": "Bölme değerleri:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "类 {0}",
|
||||
"defaultFeatureNames": "敏感特征 {0}",
|
||||
"defaultSingleFeatureName": "敏感特征",
|
||||
"defaultCustomMetricName": "自定义指标 {0}",
|
||||
"performanceTab": "性能的公平性",
|
||||
"opportunityTab": "机会的公平性",
|
||||
"modelComparisonTab": "模型比较",
|
||||
"tableTab": "详细信息视图",
|
||||
"dataSpecifications": "数据统计信息",
|
||||
"attributes": "特性",
|
||||
"singleAttributeCount": "1 个敏感特征",
|
||||
"attributesCount": "{0} 个敏感特征",
|
||||
"instanceCount": "{0} 个实例",
|
||||
"close": "关闭",
|
||||
"calculating": "正在计算...",
|
||||
"performanceMetricLegacy": "性能指标",
|
||||
"errorOnInputs": "输入出错。目前敏感特征必须是分类值。请将值映射到已装箱的类别，然后重试。",
|
||||
"Performance": {
|
||||
"header": "你想如何衡量性能?",
|
||||
"modelMakes": "模型",
|
||||
"modelsMake": "模型",
|
||||
"body": "你的数据包含 {0} 个标签和 {2} {1} 预测。根据此信息，我们推荐以下指标。请从列表中选择一个指标。",
|
||||
"binaryClassifier": "二元分类器",
|
||||
"probabilisticRegressor": "Probit 回归量",
|
||||
"regressor": "回归量",
|
||||
"binary": "二进制",
|
||||
"continuous": "连续"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "根据差异衡量公平性",
|
||||
"body": "差异指标量化了模型在所选特征之间的行为变化。有两种类型的差异指标: 即将推出更多..."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "文档"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "上一步",
|
||||
"next": "下一步"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "欢迎使用",
|
||||
"fairnessDashboard": "Fairness 仪表板",
|
||||
"introBody": "借助 Fairness 仪表板，可以评估模型的性能和公平性之间的权衡",
|
||||
"explanatoryStep": "若要设置评估，需要指定一个敏感特征和一个性能指标。",
|
||||
"getStarted": "开始使用",
|
||||
"features": "敏感特征",
|
||||
"featuresInfo": "使用敏感特征可以将数据拆分为组。通过差异指标衡量模型在这些组之间的公平性。差异指标将量化模型在这些组之间的行为变化。",
|
||||
"performance": "性能指标",
|
||||
"performanceInfo": "性能指标用于评估模型的总体质量以及在每个组中的模型质量。各组的性能指标极值之间的差异称为性能差异。"
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "模型比较",
|
||||
"howToRead": "如何阅读此图表",
|
||||
"lower": "偏低",
|
||||
"higher": "偏高",
|
||||
"howToReadText": "此图表将每个 {0} 模型表示为一个可选择的点。X 轴表示 {1}，越 {2} 越好。Y 轴表示差异，越低越好。",
|
||||
"insights": "见解",
|
||||
"insightsText1": "此图表显示 {0} 和 {1} 个模型的差异。",
|
||||
"insightsText2": "{0} 的范围是从 {1} 到 {2}。差异的范围是从 {3} 到 {4}。",
|
||||
"insightsText3": "最准确的模型可达到 {1} 的 {0} 和 {2} 的差异。",
|
||||
"insightsText4": "最低差异模型可达到 {1} 的 {0} 和 {2} 的差异。",
|
||||
"disparityInOutcomes": "预测差异",
|
||||
"disparityInPerformance": "{0} 差异",
|
||||
"howToMeasureDisparity": "如何衡量差异?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "模型 {0}",
|
||||
"title": "性能差异",
|
||||
"globalPerformanceText": "是总体 {0}",
|
||||
"performanceDisparityText": "是 {0} 的差异",
|
||||
"editConfiguration": "编辑配置",
|
||||
"backToComparisons": "多模型视图",
|
||||
"outcomesTitle": "预测差异",
|
||||
"minTag": "最小值",
|
||||
"maxTag": "最大值",
|
||||
"groupLabel": "子组",
|
||||
"underestimationError": "过低预测",
|
||||
"underpredictionExplanation": "(预测值 = 0，true = 1)",
|
||||
"overpredictionExplanation": "(预测值 = 1，true = 0)",
|
||||
"overestimationError": "过高预测",
|
||||
"classificationOutcomesHowToRead": "条形图显示每组的选择率，表示分类为 1 的点的分数。",
|
||||
"regressionOutcomesHowToRead": "箱形图显示每个组的预测值分布。各个数据点在顶部重叠。",
|
||||
"classificationPerformanceHowToRead1": "条形图显示每个组的误差分布。",
|
||||
"classificationPerformanceHowToRead2": "误差分为“过高预测”误差(true 标签为 0 时预测值为 1)和“过低预测”误差(true 标签为 1 时预测值为 0)。",
|
||||
"classificationPerformanceHowToRead3": "通过将误差数除以组总大小得到报告率。",
|
||||
"probabilityPerformanceHowToRead1": "条形图显示各组的平均绝对误差，分为“过高预测”和“过低预测”。",
|
||||
"probabilityPerformanceHowToRead2": "我们对每个样本测量预测值和标签值之间的差异。如果是正值，则视为过高预测，如果是负值，则视为过低预测。",
|
||||
"probabilityPerformanceHowToRead3": "我们报告“过高预测”误差总数与“过低预测”误差总数除以组总大小后的值。",
|
||||
"regressionPerformanceHowToRead": "误差是预测值和标签值之间的差异。箱形图显示每个组的误差分布。各个数据点在顶部重叠。",
|
||||
"distributionOfPredictions": "预测分布",
|
||||
"distributionOfErrors": "误差分布",
|
||||
"tooltipPrediction": "预测: {0}",
|
||||
"tooltipError": "误差: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "你想要通过哪些特征评估模型的公平性?",
|
||||
"body": "根据模型行为差异评估公平性。我们将根据每个所选特征的值来拆分数据，并评估这些拆分数据的模型性能指标和预测之间的差异。",
|
||||
"learnMore": "了解更多",
|
||||
"summaryCategoricalCount": "此特征具有 {0} 个唯一值",
|
||||
"summaryNumericCount": "此数值特征的范围从 {0} 到 {1}，并被划分到 {2} 个箱中。",
|
||||
"showCategories": "全部显示",
|
||||
"hideCategories": "折叠",
|
||||
"categoriesOverflow": "和 {0} 个附加类别",
|
||||
"editBinning": "编辑组",
|
||||
"subgroups": "子组"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "准确度",
|
||||
"precisionScore": "精准率",
|
||||
"recallScore": "召回",
|
||||
"zeroOneLoss": "0-1 损失",
|
||||
"specificityScore": "特异性分数",
|
||||
"missRate": "漏检率",
|
||||
"falloutRate": "错检率",
|
||||
"maxError": "最大误差",
|
||||
"meanAbsoluteError": "平均绝对误差",
|
||||
"meanSquaredError": "MSE (均方误差)",
|
||||
"meanSquaredLogError": "均方对数误差",
|
||||
"medianAbsoluteError": "中值绝对误差",
|
||||
"average": "平均预测",
|
||||
"selectionRate": "选择率",
|
||||
"overprediction": "过高预测",
|
||||
"underprediction": "过低预测",
|
||||
"r2_score": "R 平方分数",
|
||||
"rms_error": "RMSE (均方根误差)",
|
||||
"auc": "ROC 曲线下的面积",
|
||||
"balancedRootMeanSquaredError": "平衡 RMSE",
|
||||
"balancedAccuracy": "平衡准确度",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "正确分类的数据点的分数。",
|
||||
"precisionDescription": "分类为 1 的数据中正确分类的数据点的分数。",
|
||||
"recallDescription": "真实标签为 1 的数据中正确分类的数据点的分数。其他名称: 真正率、敏感度。",
|
||||
"rmseDescription": "方差平均值的平方根。",
|
||||
"mseDescription": "平方误差的平均值。",
|
||||
"meanAbsoluteErrorDescription": "误差绝对值的平均值。与 MSE 相比，对离群值更具健壮性。",
|
||||
"r2Description": "模型解释的标签差异的分数。",
|
||||
"aucDescription": "将正负样本分隔开的预测质量(以分数形式表示)。",
|
||||
"balancedRMSEDescription": "对正样本和负样本重新进行加权以具有同等的总权重。适用于基础数据高度不均衡的情况。",
|
||||
"balancedAccuracyDescription": "对正样本和负样本重新进行加权以具有同等的总权重。适用于基础数据高度不均衡的情况。",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "配置箱",
|
||||
"makeCategorical": "视为类别",
|
||||
"save": "保存",
|
||||
"cancel": "取消",
|
||||
"numberOfBins": "箱数:",
|
||||
"categoryHeader": "箱值:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,154 +0,0 @@
|
|||
{
|
||||
"loremIpsum": "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.",
|
||||
"defaultClassNames": "類別 {0}",
|
||||
"defaultFeatureNames": "敏感性特徵 {0}",
|
||||
"defaultSingleFeatureName": "敏感性特徵",
|
||||
"defaultCustomMetricName": "自訂指標 {0}",
|
||||
"performanceTab": "成效公平性",
|
||||
"opportunityTab": "商機公平性",
|
||||
"modelComparisonTab": "模型比較",
|
||||
"tableTab": "詳細資料檢視",
|
||||
"dataSpecifications": "資料統計資料",
|
||||
"attributes": "屬性",
|
||||
"singleAttributeCount": "1 個敏感性特徵",
|
||||
"attributesCount": "{0} 個敏感性特徵",
|
||||
"instanceCount": "{0} 個執行個體",
|
||||
"close": "關閉",
|
||||
"calculating": "正在計算...",
|
||||
"performanceMetricLegacy": "成效指標",
|
||||
"errorOnInputs": "輸入發生錯誤。敏感性特徵目前必須是類別目錄值。請將值對應至組界類別，然後重試。",
|
||||
"Performance": {
|
||||
"header": "要如何測量成效?",
|
||||
"modelMakes": "模型會建立",
|
||||
"modelsMake": "模型會建立",
|
||||
"body": "您的資料包含 {0} 個標籤和 {2} 個 {1} 預測。根據這些資訊，建議您使用下列指標。請從清單中選取一個指標。",
|
||||
"binaryClassifier": "二元分類器",
|
||||
"probabilisticRegressor": "概率迴歸輸入變數",
|
||||
"regressor": "迴歸輸入變數",
|
||||
"binary": "二元",
|
||||
"continuous": "連續"
|
||||
},
|
||||
"Parity": {
|
||||
"header": "依據差距進行測量的公平性",
|
||||
"body": "差距指標會在所選功能間量化模型行為的變化。有兩種差距指標: 即將推出更多...."
|
||||
},
|
||||
"Header": {
|
||||
"title": "Fairness",
|
||||
"documentation": "文件"
|
||||
},
|
||||
"Footer": {
|
||||
"back": "上一步",
|
||||
"next": "下一步"
|
||||
},
|
||||
"Intro": {
|
||||
"welcome": "歡迎使用",
|
||||
"fairnessDashboard": "Fairness 儀表板",
|
||||
"introBody": "Fairness 儀表板可讓您評定模型成效與公平性之間的權衡",
|
||||
"explanatoryStep": "若要設定評定，您必須指定敏感性特徵和成效指標。",
|
||||
"getStarted": "立即開始",
|
||||
"features": "敏感性特徵",
|
||||
"featuresInfo": "敏感性特徵會用於將資料分割成群組。這些群組的模型公平性會以差距指標測量。差距指標會量化模型行為在各群組之間的差異。",
|
||||
"performance": "成效指標",
|
||||
"performanceInfo": "成效指標可用於評估模型的整體品質，以及每個群組中的模型品質。各群組成效指標極端值之間的差異，會在成效中回報為差距。"
|
||||
},
|
||||
"ModelComparison": {
|
||||
"title": "模型比較",
|
||||
"howToRead": "如何閱讀此圖表",
|
||||
"lower": "低於",
|
||||
"higher": "高於",
|
||||
"howToReadText": "此圖表會將每個 {0} 模型表示為可選取的點。X 軸代表 {1}，越{2}越好。Y 軸代表差距，越低越好。",
|
||||
"insights": "見解",
|
||||
"insightsText1": "此圖表顯示 {0} 與 {1} 模型的差距。",
|
||||
"insightsText2": "{0} 的範圍為 {1} 到 {2}。差距的範圍為 {3} 到 {4}。",
|
||||
"insightsText3": "最準確的模型達到 {1} 的 {0} 和 {2} 的差距。",
|
||||
"insightsText4": "最低差距模型達到 {1} 的 {0} 和 {2} 的差距。",
|
||||
"disparityInOutcomes": "預測中的差距",
|
||||
"disparityInPerformance": "{0} 中的差距",
|
||||
"howToMeasureDisparity": "應如何測量差距?"
|
||||
},
|
||||
"Report": {
|
||||
"modelName": "模型 {0}",
|
||||
"title": "成效差距",
|
||||
"globalPerformanceText": "為整體 {0}",
|
||||
"performanceDisparityText": "為 {0} 中的差距",
|
||||
"editConfiguration": "編輯組態",
|
||||
"backToComparisons": "多重模型檢視",
|
||||
"outcomesTitle": "預測中的差距",
|
||||
"minTag": "最小值",
|
||||
"maxTag": "最大值",
|
||||
"groupLabel": "子群組",
|
||||
"underestimationError": "低於預測",
|
||||
"underpredictionExplanation": "(已預測 = 0，True = 1)",
|
||||
"overpredictionExplanation": "(已預測 = 1，True = 0)",
|
||||
"overestimationError": "高於預測",
|
||||
"classificationOutcomesHowToRead": "橫條圖會顯示各群組中的選擇率，表示分類為 1 的資料點。",
|
||||
"regressionOutcomesHowToRead": "盒狀圖會顯示各群組中的預測分佈。個別資料點會重疊在上方。",
|
||||
"classificationPerformanceHowToRead1": "橫條圖會顯示各群組中的誤差分佈。",
|
||||
"classificationPerformanceHowToRead2": "錯誤會分割成高於預測錯誤 (當 True 標籤為 0 時，預測 1) 及低於預測錯誤 (當 True 標籤為 1 時，預測 0)。",
|
||||
"classificationPerformanceHowToRead3": "以整體群組大小除以錯誤數目來取得回報比率。",
|
||||
"probabilityPerformanceHowToRead1": "橫條圖會顯示各群組的平均絕對誤差，分割成高於預測及低於預測。",
|
||||
"probabilityPerformanceHowToRead2": "在每個範例中，我們都會測量預測與標籤之間的差異。若是正數，我們稱為高於預測; 若是負數，我們稱為低於預測。",
|
||||
"probabilityPerformanceHowToRead3": "我們回報了高於預測誤差總和及低於預測誤差總和除以整體群組大小。",
|
||||
"regressionPerformanceHowToRead": "誤差是預測與標籤之間的差異。盒狀圖會顯示各群組中的誤差分佈。個別資料點會重疊在上方。",
|
||||
"distributionOfPredictions": "預測分佈",
|
||||
"distributionOfErrors": "誤差分佈",
|
||||
"tooltipPrediction": "預測: {0}",
|
||||
"tooltipError": "誤差: {0}"
|
||||
},
|
||||
"Feature": {
|
||||
"header": "要依循哪些特徵評估模型公平性?",
|
||||
"body": "公平性會根據模型行為中的差距部分進行評估。我們會依據各個所選特徵的值來分割資料，並評估在這些分割中，模型的成效指標與預測有何差異。",
|
||||
"learnMore": "深入了解",
|
||||
"summaryCategoricalCount": "此特徵有 {0} 個唯一的值",
|
||||
"summaryNumericCount": "此數值特徵的範圍為 {0} 到 {1}，並分成 {2} 個組界。",
|
||||
"showCategories": "全部顯示",
|
||||
"hideCategories": "摺疊",
|
||||
"categoriesOverflow": " 和其他 {0} 個類別",
|
||||
"editBinning": "編輯群組",
|
||||
"subgroups": "子群組"
|
||||
},
|
||||
"Metrics": {
|
||||
"accuracyScore": "正確性",
|
||||
"precisionScore": "精確度",
|
||||
"recallScore": "召回率",
|
||||
"zeroOneLoss": "0-1 損失",
|
||||
"specificityScore": "具體程度分數",
|
||||
"missRate": "失誤率",
|
||||
"falloutRate": "散落率",
|
||||
"maxError": "最大誤差",
|
||||
"meanAbsoluteError": "平均絕對誤差",
|
||||
"meanSquaredError": " MSE (均方誤差)",
|
||||
"meanSquaredLogError": "平均平方對數誤差",
|
||||
"medianAbsoluteError": "絕對誤差中位數",
|
||||
"average": "平均預測",
|
||||
"selectionRate": "選擇率",
|
||||
"overprediction": "高於預測",
|
||||
"underprediction": "低於預測",
|
||||
"r2_score": "R 平方分數",
|
||||
"rms_error": "RMSE (平均平方根誤差)",
|
||||
"auc": "ROC 曲線下的區域",
|
||||
"balancedRootMeanSquaredError": "平衡的 RMSE",
|
||||
"balancedAccuracy": "平衡的正確性",
|
||||
"f1Score": "F1-Score",
|
||||
"logLoss": "Log Loss",
|
||||
"accuracyDescription": "正確分類的資料點。",
|
||||
"precisionDescription": "在分類為 1 的資料點當中，正確分類的資料點。",
|
||||
"recallDescription": "在 True 標籤為 1 的資料點當中，正確分類的資料點。替代名稱: 確判為真率、敏感度。",
|
||||
"rmseDescription": "平方誤差平均的平方根。",
|
||||
"mseDescription": "平方誤差的平均。",
|
||||
"meanAbsoluteErrorDescription": "誤差絕對值的平均。比 MSE 更不易受極端值影響。",
|
||||
"r2Description": "模型所解釋的標籤變異數。",
|
||||
"aucDescription": "從負值範例分隔出正值範例的預測品質，以分數檢視。",
|
||||
"balancedRMSEDescription": "正值與負值範例會重新加權，使總權重相等。適用於高度不平衡的基礎資料。",
|
||||
"balancedAccuracyDescription": "正值與負值範例會重新加權，使總權重相等。適用於高度不平衡的基礎資料。",
|
||||
"f1ScoreDescription": "F1-Score is the harmonic mean of precision and recall."
|
||||
},
|
||||
"BinDialog": {
|
||||
"header": "設定組界",
|
||||
"makeCategorical": "視為類別",
|
||||
"save": "儲存",
|
||||
"cancel": "取消",
|
||||
"numberOfBins": "組界數目:",
|
||||
"categoryHeader": "組界值:"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { INumericRange, RangeTypes } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -12,7 +13,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../Localization/localization";
|
||||
import { BinnedResponseBuilder } from "../util/BinnedResponseBuilder";
|
||||
import { IBinnedResponse } from "../util/IBinnedResponse";
|
||||
|
||||
|
|
@ -44,13 +44,13 @@ export class BinDialog extends React.PureComponent<
|
|||
return (
|
||||
<div className={styles.frame}>
|
||||
<Text variant={"xLargePlus"} className={styles.header}>
|
||||
{localization.BinDialog.header}
|
||||
{localization.Fairness.BinDialog.header}
|
||||
</Text>
|
||||
<div className={styles.main}>
|
||||
<div className={styles.controls}>
|
||||
{this.props.range.rangeType === RangeTypes.Integer && (
|
||||
<Checkbox
|
||||
label={localization.BinDialog.makeCategorical}
|
||||
label={localization.Fairness.BinDialog.makeCategorical}
|
||||
checked={this.state.rangeType === RangeTypes.Categorical}
|
||||
onChange={this.toggleCategorical}
|
||||
/>
|
||||
|
|
@ -71,7 +71,7 @@ export class BinDialog extends React.PureComponent<
|
|||
},
|
||||
spinButtonWrapper: { maxWidth: "98px" }
|
||||
}}
|
||||
label={localization.BinDialog.numberOfBins}
|
||||
label={localization.Fairness.BinDialog.numberOfBins}
|
||||
min={BinDialog.minBins}
|
||||
max={BinDialog.maxBins}
|
||||
value={this.state.array.length.toString()}
|
||||
|
|
@ -82,7 +82,7 @@ export class BinDialog extends React.PureComponent<
|
|||
</div>
|
||||
)}
|
||||
</div>
|
||||
<Text>{localization.BinDialog.categoryHeader}</Text>
|
||||
<Text>{localization.Fairness.BinDialog.categoryHeader}</Text>
|
||||
<div className={styles.scrollArea}>
|
||||
{this.state.labelArray.map((val, i) => {
|
||||
return (
|
||||
|
|
@ -96,12 +96,12 @@ export class BinDialog extends React.PureComponent<
|
|||
<div className={styles.buttons}>
|
||||
<PrimaryButton
|
||||
className={styles.saveButton}
|
||||
text={localization.BinDialog.save}
|
||||
text={localization.Fairness.BinDialog.save}
|
||||
onClick={this.onSave}
|
||||
/>
|
||||
<DefaultButton
|
||||
className={styles.cancelButton}
|
||||
text={localization.BinDialog.cancel}
|
||||
text={localization.Fairness.BinDialog.cancel}
|
||||
onClick={this.props.onCancel}
|
||||
/>
|
||||
</div>
|
||||
|
|
|
|||
|
|
@ -1,11 +1,10 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../Localization/localization";
|
||||
|
||||
import { DataSpecificationBladeStyles } from "./DataSpecificationBlade.styles";
|
||||
|
||||
export interface IDataSpecProps {
|
||||
|
|
@ -21,19 +20,19 @@ export class DataSpecificationBlade extends React.PureComponent<
|
|||
return (
|
||||
<div className={styles.frame}>
|
||||
<Text variant={"small"} className={styles.title} block>
|
||||
{localization.dataSpecifications}
|
||||
{localization.Fairness.dataSpecifications}
|
||||
</Text>
|
||||
<Text variant={"small"} className={styles.text} block>
|
||||
{this.props.featureNames.length === 1
|
||||
? localization.singleAttributeCount
|
||||
? localization.Fairness.singleAttributeCount
|
||||
: localization.formatString(
|
||||
localization.attributesCount,
|
||||
localization.Fairness.attributesCount,
|
||||
this.props.featureNames.length
|
||||
)}
|
||||
</Text>
|
||||
<Text variant={"small"} className={styles.text} block>
|
||||
{localization.formatString(
|
||||
localization.instanceCount,
|
||||
localization.Fairness.instanceCount,
|
||||
this.props.numberRows
|
||||
)}
|
||||
</Text>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { INumericRange, RangeTypes } from "@responsible-ai/mlchartlib";
|
||||
import {
|
||||
ActionButton,
|
||||
|
|
@ -15,7 +16,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../Localization/localization";
|
||||
import { IBinnedResponse } from "../util/IBinnedResponse";
|
||||
|
||||
import { BinDialog } from "./BinDialog";
|
||||
|
|
@ -61,13 +61,13 @@ export class FeatureTab extends React.PureComponent<IFeatureTabProps, IState> {
|
|||
{
|
||||
key: "feature",
|
||||
minWidth: 75,
|
||||
name: localization.Intro.features,
|
||||
name: localization.Fairness.Intro.features,
|
||||
onRender: this.renderFeatureNameCell
|
||||
},
|
||||
{
|
||||
key: "subgroup",
|
||||
minWidth: 130,
|
||||
name: localization.Feature.subgroups,
|
||||
name: localization.Fairness.Feature.subgroups,
|
||||
onRender: this.renderSubGroupCell
|
||||
}
|
||||
];
|
||||
|
|
@ -112,10 +112,10 @@ export class FeatureTab extends React.PureComponent<IFeatureTabProps, IState> {
|
|||
</Modal>
|
||||
<Stack className={styles.main}>
|
||||
<Text variant={"mediumPlus"} className={styles.header} block>
|
||||
{localization.Feature.header}
|
||||
{localization.Fairness.Feature.header}
|
||||
</Text>
|
||||
<Text className={styles.textBody} block>
|
||||
{localization.Feature.body}
|
||||
{localization.Fairness.Feature.body}
|
||||
</Text>
|
||||
<DetailsList
|
||||
items={this.props.featureBins}
|
||||
|
|
@ -169,7 +169,7 @@ export class FeatureTab extends React.PureComponent<IFeatureTabProps, IState> {
|
|||
{item.rangeType === RangeTypes.Categorical && (
|
||||
<Text variant={"mediumPlus"} className={styles.valueCount} block>
|
||||
{localization.formatString(
|
||||
localization.Feature.summaryCategoricalCount,
|
||||
localization.Fairness.Feature.summaryCategoricalCount,
|
||||
item.array.length
|
||||
)}
|
||||
</Text>
|
||||
|
|
@ -177,7 +177,7 @@ export class FeatureTab extends React.PureComponent<IFeatureTabProps, IState> {
|
|||
{item.rangeType !== RangeTypes.Categorical && (
|
||||
<Text variant={"mediumPlus"} className={styles.valueCount} block>
|
||||
{localization.formatString(
|
||||
localization.Feature.summaryNumericCount,
|
||||
localization.Fairness.Feature.summaryNumericCount,
|
||||
(this.props.dashboardContext.modelMetadata.featureRanges[
|
||||
index
|
||||
] as INumericRange).min,
|
||||
|
|
@ -196,7 +196,7 @@ export class FeatureTab extends React.PureComponent<IFeatureTabProps, IState> {
|
|||
iconProps={{ iconName: "Edit" }}
|
||||
onClick={this.editBins.bind(this, index)}
|
||||
>
|
||||
{localization.Feature.editBinning}
|
||||
{localization.Fairness.Feature.editBinning}
|
||||
</ActionButton>
|
||||
)}
|
||||
</>
|
||||
|
|
|
|||
|
|
@ -1,10 +1,10 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { ActionButton, Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../Localization/localization";
|
||||
import { IBinnedResponse } from "../util/IBinnedResponse";
|
||||
|
||||
import { FeatureTabStyles } from "./FeatureTab.styles";
|
||||
|
|
@ -49,8 +49,8 @@ export class FeatureTabSubGroup extends React.Component<
|
|||
onClick={this.toggle}
|
||||
>
|
||||
{this.state.expanded
|
||||
? localization.Feature.hideCategories
|
||||
: localization.Feature.showCategories}
|
||||
? localization.Fairness.Feature.hideCategories
|
||||
: localization.Fairness.Feature.showCategories}
|
||||
</ActionButton>
|
||||
)}
|
||||
</div>
|
||||
|
|
|
|||
|
|
@ -1,14 +1,13 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
DefaultButton,
|
||||
PrimaryButton
|
||||
} from "office-ui-fabric-react/lib/Button";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../Localization/localization";
|
||||
|
||||
import { WizardFooterStyles } from "./WizardFooter.styles";
|
||||
|
||||
export interface IWizardFooterProps {
|
||||
|
|
@ -23,13 +22,13 @@ export class WizardFooter extends React.PureComponent<IWizardFooterProps> {
|
|||
<div className={styles.frame}>
|
||||
<PrimaryButton
|
||||
className={styles.next}
|
||||
text={localization.Footer.next}
|
||||
text={localization.Fairness.Footer.next}
|
||||
onClick={this.props.onNext}
|
||||
/>
|
||||
{!!this.props.onPrevious && (
|
||||
<DefaultButton
|
||||
className={styles.back}
|
||||
text={localization.Footer.back}
|
||||
text={localization.Fairness.Footer.back}
|
||||
onClick={this.props.onPrevious}
|
||||
/>
|
||||
)}
|
||||
|
|
|
|||
|
|
@ -1,7 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "../Localization/localization";
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
|
||||
export interface IParityOption {
|
||||
key: string;
|
||||
|
|
@ -18,31 +18,32 @@ export enum ParityModes {
|
|||
|
||||
export const parityOptions: { [key: string]: IParityOption } = {
|
||||
recall_score: {
|
||||
description: localization.Metrics.equalOpportunityDifferenceDescription,
|
||||
description:
|
||||
localization.Fairness.Metrics.equalOpportunityDifferenceDescription,
|
||||
key: "recall_score",
|
||||
parityMetric: "recall_score",
|
||||
parityMode: ParityModes.Difference,
|
||||
title: localization.Metrics.equalOpportunityDifference
|
||||
title: localization.Fairness.Metrics.equalOpportunityDifference
|
||||
},
|
||||
selection_rate: {
|
||||
description: localization.Metrics.parityDifferenceDescription,
|
||||
description: localization.Fairness.Metrics.parityDifferenceDescription,
|
||||
key: "selection_rate",
|
||||
parityMetric: "selection_rate",
|
||||
parityMode: ParityModes.Difference,
|
||||
title: localization.Metrics.parityDifference
|
||||
title: localization.Fairness.Metrics.parityDifference
|
||||
},
|
||||
selection_rate_ratio: {
|
||||
description: localization.Metrics.parityRatioDescription,
|
||||
description: localization.Fairness.Metrics.parityRatioDescription,
|
||||
key: "selection_rate_ratio",
|
||||
parityMetric: "selection_rate",
|
||||
parityMode: ParityModes.Ratio,
|
||||
title: localization.Metrics.parityRatio
|
||||
title: localization.Fairness.Metrics.parityRatio
|
||||
},
|
||||
zero_one_loss: {
|
||||
description: localization.Metrics.errorRateDifferenceDescription,
|
||||
description: localization.Fairness.Metrics.errorRateDifferenceDescription,
|
||||
key: "zero_one_loss",
|
||||
parityMetric: "zero_one_loss",
|
||||
parityMode: ParityModes.Difference,
|
||||
title: localization.Metrics.errorRateDifference
|
||||
title: localization.Fairness.Metrics.errorRateDifference
|
||||
}
|
||||
};
|
||||
|
|
|
|||
|
|
@ -1,7 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "./../Localization/localization";
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
|
||||
export interface IPerformanceOption {
|
||||
key: string;
|
||||
|
|
@ -16,192 +16,192 @@ export interface IPerformanceOption {
|
|||
|
||||
export const performanceOptions: { [key: string]: IPerformanceOption } = {
|
||||
accuracy_score: {
|
||||
description: localization.Metrics.accuracyDescription,
|
||||
description: localization.Fairness.Metrics.accuracyDescription,
|
||||
isMinimization: false,
|
||||
isPercentage: true,
|
||||
key: "accuracy_score",
|
||||
title: localization.Metrics.accuracyScore,
|
||||
title: localization.Fairness.Metrics.accuracyScore,
|
||||
userVisible: true
|
||||
},
|
||||
auc: {
|
||||
description: localization.Metrics.aucDescription,
|
||||
description: localization.Fairness.Metrics.aucDescription,
|
||||
isMinimization: false,
|
||||
isPercentage: false,
|
||||
key: "auc",
|
||||
title: localization.Metrics.auc,
|
||||
title: localization.Fairness.Metrics.auc,
|
||||
userVisible: true
|
||||
},
|
||||
average: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: false,
|
||||
isPercentage: false,
|
||||
key: "average",
|
||||
title: localization.Metrics.average
|
||||
title: localization.Fairness.Metrics.average
|
||||
},
|
||||
balanced_accuracy_score: {
|
||||
description: localization.Metrics.balancedAccuracyDescription,
|
||||
description: localization.Fairness.Metrics.balancedAccuracyDescription,
|
||||
isMinimization: false,
|
||||
isPercentage: true,
|
||||
key: "balanced_accuracy_score",
|
||||
title: localization.Metrics.balancedAccuracy,
|
||||
title: localization.Fairness.Metrics.balancedAccuracy,
|
||||
userVisible: true
|
||||
},
|
||||
balanced_root_mean_squared_error: {
|
||||
description: localization.Metrics.balancedRMSEDescription,
|
||||
description: localization.Fairness.Metrics.balancedRMSEDescription,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "balanced_root_mean_squared_error",
|
||||
title: localization.Metrics.balancedRootMeanSquaredError,
|
||||
title: localization.Fairness.Metrics.balancedRootMeanSquaredError,
|
||||
userVisible: true
|
||||
},
|
||||
f1_score: {
|
||||
alwaysUpperCase: true,
|
||||
description: localization.Metrics.f1ScoreDescription,
|
||||
description: localization.Fairness.Metrics.f1ScoreDescription,
|
||||
isMinimization: false,
|
||||
isPercentage: false,
|
||||
key: "f1_score",
|
||||
title: localization.Metrics.f1Score,
|
||||
title: localization.Fairness.Metrics.f1Score,
|
||||
userVisible: true
|
||||
},
|
||||
fallout_rate: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: true,
|
||||
key: "fallout_rate",
|
||||
title: localization.Metrics.falloutRate
|
||||
title: localization.Fairness.Metrics.falloutRate
|
||||
},
|
||||
false_negative_rate: {
|
||||
description: localization.Metrics.falseNegativeRateDescription,
|
||||
description: localization.Fairness.Metrics.falseNegativeRateDescription,
|
||||
isMinimization: true,
|
||||
isPercentage: true,
|
||||
key: "false_negative_rate",
|
||||
title: localization.Metrics.falseNegativeRate
|
||||
title: localization.Fairness.Metrics.falseNegativeRate
|
||||
},
|
||||
false_positive_rate: {
|
||||
description: localization.Metrics.falsePositiveRateDescription,
|
||||
description: localization.Fairness.Metrics.falsePositiveRateDescription,
|
||||
isMinimization: true,
|
||||
isPercentage: true,
|
||||
key: "false_positive_rate",
|
||||
title: localization.Metrics.falsePositiveRate
|
||||
title: localization.Fairness.Metrics.falsePositiveRate
|
||||
},
|
||||
log_loss: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "log_loss",
|
||||
title: localization.Metrics.logLoss
|
||||
title: localization.Fairness.Metrics.logLoss
|
||||
},
|
||||
max_error: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "max_error",
|
||||
title: localization.Metrics.maxError
|
||||
title: localization.Fairness.Metrics.maxError
|
||||
},
|
||||
mean_absolute_error: {
|
||||
description: localization.Metrics.meanAbsoluteErrorDescription,
|
||||
description: localization.Fairness.Metrics.meanAbsoluteErrorDescription,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "mean_absolute_error",
|
||||
title: localization.Metrics.meanAbsoluteError,
|
||||
title: localization.Fairness.Metrics.meanAbsoluteError,
|
||||
userVisible: true
|
||||
},
|
||||
mean_squared_error: {
|
||||
description: localization.Metrics.mseDescription,
|
||||
description: localization.Fairness.Metrics.mseDescription,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "mean_squared_error",
|
||||
title: localization.Metrics.meanSquaredError,
|
||||
title: localization.Fairness.Metrics.meanSquaredError,
|
||||
userVisible: true
|
||||
},
|
||||
mean_squared_log_error: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "mean_squared_log_error",
|
||||
title: localization.Metrics.meanSquaredLogError
|
||||
title: localization.Fairness.Metrics.meanSquaredLogError
|
||||
},
|
||||
median_absolute_error: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "median_absolute_error",
|
||||
title: localization.Metrics.medianAbsoluteError
|
||||
title: localization.Fairness.Metrics.medianAbsoluteError
|
||||
},
|
||||
miss_rate: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: true,
|
||||
key: "miss_rate",
|
||||
title: localization.Metrics.missRate
|
||||
title: localization.Fairness.Metrics.missRate
|
||||
},
|
||||
overprediction: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "overprediction",
|
||||
title: localization.Metrics.overprediction
|
||||
title: localization.Fairness.Metrics.overprediction
|
||||
},
|
||||
precision_score: {
|
||||
description: localization.Metrics.precisionDescription,
|
||||
description: localization.Fairness.Metrics.precisionDescription,
|
||||
isMinimization: false,
|
||||
isPercentage: true,
|
||||
key: "precision_score",
|
||||
title: localization.Metrics.precisionScore,
|
||||
title: localization.Fairness.Metrics.precisionScore,
|
||||
userVisible: true
|
||||
},
|
||||
r2_score: {
|
||||
alwaysUpperCase: true,
|
||||
description: localization.Metrics.r2Description,
|
||||
description: localization.Fairness.Metrics.r2Description,
|
||||
isMinimization: false,
|
||||
isPercentage: false,
|
||||
key: "r2_score",
|
||||
title: localization.Metrics.r2_score,
|
||||
title: localization.Fairness.Metrics.r2_score,
|
||||
userVisible: true
|
||||
},
|
||||
recall_score: {
|
||||
description: localization.Metrics.recallDescription,
|
||||
description: localization.Fairness.Metrics.recallDescription,
|
||||
isMinimization: false,
|
||||
isPercentage: true,
|
||||
key: "recall_score",
|
||||
title: localization.Metrics.recallScore,
|
||||
title: localization.Fairness.Metrics.recallScore,
|
||||
userVisible: true
|
||||
},
|
||||
root_mean_squared_error: {
|
||||
alwaysUpperCase: true,
|
||||
description: localization.Metrics.rmseDescription,
|
||||
description: localization.Fairness.Metrics.rmseDescription,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "root_mean_squared_error",
|
||||
title: localization.Metrics.rms_error,
|
||||
title: localization.Fairness.Metrics.rms_error,
|
||||
userVisible: true
|
||||
},
|
||||
selection_rate: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: false,
|
||||
isPercentage: true,
|
||||
key: "selection_rate",
|
||||
title: localization.Metrics.selectionRate
|
||||
title: localization.Fairness.Metrics.selectionRate
|
||||
},
|
||||
specificity_score: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: false,
|
||||
isPercentage: true,
|
||||
key: "specificity_score",
|
||||
title: localization.Metrics.specificityScore
|
||||
title: localization.Fairness.Metrics.specificityScore
|
||||
},
|
||||
underprediction: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: false,
|
||||
key: "underprediction",
|
||||
title: localization.Metrics.underprediction
|
||||
title: localization.Fairness.Metrics.underprediction
|
||||
},
|
||||
zero_one_loss: {
|
||||
description: localization.loremIpsum,
|
||||
description: localization.Fairness.loremIpsum,
|
||||
isMinimization: true,
|
||||
isPercentage: true,
|
||||
key: "zero_one_loss",
|
||||
title: localization.Metrics.zeroOneLoss
|
||||
title: localization.Fairness.Metrics.zeroOneLoss
|
||||
}
|
||||
};
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
ModelMetadata,
|
||||
ICategoricalRange,
|
||||
|
|
@ -18,7 +19,6 @@ import {
|
|||
PredictionTypes,
|
||||
IPreComputedData
|
||||
} from "../IFairnessProps";
|
||||
import { localization } from "../Localization/localization";
|
||||
|
||||
import { BinnedResponseBuilder } from "./BinnedResponseBuilder";
|
||||
import { IBinnedResponse } from "./IBinnedResponse";
|
||||
|
|
@ -89,7 +89,10 @@ export class WizardBuilder {
|
|||
featureNames = props.precomputedFeatureBins.map((binObject, index) => {
|
||||
return (
|
||||
binObject.featureBinName ||
|
||||
localization.formatString(localization.defaultFeatureNames, index)
|
||||
localization.formatString(
|
||||
localization.Fairness.defaultFeatureNames,
|
||||
index
|
||||
)
|
||||
);
|
||||
});
|
||||
}
|
||||
|
|
@ -97,7 +100,7 @@ export class WizardBuilder {
|
|||
props.dataSummary?.classNames ||
|
||||
this.buildIndexedNames(
|
||||
this.getClassLength(props),
|
||||
localization.defaultClassNames
|
||||
localization.Fairness.defaultClassNames
|
||||
);
|
||||
const featureRanges = props.precomputedFeatureBins.map((binMeta) => {
|
||||
return {
|
||||
|
|
@ -126,17 +129,17 @@ export class WizardBuilder {
|
|||
}
|
||||
featureNames =
|
||||
featureLength === 1
|
||||
? [localization.defaultSingleFeatureName]
|
||||
? [localization.Fairness.defaultSingleFeatureName]
|
||||
: this.buildIndexedNames(
|
||||
featureLength,
|
||||
localization.defaultFeatureNames
|
||||
localization.Fairness.defaultFeatureNames
|
||||
);
|
||||
}
|
||||
const classNames =
|
||||
props.dataSummary?.classNames ||
|
||||
this.buildIndexedNames(
|
||||
this.getClassLength(props),
|
||||
localization.defaultClassNames
|
||||
localization.Fairness.defaultClassNames
|
||||
);
|
||||
const featureIsCategorical = ModelMetadata.buildIsCategorical(
|
||||
featureNames.length,
|
||||
|
|
@ -224,7 +227,7 @@ export class WizardBuilder {
|
|||
title:
|
||||
customMetric?.name ||
|
||||
localization.formatString(
|
||||
localization.defaultCustomMetricName,
|
||||
localization.Fairness.defaultCustomMetricName,
|
||||
customMetrics.length
|
||||
)
|
||||
});
|
||||
|
|
|
|||
|
|
@ -1,11 +1,10 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { ActionButton, Stack, Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
|
||||
import { IntroTabStyles } from "./IntroTab.styles";
|
||||
|
||||
export interface IIntroTabProps {
|
||||
|
|
@ -19,29 +18,31 @@ export class IntroTab extends React.PureComponent<IIntroTabProps> {
|
|||
<Stack style={{ height: "100%" }}>
|
||||
<div className={styles.firstSection}>
|
||||
<Text className={styles.firstSectionTitle} block>
|
||||
{localization.Intro.welcome}
|
||||
{localization.Fairness.Intro.welcome}
|
||||
</Text>
|
||||
<Text className={styles.firstSectionSubtitle} block>
|
||||
{localization.Intro.fairnessDashboard}
|
||||
{localization.Fairness.Intro.fairnessDashboard}
|
||||
</Text>
|
||||
<Text variant={"large"} block>
|
||||
{localization.Intro.introBody}
|
||||
{localization.Fairness.Intro.introBody}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.lowerSection}>
|
||||
<div className={styles.stepsContainer}>
|
||||
<Text variant={"large"} className={styles.boldStep}>
|
||||
{localization.Intro.explanatoryStep}
|
||||
{localization.Fairness.Intro.explanatoryStep}
|
||||
</Text>
|
||||
<div className={styles.explanatoryStep}>
|
||||
<div>
|
||||
<Text variant={"large"} className={styles.numericLabel}>
|
||||
01
|
||||
</Text>
|
||||
<Text variant={"large"}>{localization.Intro.features}</Text>
|
||||
<Text variant={"large"}>
|
||||
{localization.Fairness.Intro.features}
|
||||
</Text>
|
||||
</div>
|
||||
<Text className={styles.explanatoryText} block>
|
||||
{localization.Intro.featuresInfo}
|
||||
{localization.Fairness.Intro.featuresInfo}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.explanatoryStep}>
|
||||
|
|
@ -49,10 +50,12 @@ export class IntroTab extends React.PureComponent<IIntroTabProps> {
|
|||
<Text variant={"large"} className={styles.numericLabel}>
|
||||
02
|
||||
</Text>
|
||||
<Text variant={"large"}>{localization.Intro.performance}</Text>
|
||||
<Text variant={"large"}>
|
||||
{localization.Fairness.Intro.performance}
|
||||
</Text>
|
||||
</div>
|
||||
<Text className={styles.explanatoryText} block>
|
||||
{localization.Intro.performanceInfo}
|
||||
{localization.Fairness.Intro.performanceInfo}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -62,7 +65,7 @@ export class IntroTab extends React.PureComponent<IIntroTabProps> {
|
|||
className={styles.getStarted}
|
||||
onClick={this.props.onNext}
|
||||
>
|
||||
{localization.Intro.getStarted}
|
||||
{localization.Fairness.Intro.getStarted}
|
||||
</ActionButton>
|
||||
</Stack>
|
||||
</div>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
AccessibleChart,
|
||||
ChartBuilder,
|
||||
|
|
@ -21,7 +22,6 @@ import {
|
|||
import React from "react";
|
||||
|
||||
import { PredictionTypes } from "../../IFairnessProps";
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { FormatMetrics } from "../../util/FormatMetrics";
|
||||
import { IFairnessContext } from "../../util/IFairnessContext";
|
||||
import { MetricsCache } from "../../util/MetricsCache";
|
||||
|
|
@ -142,7 +142,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
<Spinner
|
||||
className={styles.spinner}
|
||||
size={SpinnerSize.large}
|
||||
label={localization.calculating}
|
||||
label={localization.Fairness.calculating}
|
||||
/>
|
||||
);
|
||||
}
|
||||
|
|
@ -211,7 +211,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
) ||
|
||||
this.props.performancePickerProps.performanceOptions[0];
|
||||
const insights2 = localization.formatString(
|
||||
localization.ModelComparison.insightsText2,
|
||||
localization.Fairness.ModelComparison.insightsText2,
|
||||
selectedMetric.title,
|
||||
formattedMinPerformance,
|
||||
formattedMaxPerformance,
|
||||
|
|
@ -222,7 +222,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
? selectedMetric.title
|
||||
: selectedMetric.title.toLowerCase();
|
||||
const insights3 = localization.formatString(
|
||||
localization.ModelComparison.insightsText3,
|
||||
localization.Fairness.ModelComparison.insightsText3,
|
||||
metricTitleAppropriateCase,
|
||||
selectedMetric.isMinimization
|
||||
? formattedMinPerformance
|
||||
|
|
@ -240,7 +240,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
);
|
||||
|
||||
const insights4 = localization.formatString(
|
||||
localization.ModelComparison.insightsText4,
|
||||
localization.Fairness.ModelComparison.insightsText4,
|
||||
metricTitleAppropriateCase,
|
||||
FormatMetrics.formatNumbers(
|
||||
this.state.performanceArray[minDisparityIndex],
|
||||
|
|
@ -250,12 +250,12 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
);
|
||||
|
||||
const howToReadText = localization.formatString(
|
||||
localization.ModelComparison.howToReadText,
|
||||
localization.Fairness.ModelComparison.howToReadText,
|
||||
this.props.modelCount.toString(),
|
||||
metricTitleAppropriateCase,
|
||||
selectedMetric.isMinimization
|
||||
? localization.ModelComparison.lower
|
||||
: localization.ModelComparison.higher
|
||||
? localization.Fairness.ModelComparison.lower
|
||||
: localization.Fairness.ModelComparison.higher
|
||||
);
|
||||
|
||||
const props = _.cloneDeep(this.plotlyProps);
|
||||
|
|
@ -272,9 +272,9 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
}
|
||||
if (props.layout?.yaxis) {
|
||||
props.layout.yaxis.title = this.state.disparityInOutcomes
|
||||
? localization.ModelComparison.disparityInOutcomes
|
||||
? localization.Fairness.ModelComparison.disparityInOutcomes
|
||||
: localization.formatString(
|
||||
localization.ModelComparison.disparityInPerformance,
|
||||
localization.Fairness.ModelComparison.disparityInPerformance,
|
||||
metricTitleAppropriateCase
|
||||
);
|
||||
}
|
||||
|
|
@ -282,7 +282,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
<Stack className={styles.frame}>
|
||||
<div className={styles.header}>
|
||||
<Text variant={"large"} className={styles.headerTitle} block>
|
||||
{localization.ModelComparison.title}
|
||||
{localization.Fairness.ModelComparison.title}
|
||||
</Text>
|
||||
<ActionButton
|
||||
iconProps={{ iconName: "Edit" }}
|
||||
|
|
@ -290,7 +290,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
className={styles.editButton}
|
||||
autoFocus={true}
|
||||
>
|
||||
{localization.Report.editConfiguration}
|
||||
{localization.Fairness.Report.editConfiguration}
|
||||
</ActionButton>
|
||||
</div>
|
||||
<div className={styles.main}>
|
||||
|
|
@ -303,13 +303,13 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
</div>
|
||||
<div className={styles.mainRight}>
|
||||
<Text className={styles.rightTitle} block>
|
||||
{localization.ModelComparison.howToRead}
|
||||
{localization.Fairness.ModelComparison.howToRead}
|
||||
</Text>
|
||||
<Text className={styles.rightText} block>
|
||||
{howToReadText}
|
||||
</Text>
|
||||
<Text className={styles.insights} block>
|
||||
{localization.ModelComparison.insightsLegacy}
|
||||
{localization.Fairness.ModelComparison.insightsLegacy}
|
||||
</Text>
|
||||
<div className={styles.insightsText}>
|
||||
<Text className={styles.textSection} block>
|
||||
|
|
@ -335,18 +335,18 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
key: "performance",
|
||||
styles: { choiceFieldWrapper: styles.radioOptions },
|
||||
text: localization.formatString(
|
||||
localization.ModelComparison.disparityInPerformance,
|
||||
localization.Fairness.ModelComparison.disparityInPerformance,
|
||||
metricTitleAppropriateCase
|
||||
)
|
||||
},
|
||||
{
|
||||
key: "outcomes",
|
||||
styles: { choiceFieldWrapper: styles.radioOptions },
|
||||
text: localization.ModelComparison.disparityInOutcomes
|
||||
text: localization.Fairness.ModelComparison.disparityInOutcomes
|
||||
}
|
||||
]}
|
||||
onChange={this.disparityChanged}
|
||||
label={localization.ModelComparison.howToMeasureDisparity}
|
||||
label={localization.Fairness.ModelComparison.howToMeasureDisparity}
|
||||
required={false}
|
||||
></ChoiceGroup>
|
||||
</div>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
IconButton,
|
||||
DefaultButton,
|
||||
|
|
@ -12,7 +13,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IParityPickerPropsV1 } from "../FairnessWizard";
|
||||
|
||||
interface IState {
|
||||
|
|
@ -69,7 +69,7 @@ export class ParityPicker extends React.PureComponent<
|
|||
>
|
||||
<div>
|
||||
<DefaultButton onClick={this.onDismiss}>
|
||||
{localization.close}
|
||||
{localization.Fairness.close}
|
||||
</DefaultButton>
|
||||
</div>
|
||||
</Callout>
|
||||
|
|
|
|||
|
|
@ -1,13 +1,13 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { Stack, StackItem, Separator } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { DataSpecificationBlade } from "../../components/DataSpecificationBlade";
|
||||
import { IWizardTabProps } from "../../components/IWizardTabProps";
|
||||
import { WizardFooter } from "../../components/WizardFooter";
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IParityPickerPropsV1 } from "../FairnessWizard";
|
||||
|
||||
import { TileList, ITileProp } from "./TileList";
|
||||
|
|
@ -22,8 +22,10 @@ export class ParityTab extends React.PureComponent<IParityTabProps> {
|
|||
<Stack horizontal>
|
||||
<StackItem grow={2}>
|
||||
<Stack>
|
||||
<h2 style={{ fontWeight: "bold" }}>{localization.Parity.header}</h2>
|
||||
<p>{localization.Parity.bodyLegacy}</p>
|
||||
<h2 style={{ fontWeight: "bold" }}>
|
||||
{localization.Fairness.Parity.header}
|
||||
</h2>
|
||||
<p>{localization.Fairness.Parity.bodyLegacy}</p>
|
||||
<StackItem grow={2}>
|
||||
<TileList
|
||||
items={this.props.parityPickerProps.parityOptions.map(
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
IconButton,
|
||||
DefaultButton,
|
||||
|
|
@ -12,7 +13,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IPerformancePickerPropsV1 } from "../FairnessWizard";
|
||||
|
||||
interface IState {
|
||||
|
|
@ -68,7 +68,7 @@ export class PerformancePicker extends React.PureComponent<
|
|||
>
|
||||
<div>
|
||||
<DefaultButton onClick={this.onDismiss}>
|
||||
{localization.close}
|
||||
{localization.Fairness.close}
|
||||
</DefaultButton>
|
||||
</div>
|
||||
</Callout>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { Text, FocusZone } from "office-ui-fabric-react";
|
||||
import { Stack, StackItem } from "office-ui-fabric-react/lib/Stack";
|
||||
import React from "react";
|
||||
|
|
@ -9,7 +10,6 @@ import { DataSpecificationBlade } from "../../components/DataSpecificationBlade"
|
|||
import { IWizardTabProps } from "../../components/IWizardTabProps";
|
||||
import { WizardFooter } from "../../components/WizardFooter";
|
||||
import { PredictionTypes } from "../../IFairnessProps";
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IPerformancePickerPropsV1 } from "../FairnessWizard";
|
||||
|
||||
import { PerformanceTabStyles } from "./PerformanceTab.styles";
|
||||
|
|
@ -32,22 +32,22 @@ export class PerformanceTab extends React.PureComponent<
|
|||
>
|
||||
<Stack className={styles.main}>
|
||||
<Text className={styles.header} block>
|
||||
{localization.Performance.header}
|
||||
{localization.Fairness.Performance.header}
|
||||
</Text>
|
||||
<Text className={styles.textBody} block>
|
||||
{localization.formatString(
|
||||
localization.Performance.body,
|
||||
localization.Fairness.Performance.body,
|
||||
this.props.dashboardContext.modelMetadata.PredictionType !==
|
||||
PredictionTypes.Regression
|
||||
? localization.Performance.binary
|
||||
: localization.Performance.continuous,
|
||||
? localization.Fairness.Performance.binary
|
||||
: localization.Fairness.Performance.continuous,
|
||||
this.props.dashboardContext.modelMetadata.PredictionType ===
|
||||
PredictionTypes.BinaryClassification
|
||||
? localization.Performance.binary
|
||||
: localization.Performance.continuous,
|
||||
? localization.Fairness.Performance.binary
|
||||
: localization.Fairness.Performance.continuous,
|
||||
this.props.dashboardContext.predictions.length === 1
|
||||
? localization.Performance.modelMakes
|
||||
: localization.Performance.modelsMake
|
||||
? localization.Fairness.Performance.modelMakes
|
||||
: localization.Fairness.Performance.modelsMake
|
||||
)}
|
||||
</Text>
|
||||
<StackItem grow={2} className={styles.itemsList}>
|
||||
|
|
|
|||
|
|
@ -1,11 +1,10 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { Stack, Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
|
||||
import { SummaryTableStyles } from "./SummaryTable.styles";
|
||||
|
||||
export interface ISummaryTableProps {
|
||||
|
|
@ -55,12 +54,12 @@ export class SummaryTable extends React.PureComponent<ISummaryTableProps> {
|
|||
<Stack horizontal>
|
||||
{minIndexes.includes(index) && (
|
||||
<Text variant={"xSmall"} className={styles.minMaxLabel}>
|
||||
{localization.Report.minTag}
|
||||
{localization.Fairness.Report.minTag}
|
||||
</Text>
|
||||
)}
|
||||
{maxIndexes.includes(index) && (
|
||||
<Text variant={"xSmall"} className={styles.minMaxLabel}>
|
||||
{localization.Report.maxTag}
|
||||
{localization.Fairness.Report.maxTag}
|
||||
</Text>
|
||||
)}
|
||||
</Stack>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { RangeTypes } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -27,7 +28,6 @@ import {
|
|||
} from "../util/PerformanceMetrics";
|
||||
import { WizardBuilder } from "../util/WizardBuilder";
|
||||
|
||||
import { localization } from "./../Localization/localization";
|
||||
import { IntroTab } from "./Controls/IntroTab";
|
||||
import { ModelComparisonChart } from "./Controls/ModelComparisonChart";
|
||||
import { ParityTab } from "./Controls/ParityTab";
|
||||
|
|
@ -197,11 +197,13 @@ export class FairnessWizardV1 extends React.PureComponent<
|
|||
className={styles.thinHeader}
|
||||
>
|
||||
{/* <Text variant={"mediumPlus"} className={styles.headerLeft}>
|
||||
{localization.Header.title}
|
||||
{localization.Fairness.Header.title}
|
||||
</Text> */}
|
||||
</Stack>
|
||||
<Stack.Item grow={2} className={styles.body}>
|
||||
<Text variant={"mediumPlus"}>{localization.errorOnInputs}</Text>
|
||||
<Text variant={"mediumPlus"}>
|
||||
{localization.Fairness.errorOnInputs}
|
||||
</Text>
|
||||
</Stack.Item>
|
||||
</Stack>
|
||||
);
|
||||
|
|
@ -215,7 +217,7 @@ export class FairnessWizardV1 extends React.PureComponent<
|
|||
className={styles.thinHeader}
|
||||
>
|
||||
{/* <Text variant={"mediumPlus"} className={styles.headerLeft}>
|
||||
{localization.Header.title}
|
||||
{localization.Fairness.Header.title}
|
||||
</Text> */}
|
||||
</Stack>
|
||||
{this.state.activeTabKey === introTabKey && (
|
||||
|
|
@ -238,7 +240,7 @@ export class FairnessWizardV1 extends React.PureComponent<
|
|||
onLinkClick={this.handleTabClick}
|
||||
>
|
||||
<PivotItem
|
||||
headerText={localization.Intro.features}
|
||||
headerText={localization.Fairness.Intro.features}
|
||||
itemKey={featureBinTabKey}
|
||||
style={{ height: "100%", paddingLeft: "8px" }}
|
||||
>
|
||||
|
|
@ -252,7 +254,7 @@ export class FairnessWizardV1 extends React.PureComponent<
|
|||
/>
|
||||
</PivotItem>
|
||||
<PivotItem
|
||||
headerText={localization.performanceMetricLegacy}
|
||||
headerText={localization.Fairness.performanceMetricLegacy}
|
||||
itemKey={performanceTabKey}
|
||||
style={{ height: "100%", paddingLeft: "8px" }}
|
||||
>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { AccessibleChart, IPlotlyProperty } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -18,7 +19,6 @@ import { FormatMetrics } from "../util/FormatMetrics";
|
|||
import { ParityModes } from "../util/ParityMetrics";
|
||||
import { performanceOptions } from "../util/PerformanceMetrics";
|
||||
|
||||
import { localization } from "./../Localization/localization";
|
||||
import { IModelComparisonProps } from "./Controls/ModelComparisonChart";
|
||||
import { SummaryTable } from "./Controls/SummaryTable";
|
||||
import { WizardReportStyles } from "./WizardReport.styles";
|
||||
|
|
@ -115,7 +115,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
<Spinner
|
||||
className={styles.spinner}
|
||||
size={SpinnerSize.large}
|
||||
label={localization.calculating}
|
||||
label={localization.Fairness.calculating}
|
||||
/>
|
||||
);
|
||||
}
|
||||
|
|
@ -153,7 +153,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
{
|
||||
fillcolor: chartColors[0],
|
||||
hoverinfo: "skip",
|
||||
name: localization.Metrics.overprediction,
|
||||
name: localization.Fairness.Metrics.overprediction,
|
||||
orientation: "h",
|
||||
text: this.state.metrics.binnedOverprediction?.map((num) =>
|
||||
FormatMetrics.formatNumbers(num, "accuracy_score", false, 2)
|
||||
|
|
@ -167,7 +167,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
{
|
||||
fillcolor: chartColors[1],
|
||||
hoverinfo: "skip",
|
||||
name: localization.Metrics.underprediction,
|
||||
name: localization.Fairness.Metrics.underprediction,
|
||||
orientation: "h",
|
||||
text: this.state.metrics.binnedUnderprediction?.map((num) =>
|
||||
FormatMetrics.formatNumbers(num, "accuracy_score", false, 2)
|
||||
|
|
@ -185,7 +185,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
{
|
||||
font: { color: theme.semanticColors.bodySubtext, size: 10 },
|
||||
showarrow: false,
|
||||
text: localization.Report.underestimationError,
|
||||
text: localization.Fairness.Report.underestimationError,
|
||||
x: 0.02,
|
||||
xref: "paper",
|
||||
y: 1,
|
||||
|
|
@ -194,7 +194,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
{
|
||||
font: { color: theme.semanticColors.bodySubtext, size: 10 },
|
||||
showarrow: false,
|
||||
text: localization.Report.overestimationError,
|
||||
text: localization.Fairness.Report.overestimationError,
|
||||
x: 0.98,
|
||||
xref: "paper",
|
||||
y: 1,
|
||||
|
|
@ -231,9 +231,11 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
style={{ backgroundColor: chartColors[1] }}
|
||||
/>
|
||||
<div>
|
||||
<Text block>{localization.Report.underestimationError}</Text>
|
||||
<Text block>
|
||||
{localization.Report.underpredictionExplanation}
|
||||
{localization.Fairness.Report.underestimationError}
|
||||
</Text>
|
||||
<Text block>
|
||||
{localization.Fairness.Report.underpredictionExplanation}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -243,24 +245,28 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
style={{ backgroundColor: chartColors[0] }}
|
||||
/>
|
||||
<div>
|
||||
<Text block>{localization.Report.overestimationError}</Text>
|
||||
<Text block>{localization.Report.overpredictionExplanation}</Text>
|
||||
<Text block>
|
||||
{localization.Fairness.Report.overestimationError}
|
||||
</Text>
|
||||
<Text block>
|
||||
{localization.Fairness.Report.overpredictionExplanation}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
<Text block>
|
||||
{localization.Report.classificationPerformanceHowToRead1}
|
||||
{localization.Fairness.Report.classificationPerformanceHowToRead1}
|
||||
</Text>
|
||||
<Text block>
|
||||
{localization.Report.classificationPerformanceHowToRead2}
|
||||
{localization.Fairness.Report.classificationPerformanceHowToRead2}
|
||||
</Text>
|
||||
<Text block>
|
||||
{localization.Report.classificationPerformanceHowToRead3}
|
||||
{localization.Fairness.Report.classificationPerformanceHowToRead3}
|
||||
</Text>
|
||||
</div>
|
||||
);
|
||||
howToReadOutcomesSection = (
|
||||
<Text className={styles.textRow} block>
|
||||
{localization.Report.classificationOutcomesHowToRead}
|
||||
{localization.Fairness.Report.classificationOutcomesHowToRead}
|
||||
</Text>
|
||||
);
|
||||
}
|
||||
|
|
@ -272,7 +278,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
{
|
||||
fillcolor: chartColors[0],
|
||||
hoverinfo: "skip",
|
||||
name: localization.Metrics.overprediction,
|
||||
name: localization.Fairness.Metrics.overprediction,
|
||||
orientation: "h",
|
||||
text: this.state.metrics.binnedOverprediction?.map((num) =>
|
||||
FormatMetrics.formatNumbers(num, "overprediction", false, 2)
|
||||
|
|
@ -286,7 +292,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
{
|
||||
fillcolor: chartColors[1],
|
||||
hoverinfo: "skip",
|
||||
name: localization.Metrics.underprediction,
|
||||
name: localization.Fairness.Metrics.underprediction,
|
||||
orientation: "h",
|
||||
text: this.state.metrics.binnedUnderprediction?.map((num) =>
|
||||
FormatMetrics.formatNumbers(num, "underprediction", false, 2)
|
||||
|
|
@ -304,7 +310,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
{
|
||||
font: { color: theme.semanticColors.bodyText, size: 10 },
|
||||
showarrow: false,
|
||||
text: localization.Report.underestimationError,
|
||||
text: localization.Fairness.Report.underestimationError,
|
||||
x: 0.1,
|
||||
xref: "paper",
|
||||
y: 1,
|
||||
|
|
@ -313,7 +319,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
{
|
||||
font: { color: theme.semanticColors.bodyText, size: 10 },
|
||||
showarrow: false,
|
||||
text: localization.Report.overestimationError,
|
||||
text: localization.Fairness.Report.overestimationError,
|
||||
x: 0.9,
|
||||
xref: "paper",
|
||||
y: 1,
|
||||
|
|
@ -323,7 +329,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
}
|
||||
const opportunityText = this.state.metrics.predictions?.map((val) => {
|
||||
return localization.formatString(
|
||||
localization.Report.tooltipPrediction,
|
||||
localization.Fairness.Report.tooltipPrediction,
|
||||
FormatMetrics.formatNumbers(val, "average", false, 3)
|
||||
);
|
||||
});
|
||||
|
|
@ -350,34 +356,39 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
className={styles.colorBlock}
|
||||
style={{ backgroundColor: chartColors[0] }}
|
||||
/>
|
||||
<Text block>{localization.Report.overestimationError}</Text>
|
||||
<Text block>
|
||||
{localization.Fairness.Report.overestimationError}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.textRow}>
|
||||
<div
|
||||
className={styles.colorBlock}
|
||||
style={{ backgroundColor: chartColors[1] }}
|
||||
/>
|
||||
<Text block>{localization.Report.underestimationError}</Text>
|
||||
<Text block>
|
||||
{localization.Fairness.Report.underestimationError}
|
||||
</Text>
|
||||
</div>
|
||||
<Text className={styles.textRow} block>
|
||||
{localization.Report.probabilityPerformanceHowToRead1}
|
||||
{localization.Fairness.Report.probabilityPerformanceHowToRead1}
|
||||
</Text>
|
||||
<Text className={styles.textRow} block>
|
||||
{localization.Report.probabilityPerformanceHowToRead2}
|
||||
{localization.Fairness.Report.probabilityPerformanceHowToRead2}
|
||||
</Text>
|
||||
<Text className={styles.textRow} block>
|
||||
{localization.Report.probabilityPerformanceHowToRead3}
|
||||
{localization.Fairness.Report.probabilityPerformanceHowToRead3}
|
||||
</Text>
|
||||
</div>
|
||||
);
|
||||
howToReadOutcomesSection = (
|
||||
<div>
|
||||
<Text className={styles.textRow} block>
|
||||
{localization.Report.regressionOutcomesHowToRead}
|
||||
{localization.Fairness.Report.regressionOutcomesHowToRead}
|
||||
</Text>
|
||||
</div>
|
||||
);
|
||||
opportunityChartHeader = localization.Report.distributionOfPredictions;
|
||||
opportunityChartHeader =
|
||||
localization.Fairness.Report.distributionOfPredictions;
|
||||
}
|
||||
if (
|
||||
this.props.dashboardContext.modelMetadata.PredictionType ===
|
||||
|
|
@ -385,14 +396,14 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
) {
|
||||
const opportunityText = this.state.metrics.predictions?.map((val) => {
|
||||
return localization.formatString(
|
||||
localization.Report.tooltipPrediction,
|
||||
localization.Fairness.Report.tooltipPrediction,
|
||||
val
|
||||
);
|
||||
});
|
||||
const performanceText = this.state.metrics.predictions?.map(
|
||||
(val, index) => {
|
||||
return `${localization.formatString(
|
||||
localization.Report.tooltipError,
|
||||
localization.Fairness.Report.tooltipError,
|
||||
FormatMetrics.formatNumbers(
|
||||
this.state.metrics?.errors
|
||||
? this.state.metrics.errors[index]
|
||||
|
|
@ -402,7 +413,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
3
|
||||
)
|
||||
)}<br>${localization.formatString(
|
||||
localization.Report.tooltipPrediction,
|
||||
localization.Fairness.Report.tooltipPrediction,
|
||||
FormatMetrics.formatNumbers(val, "average", false, 3)
|
||||
)}`;
|
||||
}
|
||||
|
|
@ -442,19 +453,21 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
howToReadPerformanceSection = (
|
||||
<div>
|
||||
<Text className={styles.textRow} block>
|
||||
{localization.Report.regressionPerformanceHowToRead}
|
||||
{localization.Fairness.Report.regressionPerformanceHowToRead}
|
||||
</Text>
|
||||
</div>
|
||||
);
|
||||
howToReadOutcomesSection = (
|
||||
<div>
|
||||
<Text className={styles.textRow} block>
|
||||
{localization.Report.regressionOutcomesHowToRead}
|
||||
{localization.Fairness.Report.regressionOutcomesHowToRead}
|
||||
</Text>
|
||||
</div>
|
||||
);
|
||||
opportunityChartHeader = localization.Report.distributionOfPredictions;
|
||||
performanceChartHeader = localization.Report.distributionOfErrors;
|
||||
opportunityChartHeader =
|
||||
localization.Fairness.Report.distributionOfPredictions;
|
||||
performanceChartHeader =
|
||||
localization.Fairness.Report.distributionOfErrors;
|
||||
}
|
||||
|
||||
const globalPerformanceString = FormatMetrics.formatNumbers(
|
||||
|
|
@ -499,7 +512,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
iconProps={{ iconName: "ChevronLeft" }}
|
||||
onClick={this.clearModelSelection}
|
||||
>
|
||||
{localization.Report.backToComparisonsLegacy}
|
||||
{localization.Fairness.Report.backToComparisonsLegacy}
|
||||
</ActionButton>
|
||||
<Text variant={"large"} className={styles.modelLabel}>
|
||||
{
|
||||
|
|
@ -511,7 +524,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
</div>
|
||||
)}
|
||||
<Text variant={"mediumPlus"} className={styles.headerTitle}>
|
||||
{localization.Report.title}
|
||||
{localization.Fairness.Report.title}
|
||||
</Text>
|
||||
<div className={styles.bannerWrapper}>
|
||||
<div className={styles.headerBanner}>
|
||||
|
|
@ -520,7 +533,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
</Text>
|
||||
<Text className={styles.firstMetricLabel} block>
|
||||
{localization.formatString(
|
||||
localization.Report.globalPerformanceText,
|
||||
localization.Fairness.Report.globalPerformanceText,
|
||||
selectedMetric.alwaysUpperCase
|
||||
? selectedMetric.title
|
||||
: selectedMetric.title.toLowerCase()
|
||||
|
|
@ -531,7 +544,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
</Text>
|
||||
<Text className={styles.metricLabel} block>
|
||||
{localization.formatString(
|
||||
localization.Report.performanceDisparityText,
|
||||
localization.Fairness.Report.performanceDisparityText,
|
||||
selectedMetric.alwaysUpperCase
|
||||
? selectedMetric.title
|
||||
: selectedMetric.title.toLowerCase()
|
||||
|
|
@ -543,7 +556,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
onClick={this.onEditConfigs}
|
||||
autoFocus={true}
|
||||
>
|
||||
{localization.Report.editConfiguration}
|
||||
{localization.Fairness.Report.editConfiguration}
|
||||
</ActionButton>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -575,14 +588,14 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
</div>
|
||||
<div className={styles.mainRight}>
|
||||
<Text className={styles.rightTitle} block>
|
||||
{localization.ModelComparison.howToRead}
|
||||
{localization.Fairness.ModelComparison.howToRead}
|
||||
</Text>
|
||||
{howToReadPerformanceSection}
|
||||
</div>
|
||||
</div>
|
||||
<div className={styles.header}>
|
||||
<Text variant={"mediumPlus"} className={styles.headerTitle}>
|
||||
{localization.Report.outcomesTitle}
|
||||
{localization.Fairness.Report.outcomesTitle}
|
||||
</Text>
|
||||
<div className={styles.bannerWrapper}>
|
||||
<div className={styles.headerBanner}>
|
||||
|
|
@ -591,7 +604,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
</Text>
|
||||
<Text className={styles.firstMetricLabel} block>
|
||||
{localization.formatString(
|
||||
localization.Report.globalPerformanceText,
|
||||
localization.Fairness.Report.globalPerformanceText,
|
||||
outcomeMetric.title.toLowerCase()
|
||||
)}
|
||||
</Text>
|
||||
|
|
@ -600,7 +613,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
</Text>
|
||||
<Text className={styles.metricLabel} block>
|
||||
{localization.formatString(
|
||||
localization.Report.performanceDisparityText,
|
||||
localization.Fairness.Report.performanceDisparityText,
|
||||
outcomeMetric.title.toLowerCase()
|
||||
)}
|
||||
</Text>
|
||||
|
|
@ -635,7 +648,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
</div>
|
||||
<div className={styles.mainRight}>
|
||||
<Text className={styles.rightTitle} block>
|
||||
{localization.ModelComparison.howToRead}
|
||||
{localization.Fairness.ModelComparison.howToRead}
|
||||
</Text>
|
||||
<Text className={styles.rightText} block>
|
||||
{howToReadOutcomesSection}
|
||||
|
|
|
|||
|
|
@ -1,11 +1,10 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { PrimaryButton, Stack, Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
|
||||
import { IntroTabStyles } from "./IntroTab.styles";
|
||||
import { ReactComponent } from "./IntroTabIcon.svg";
|
||||
|
||||
|
|
@ -25,13 +24,13 @@ export class IntroTab extends React.PureComponent<IIntroTabProps> {
|
|||
className={styles.firstSectionContainer}
|
||||
>
|
||||
<Text className={styles.firstSectionTitle} block>
|
||||
{localization.Intro.welcome}
|
||||
{localization.Fairness.Intro.welcome}
|
||||
</Text>
|
||||
<Text className={styles.firstSectionSubtitle} block>
|
||||
{localization.Intro.fairnessDashboard}
|
||||
{localization.Fairness.Intro.fairnessDashboard}
|
||||
</Text>
|
||||
<Text className={styles.firstSectionBody} variant={"large"} block>
|
||||
{localization.Intro.introBody}
|
||||
{localization.Fairness.Intro.introBody}
|
||||
</Text>
|
||||
<div className={styles.firstSectionGraphics}>
|
||||
<ReactComponent />
|
||||
|
|
@ -41,17 +40,19 @@ export class IntroTab extends React.PureComponent<IIntroTabProps> {
|
|||
<div className={styles.lowerSection}>
|
||||
<div className={styles.stepsContainer}>
|
||||
<Text variant={"large"} className={styles.boldStep}>
|
||||
{localization.Intro.explanatoryStep}
|
||||
{localization.Fairness.Intro.explanatoryStep}
|
||||
</Text>
|
||||
<div className={styles.explanatoryStep}>
|
||||
<div>
|
||||
<Text variant={"large"} className={styles.numericLabel}>
|
||||
01
|
||||
</Text>
|
||||
<Text variant={"large"}>{localization.Intro.features}</Text>
|
||||
<Text variant={"large"}>
|
||||
{localization.Fairness.Intro.features}
|
||||
</Text>
|
||||
</div>
|
||||
<Text className={styles.explanatoryText} block>
|
||||
{localization.Intro.featuresInfo}
|
||||
{localization.Fairness.Intro.featuresInfo}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.explanatoryStep}>
|
||||
|
|
@ -59,10 +60,12 @@ export class IntroTab extends React.PureComponent<IIntroTabProps> {
|
|||
<Text variant={"large"} className={styles.numericLabel}>
|
||||
02
|
||||
</Text>
|
||||
<Text variant={"large"}>{localization.Intro.performance}</Text>
|
||||
<Text variant={"large"}>
|
||||
{localization.Fairness.Intro.performance}
|
||||
</Text>
|
||||
</div>
|
||||
<Text className={styles.explanatoryText} block>
|
||||
{localization.Intro.performanceInfo}
|
||||
{localization.Fairness.Intro.performanceInfo}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.explanatoryStep}>
|
||||
|
|
@ -70,10 +73,12 @@ export class IntroTab extends React.PureComponent<IIntroTabProps> {
|
|||
<Text variant={"large"} className={styles.numericLabel}>
|
||||
03
|
||||
</Text>
|
||||
<Text variant={"large"}>{localization.Intro.parity}</Text>
|
||||
<Text variant={"large"}>
|
||||
{localization.Fairness.Intro.parity}
|
||||
</Text>
|
||||
</div>
|
||||
<Text className={styles.explanatoryText} block>
|
||||
{localization.Intro.parityInfo}
|
||||
{localization.Fairness.Intro.parityInfo}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -83,7 +88,7 @@ export class IntroTab extends React.PureComponent<IIntroTabProps> {
|
|||
className={styles.getStarted}
|
||||
onClick={this.props.onNext}
|
||||
>
|
||||
{localization.Intro.getStarted}
|
||||
{localization.Fairness.Intro.getStarted}
|
||||
</PrimaryButton>
|
||||
</Stack>
|
||||
</div>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
AccessibleChart,
|
||||
ChartBuilder,
|
||||
|
|
@ -27,7 +28,6 @@ import {
|
|||
import React from "react";
|
||||
|
||||
import { PredictionTypes } from "../../IFairnessProps";
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { FormatMetrics } from "../../util/FormatMetrics";
|
||||
import { IFairnessContext } from "../../util/IFairnessContext";
|
||||
import { MetricsCache } from "../../util/MetricsCache";
|
||||
|
|
@ -195,7 +195,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
<Spinner
|
||||
className={styles.spinner}
|
||||
size={SpinnerSize.large}
|
||||
label={localization.calculating}
|
||||
label={localization.Fairness.calculating}
|
||||
/>
|
||||
);
|
||||
} else {
|
||||
|
|
@ -260,7 +260,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
);
|
||||
|
||||
const insights2 = localization.formatString(
|
||||
localization.ModelComparison.insightsText2,
|
||||
localization.Fairness.ModelComparison.insightsText2,
|
||||
selectedMetric.title,
|
||||
formattedMinPerformance,
|
||||
formattedMaxPerformance,
|
||||
|
|
@ -269,7 +269,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
);
|
||||
|
||||
const insights3 = localization.formatString(
|
||||
localization.ModelComparison.insightsText3,
|
||||
localization.Fairness.ModelComparison.insightsText3,
|
||||
selectedMetric.title.toLowerCase(),
|
||||
selectedMetric.isMinimization
|
||||
? formattedMinPerformance
|
||||
|
|
@ -285,7 +285,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
);
|
||||
|
||||
const insights4 = localization.formatString(
|
||||
localization.ModelComparison.insightsText4,
|
||||
localization.Fairness.ModelComparison.insightsText4,
|
||||
selectedMetric.title.toLowerCase(),
|
||||
FormatMetrics.formatNumbers(
|
||||
this.state.performanceArray[minDisparityIndex],
|
||||
|
|
@ -337,20 +337,20 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
/>
|
||||
</div>
|
||||
<p className={styles.modalContentIntroText}>
|
||||
{localization.ModelComparison.introModalText}
|
||||
{localization.Fairness.ModelComparison.introModalText}
|
||||
</p>
|
||||
<div style={{ display: "flex", paddingBottom: "20px" }}>
|
||||
<PrimaryButton
|
||||
className={styles.doneButton}
|
||||
onClick={this.handleCloseModalIntro}
|
||||
>
|
||||
{localization.done}
|
||||
{localization.Fairness.done}
|
||||
</PrimaryButton>
|
||||
</div>
|
||||
</Modal>
|
||||
<ActionButton onClick={this.handleOpenModalHelp}>
|
||||
<div className={styles.infoButton}>i</div>
|
||||
{localization.ModelComparison.howToRead}
|
||||
{localization.Fairness.ModelComparison.howToRead}
|
||||
</ActionButton>
|
||||
<Modal
|
||||
titleAriaId="help modal"
|
||||
|
|
@ -368,17 +368,17 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
/>
|
||||
</div>
|
||||
<p className={styles.modalContentHelpText}>
|
||||
{localization.ModelComparison.helpModalText1}
|
||||
{localization.Fairness.ModelComparison.helpModalText1}
|
||||
<br />
|
||||
<br />
|
||||
{localization.ModelComparison.helpModalText2}
|
||||
{localization.Fairness.ModelComparison.helpModalText2}
|
||||
</p>
|
||||
<div style={{ display: "flex", paddingBottom: "20px" }}>
|
||||
<PrimaryButton
|
||||
className={styles.doneButton}
|
||||
onClick={this.handleCloseModalHelp}
|
||||
>
|
||||
{localization.done}
|
||||
{localization.Fairness.done}
|
||||
</PrimaryButton>
|
||||
</div>
|
||||
</Modal>
|
||||
|
|
@ -398,7 +398,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
className={styles.insightsIcon}
|
||||
/>
|
||||
<Text className={styles.insights} block>
|
||||
{localization.ModelComparison.insights}
|
||||
{localization.Fairness.ModelComparison.insights}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.insightsText}>
|
||||
|
|
@ -415,7 +415,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
<div className={styles.downloadReport}>
|
||||
<Icon iconName="Download" className={styles.downloadIcon} />
|
||||
<Text style={{ verticalAlign: "middle" }}>
|
||||
{localization.ModelComparison.downloadReport}
|
||||
{localization.Fairness.ModelComparison.downloadReport}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -427,7 +427,7 @@ export class ModelComparisonChart extends React.PureComponent<
|
|||
<Stack className={styles.frame}>
|
||||
<div className={styles.header}>
|
||||
<Text variant={"large"} className={styles.headerTitle} block>
|
||||
{localization.ModelComparison.title} <b>assessment</b>
|
||||
{localization.Fairness.ModelComparison.title} <b>assessment</b>
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.headerOptions}>
|
||||
|
|
|
|||
|
|
@ -1,10 +1,9 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
|
||||
import { OverallTableStyles } from "./OverallTable.styles";
|
||||
|
||||
export interface IOverallTableProps {
|
||||
|
|
@ -56,7 +55,7 @@ export class OverallTable extends React.PureComponent<IOverallTableProps> {
|
|||
<div className={styles.flexCol}>
|
||||
<div className={styles.binBox}>
|
||||
<div className={styles.binTitle}>
|
||||
{localization.Report.overallLabel}
|
||||
{localization.Fairness.Report.overallLabel}
|
||||
</div>
|
||||
</div>
|
||||
{this.props.binLabels.map((label, index) => {
|
||||
|
|
@ -65,8 +64,8 @@ export class OverallTable extends React.PureComponent<IOverallTableProps> {
|
|||
<div className={styles.binBox} key={index}>
|
||||
<div className={styles.binLabel}>{label}</div>
|
||||
{/* <Stack horizontal>
|
||||
{minIndexes.includes(index) && <div className={styles.minMaxLabel}>{localization.Report.minTag}</div>}
|
||||
{maxIndexes.includes(index) && <div className={styles.minMaxLabel}>{localization.Report.maxTag}</div>}
|
||||
{minIndexes.includes(index) && <div className={styles.minMaxLabel}>{localization.Fairness.Report.minTag}</div>}
|
||||
{maxIndexes.includes(index) && <div className={styles.minMaxLabel}>{localization.Fairness.Report.maxTag}</div>}
|
||||
</Stack> */}
|
||||
</div>
|
||||
);
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
IconButton,
|
||||
DefaultButton,
|
||||
|
|
@ -12,7 +13,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IParityPickerPropsV2 } from "../FairnessWizard";
|
||||
|
||||
interface IState {
|
||||
|
|
@ -69,7 +69,7 @@ export class ParityPicker extends React.PureComponent<
|
|||
>
|
||||
<div>
|
||||
<DefaultButton onClick={this.onDismiss}>
|
||||
{localization.close}
|
||||
{localization.Fairness.close}
|
||||
</DefaultButton>
|
||||
</div>
|
||||
</Callout>
|
||||
|
|
|
|||
|
|
@ -1,13 +1,13 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { Stack, StackItem, Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { DataSpecificationBlade } from "../../components/DataSpecificationBlade";
|
||||
import { IWizardTabProps } from "../../components/IWizardTabProps";
|
||||
import { WizardFooter } from "../../components/WizardFooter";
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IParityPickerPropsV2 } from "../FairnessWizard";
|
||||
|
||||
import { ParityTabStyles } from "./ParityTab.styles";
|
||||
|
|
@ -29,10 +29,10 @@ export class ParityTab extends React.PureComponent<IParityTabProps> {
|
|||
<StackItem grow={2}>
|
||||
<Stack className={styles.main}>
|
||||
<Text className={styles.header} block>
|
||||
{localization.Performance.header}
|
||||
{localization.Fairness.Performance.header}
|
||||
</Text>
|
||||
<Text className={styles.textBody} block>
|
||||
{localization.Parity.body}
|
||||
{localization.Fairness.Parity.body}
|
||||
</Text>
|
||||
<StackItem grow={2} className={styles.itemsList}>
|
||||
<TileList
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
IconButton,
|
||||
DefaultButton,
|
||||
|
|
@ -12,7 +13,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IPerformancePickerPropsV2 } from "../FairnessWizard";
|
||||
|
||||
interface IState {
|
||||
|
|
@ -68,7 +68,7 @@ export class PerformancePicker extends React.PureComponent<
|
|||
>
|
||||
<div>
|
||||
<DefaultButton onClick={this.onDismiss}>
|
||||
{localization.close}
|
||||
{localization.Fairness.close}
|
||||
</DefaultButton>
|
||||
</div>
|
||||
</Callout>
|
||||
|
|
|
|||
|
|
@ -1,13 +1,13 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { AccessibleChart } from "@responsible-ai/mlchartlib";
|
||||
import { getTheme } from "@uifabric/styling";
|
||||
import { ITheme } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { PredictionTypes } from "../../IFairnessProps";
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { chartColors } from "../../util/chartColors";
|
||||
import { FormatMetrics } from "../../util/FormatMetrics";
|
||||
import { IFairnessContext } from "../../util/IFairnessContext";
|
||||
|
|
@ -36,7 +36,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
{
|
||||
color: chartColors[0],
|
||||
hoverinfo: "skip",
|
||||
name: localization.Metrics.falsePositiveRate,
|
||||
name: localization.Fairness.Metrics.falsePositiveRate,
|
||||
orientation: "h",
|
||||
text: this.props.metrics.falsePositiveRates?.bins.map((num) =>
|
||||
FormatMetrics.formatNumbers(num, "false_positive_rate", false, 2)
|
||||
|
|
@ -50,7 +50,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
{
|
||||
color: chartColors[1],
|
||||
hoverinfo: "skip",
|
||||
name: localization.Metrics.falseNegativeRate,
|
||||
name: localization.Fairness.Metrics.falseNegativeRate,
|
||||
orientation: "h",
|
||||
text: this.props.metrics.falseNegativeRates?.bins.map((num) =>
|
||||
FormatMetrics.formatNumbers(num, "false_negative_rate", false, 2)
|
||||
|
|
@ -71,7 +71,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
size: 10
|
||||
},
|
||||
showarrow: false,
|
||||
text: localization.Report.falseNegativeRate,
|
||||
text: localization.Fairness.Report.falseNegativeRate,
|
||||
x: 0.02,
|
||||
xref: "paper",
|
||||
y: 1,
|
||||
|
|
@ -83,7 +83,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
size: 10
|
||||
},
|
||||
showarrow: false,
|
||||
text: localization.Report.falsePositiveRate,
|
||||
text: localization.Fairness.Report.falsePositiveRate,
|
||||
x: 0.98,
|
||||
xref: "paper",
|
||||
y: 1,
|
||||
|
|
@ -103,7 +103,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
{
|
||||
color: chartColors[0],
|
||||
hoverinfo: "skip",
|
||||
name: localization.Metrics.overprediction,
|
||||
name: localization.Fairness.Metrics.overprediction,
|
||||
orientation: "h",
|
||||
text: this.props.metrics.overpredictions?.bins.map((num) =>
|
||||
FormatMetrics.formatNumbers(num, "overprediction", false, 2)
|
||||
|
|
@ -117,7 +117,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
{
|
||||
color: chartColors[1],
|
||||
hoverinfo: "skip",
|
||||
name: localization.Metrics.underprediction,
|
||||
name: localization.Fairness.Metrics.underprediction,
|
||||
orientation: "h",
|
||||
text: this.props.metrics.underpredictions?.bins.map((num) =>
|
||||
FormatMetrics.formatNumbers(num, "underprediction", false, 2)
|
||||
|
|
@ -137,7 +137,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
size: 10
|
||||
},
|
||||
showarrow: false,
|
||||
text: localization.Report.underestimationError,
|
||||
text: localization.Fairness.Report.underestimationError,
|
||||
x: 0.1,
|
||||
xref: "paper",
|
||||
y: 1,
|
||||
|
|
@ -149,7 +149,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
size: 10
|
||||
},
|
||||
showarrow: false,
|
||||
text: localization.Report.overestimationError,
|
||||
text: localization.Fairness.Report.overestimationError,
|
||||
x: 0.9,
|
||||
xref: "paper",
|
||||
y: 1,
|
||||
|
|
@ -165,7 +165,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
const performanceText = this.props.metrics.predictions?.map(
|
||||
(val, index) => {
|
||||
return `${localization.formatString(
|
||||
localization.Report.tooltipError,
|
||||
localization.Fairness.Report.tooltipError,
|
||||
FormatMetrics.formatNumbers(
|
||||
this.props.metrics?.errors?.[index],
|
||||
"average",
|
||||
|
|
@ -173,7 +173,7 @@ export class PerformancePlot extends React.PureComponent<
|
|||
3
|
||||
)
|
||||
)}<br>${localization.formatString(
|
||||
localization.Report.tooltipPrediction,
|
||||
localization.Fairness.Report.tooltipPrediction,
|
||||
FormatMetrics.formatNumbers(val, "average", false, 3)
|
||||
)}`;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { Text, FocusZone } from "office-ui-fabric-react";
|
||||
import { Stack, StackItem } from "office-ui-fabric-react/lib/Stack";
|
||||
import React from "react";
|
||||
|
|
@ -9,7 +10,6 @@ import { DataSpecificationBlade } from "../../components/DataSpecificationBlade"
|
|||
import { IWizardTabProps } from "../../components/IWizardTabProps";
|
||||
import { WizardFooter } from "../../components/WizardFooter";
|
||||
import { PredictionTypes } from "../../IFairnessProps";
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IPerformancePickerPropsV2 } from "../FairnessWizard";
|
||||
|
||||
import { PerformanceTabStyles } from "./PerformanceTab.styles";
|
||||
|
|
@ -32,22 +32,22 @@ export class PerformanceTab extends React.PureComponent<
|
|||
>
|
||||
<Stack className={styles.main}>
|
||||
<Text className={styles.header} block>
|
||||
{localization.Performance.header}
|
||||
{localization.Fairness.Performance.header}
|
||||
</Text>
|
||||
<Text className={styles.textBody} block>
|
||||
{localization.formatString(
|
||||
localization.Performance.body,
|
||||
localization.Fairness.Performance.body,
|
||||
this.props.dashboardContext.modelMetadata.PredictionType !==
|
||||
PredictionTypes.Regression
|
||||
? localization.Performance.binary
|
||||
: localization.Performance.continuous,
|
||||
? localization.Fairness.Performance.binary
|
||||
: localization.Fairness.Performance.continuous,
|
||||
this.props.dashboardContext.modelMetadata.PredictionType ===
|
||||
PredictionTypes.BinaryClassification
|
||||
? localization.Performance.binary
|
||||
: localization.Performance.continuous,
|
||||
? localization.Fairness.Performance.binary
|
||||
: localization.Fairness.Performance.continuous,
|
||||
this.props.dashboardContext.predictions.length === 1
|
||||
? localization.Performance.modelMakes
|
||||
: localization.Performance.modelsMake
|
||||
? localization.Fairness.Performance.modelMakes
|
||||
: localization.Fairness.Performance.modelsMake
|
||||
)}
|
||||
</Text>
|
||||
<StackItem grow={2} className={styles.itemsList}>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { RangeTypes } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -27,7 +28,6 @@ import {
|
|||
} from "../util/PerformanceMetrics";
|
||||
import { WizardBuilder } from "../util/WizardBuilder";
|
||||
|
||||
import { localization } from "./../Localization/localization";
|
||||
import { IntroTab } from "./Controls/IntroTab";
|
||||
import { ModelComparisonChart } from "./Controls/ModelComparisonChart";
|
||||
import { ParityTab } from "./Controls/ParityTab";
|
||||
|
|
@ -237,7 +237,7 @@ export class FairnessWizardV2 extends React.PureComponent<
|
|||
onLinkClick={this.handleTabClick}
|
||||
>
|
||||
<PivotItem
|
||||
headerText={localization.sensitiveFeatures}
|
||||
headerText={localization.Fairness.sensitiveFeatures}
|
||||
itemKey={featureBinTabKey}
|
||||
style={{ height: "100%", paddingLeft: "8px" }}
|
||||
>
|
||||
|
|
@ -251,7 +251,7 @@ export class FairnessWizardV2 extends React.PureComponent<
|
|||
/>
|
||||
</PivotItem>
|
||||
<PivotItem
|
||||
headerText={localization.performanceMetric}
|
||||
headerText={localization.Fairness.performanceMetric}
|
||||
itemKey={performanceTabKey}
|
||||
style={{ height: "100%", paddingLeft: "8px" }}
|
||||
>
|
||||
|
|
@ -267,7 +267,7 @@ export class FairnessWizardV2 extends React.PureComponent<
|
|||
</PivotItem>
|
||||
{flights.skipDisparity === false && (
|
||||
<PivotItem
|
||||
headerText={localization.disparityMetric}
|
||||
headerText={localization.Fairness.disparityMetric}
|
||||
itemKey={disparityTabKey}
|
||||
style={{ height: "100%", paddingLeft: "8px" }}
|
||||
>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { getTheme } from "@uifabric/styling";
|
||||
import {
|
||||
IDropdownStyles,
|
||||
|
|
@ -25,7 +26,6 @@ import { FormatMetrics } from "../util/FormatMetrics";
|
|||
import { ParityModes } from "../util/ParityMetrics";
|
||||
import { performanceOptions } from "../util/PerformanceMetrics";
|
||||
|
||||
import { localization } from "./../Localization/localization";
|
||||
import { BarPlotlyProps } from "./BarPlotlyProps";
|
||||
import { IModelComparisonProps } from "./Controls/ModelComparisonChart";
|
||||
import { OverallTable } from "./Controls/OverallTable";
|
||||
|
|
@ -104,7 +104,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
<Spinner
|
||||
className={styles.spinner}
|
||||
size={SpinnerSize.large}
|
||||
label={localization.calculating}
|
||||
label={localization.Fairness.calculating}
|
||||
/>
|
||||
);
|
||||
} else {
|
||||
|
|
@ -139,9 +139,9 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
// style={{ backgroundColor: ChartColors[1] }}
|
||||
// />
|
||||
// <div>
|
||||
// <Text block>{localization.Report.underestimationError}</Text>
|
||||
// <Text block>{localization.Fairness.Report.underestimationError}</Text>
|
||||
// <Text block>
|
||||
// {localization.Report.underpredictionExplanation}
|
||||
// {localization.Fairness.Report.underpredictionExplanation}
|
||||
// </Text>
|
||||
// </div>
|
||||
// </div>
|
||||
|
|
@ -151,26 +151,26 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
// style={{ backgroundColor: ChartColors[0] }}
|
||||
// />
|
||||
// <div>
|
||||
// <Text block>{localization.Report.overestimationError}</Text>
|
||||
// <Text block>{localization.Fairness.Report.overestimationError}</Text>
|
||||
// <Text block>
|
||||
// {localization.Report.overpredictionExplanation}
|
||||
// {localization.Fairness.Report.overpredictionExplanation}
|
||||
// </Text>
|
||||
// </div>
|
||||
// </div>
|
||||
// <Text block>
|
||||
// {localization.Report.classificationPerformanceHowToRead1}
|
||||
// {localization.Fairness.Report.classificationPerformanceHowToRead1}
|
||||
// </Text>
|
||||
// <Text block>
|
||||
// {localization.Report.classificationPerformanceHowToRead2}
|
||||
// {localization.Fairness.Report.classificationPerformanceHowToRead2}
|
||||
// </Text>
|
||||
// <Text block>
|
||||
// {localization.Report.classificationPerformanceHowToRead3}
|
||||
// {localization.Fairness.Report.classificationPerformanceHowToRead3}
|
||||
// </Text>
|
||||
// </div>
|
||||
// );
|
||||
// howToReadOutcomesSection = (
|
||||
// <Text className={styles.textRow} block>
|
||||
// {localization.Report.classificationOutcomesHowToRead}
|
||||
// {localization.Fairness.Report.classificationOutcomesHowToRead}
|
||||
// </Text>
|
||||
// );
|
||||
}
|
||||
|
|
@ -181,7 +181,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
// TODO: this plot doesn't exist anymore, does it?
|
||||
const opportunityText = this.state.metrics.predictions?.map((val) => {
|
||||
return localization.formatString(
|
||||
localization.Report.tooltipPrediction,
|
||||
localization.Fairness.Report.tooltipPrediction,
|
||||
FormatMetrics.formatNumbers(val, "average", false, 3)
|
||||
);
|
||||
});
|
||||
|
|
@ -208,7 +208,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
) {
|
||||
const opportunityText = this.state.metrics.predictions?.map((val) => {
|
||||
return localization.formatString(
|
||||
localization.Report.tooltipPrediction,
|
||||
localization.Fairness.Report.tooltipPrediction,
|
||||
val
|
||||
);
|
||||
});
|
||||
|
|
@ -228,7 +228,8 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
y: this.props.dashboardContext.binVector
|
||||
} as any
|
||||
];
|
||||
performanceChartHeader = localization.Report.distributionOfErrors;
|
||||
performanceChartHeader =
|
||||
localization.Fairness.Report.distributionOfErrors;
|
||||
}
|
||||
|
||||
// define task-specific metrics to show by default
|
||||
|
|
@ -382,18 +383,18 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
))}
|
||||
<Text>
|
||||
{(this.state.expandAttributes &&
|
||||
localization.Report.collapseSensitiveAttributes) ||
|
||||
localization.Fairness.Report.collapseSensitiveAttributes) ||
|
||||
(!this.state.expandAttributes &&
|
||||
localization.Report.expandSensitiveAttributes)}
|
||||
localization.Fairness.Report.expandSensitiveAttributes)}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.equalizedOdds}>
|
||||
<Text>{localization.Report.equalizedOddsDisparity}</Text>
|
||||
<Text>{localization.Fairness.Report.equalizedOddsDisparity}</Text>
|
||||
</div>
|
||||
<div className={styles.howTo}>
|
||||
<ActionButton onClick={this.handleOpenModalHelp}>
|
||||
<div className={styles.infoButton}>i</div>
|
||||
{localization.ModelComparison.howToRead}
|
||||
{localization.Fairness.ModelComparison.howToRead}
|
||||
</ActionButton>
|
||||
<Modal
|
||||
titleAriaId="intro modal"
|
||||
|
|
@ -411,13 +412,22 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
/>
|
||||
</div>
|
||||
<p className={styles.modalContentHelpText}>
|
||||
{localization.Report.classificationPerformanceHowToRead1}
|
||||
{
|
||||
localization.Fairness.Report
|
||||
.classificationPerformanceHowToRead1
|
||||
}
|
||||
<br />
|
||||
<br />
|
||||
{localization.Report.classificationPerformanceHowToRead2}
|
||||
{
|
||||
localization.Fairness.Report
|
||||
.classificationPerformanceHowToRead2
|
||||
}
|
||||
<br />
|
||||
<br />
|
||||
{localization.Report.classificationPerformanceHowToRead3}
|
||||
{
|
||||
localization.Fairness.Report
|
||||
.classificationPerformanceHowToRead3
|
||||
}
|
||||
<br />
|
||||
<br />
|
||||
</p>
|
||||
|
|
@ -426,7 +436,7 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
className={styles.doneButton}
|
||||
onClick={this.handleCloseModalHelp}
|
||||
>
|
||||
{localization.done}
|
||||
{localization.Fairness.done}
|
||||
</PrimaryButton>
|
||||
</div>
|
||||
</Modal>
|
||||
|
|
@ -468,10 +478,10 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
/>
|
||||
<div>
|
||||
<div className={styles.legendTitle}>
|
||||
{localization.Report.underestimationError}
|
||||
{localization.Fairness.Report.underestimationError}
|
||||
</div>
|
||||
<div className={styles.legendSubtitle}>
|
||||
{localization.Report.underpredictionExplanation}
|
||||
{localization.Fairness.Report.underpredictionExplanation}
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -482,10 +492,10 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
/>
|
||||
<div>
|
||||
<div className={styles.legendTitle}>
|
||||
{localization.Report.overestimationError}
|
||||
{localization.Fairness.Report.overestimationError}
|
||||
</div>
|
||||
<div className={styles.legendSubtitle}>
|
||||
{localization.Report.overpredictionExplanation}
|
||||
{localization.Fairness.Report.overpredictionExplanation}
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -498,14 +508,16 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
className={styles.insightsIcon}
|
||||
/>
|
||||
<Text style={{ verticalAlign: "middle" }}>
|
||||
{localization.ModelComparison.insights}
|
||||
{localization.Fairness.ModelComparison.insights}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={styles.insightsText}>{localization.loremIpsum}</div>
|
||||
<div className={styles.insightsText}>
|
||||
{localization.Fairness.loremIpsum}
|
||||
</div>
|
||||
<div className={styles.downloadReport}>
|
||||
<Icon iconName="Download" className={styles.downloadIcon} />
|
||||
<Text style={{ verticalAlign: "middle" }}>
|
||||
{localization.ModelComparison.downloadReport}
|
||||
{localization.Fairness.ModelComparison.downloadReport}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -523,12 +535,12 @@ export class WizardReport extends React.PureComponent<IReportProps, IState> {
|
|||
iconProps={{ iconName: "ChevronLeft" }}
|
||||
onClick={this.clearModelSelection}
|
||||
>
|
||||
{localization.Report.backToComparisons}
|
||||
{localization.Fairness.Report.backToComparisons}
|
||||
</ActionButton>
|
||||
</div>
|
||||
)}
|
||||
<div className={styles.modelLabel}>
|
||||
{localization.Report.assessmentResults}{" "}
|
||||
{localization.Fairness.Report.assessmentResults}{" "}
|
||||
<b>
|
||||
{
|
||||
this.props.dashboardContext.modelNames[
|
||||
|
|
|
|||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Pokud si chcete zobrazit místní vysvětlení určitého bodu, vyberte daný bod.",
|
||||
"defaultClassNames": "Třída {0}",
|
||||
"defaultFeatureNames": "Příznak {0}",
|
||||
"absoluteAverage": "Průměr absolutní hodnoty",
|
||||
"predictedClass": "Predikovaná třída",
|
||||
"datasetExplorer": "Průzkumník datových sad",
|
||||
"dataExploration": "Průzkum datové sady",
|
||||
"aggregateFeatureImportance": "Důležitost agregovaného příznaku",
|
||||
"globalImportance": "Globální důležitost",
|
||||
"explanationExploration": "Průzkum vysvětlení",
|
||||
"individualAndWhatIf": "Důležitost jednotlivých příznaků a citlivostní analýza",
|
||||
"summaryImportance": "Důležitost souhrnu",
|
||||
"featureImportance": "Významnost atributu",
|
||||
"featureImportanceOf": "Důležitost příznaku {0}",
|
||||
"perturbationExploration": "Průzkum perturbace",
|
||||
"localFeatureImportance": "Významnost místního atributu",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Vymazat výběr",
|
||||
"feature": "Funkce:",
|
||||
"intercept": "Zachycení",
|
||||
"modelPerformance": "Výkon modelu",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Data: {0}",
|
||||
"importanceLabel": "Důležitost: {0}",
|
||||
"predictedY": "Predikované Y",
|
||||
"index": "Index",
|
||||
"dataGroupLabel": "Data",
|
||||
"output": "Výstup",
|
||||
"probabilityLabel": "Pravděpodobnost: {0}",
|
||||
"trueY": "Skutečné Y",
|
||||
"class": "třída: ",
|
||||
"xValue": "Hodnota X:",
|
||||
"yValue": "Hodnota Y:",
|
||||
"colorValue": "Barva:",
|
||||
"count": "Počet"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Vážení mezi třídami:",
|
||||
"info": "Informace o výpočtu mezi třídami",
|
||||
"overviewInfo": "Modely s více třídami generují pro každou třídu vektor významností nezávislých atributů. Vektor významností atributů jednotlivých tříd ukazuje, kvůli kterým příznakům je určitá třída pravděpodobnější nebo méně pravděpodobná. Můžete vybrat, jak se váhy vektorů významností atributů pro jednotlivé třídy shrnují do jedné hodnoty:",
|
||||
"absoluteValInfo": "Průměr absolutní hodnoty: Udává součet důležitosti příznaku ve všech možných třídách vydělený počtem tříd.",
|
||||
"predictedClassInfo": "Predikovaná třída: Udává hodnotu významnosti atributu pro predikovanou třídu daného bodu.",
|
||||
"enumeratedClassInfo": "Výčet názvů tříd: Udává jen významnost atributu zadané třídy ve všech datových bodech.",
|
||||
"close": "Zavřít",
|
||||
"crossClassWeights": "Váhy mezi třídami"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Škálovaná hodnota příznaku",
|
||||
"low": "Nízká",
|
||||
"high": "Vysoká",
|
||||
"featureLabel": "Příznak: {0}",
|
||||
"valueLabel": "Hodnota příznaku: {0}",
|
||||
"importanceLabel": "Důležitost: {0}",
|
||||
"predictedClassTooltip": "Predikovaná třída: {0}",
|
||||
"trueClassTooltip": "Skutečná třída: {0}",
|
||||
"predictedOutputTooltip": "Predikovaný výstup: {0}",
|
||||
"trueOutputTooltip": "Skutečný výstup: {0}",
|
||||
"topKFeatures": "Hlavních K příznaků:",
|
||||
"topKInfo": "Jak se počítá prvních k",
|
||||
"predictedValue": "Předpokládaná hodnota",
|
||||
"predictedClass": "Predikovaná třída",
|
||||
"trueValue": "Pravdivá hodnota",
|
||||
"trueClass": "Skutečná třída",
|
||||
"noColor": "Žádné",
|
||||
"tooManyRows": "Poskytnutá datová sada je větší, než jakou tento graf podporuje."
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Třída: {0}",
|
||||
"sortBy": "Řadit podle",
|
||||
"noData": "Žádná data",
|
||||
"absoluteGlobal": "Absolutní globální",
|
||||
"absoluteLocal": "Absolutní lokální",
|
||||
"calculatingExplanation": "Počítá se vysvětlení."
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Musí to být číslo.",
|
||||
"integerError": "Musí to být celé číslo.",
|
||||
"prediction": "Predikce",
|
||||
"predictedProbability": "Predikovaná pravděpodobnost",
|
||||
"predictionLabel": "Predikce: {0}",
|
||||
"probabilityLabel": "Pravděpodobnost: {0}",
|
||||
"noModelError": "Pokud chcete prozkoumat predikce v grafech ICE, poskytněte prosím zprovozněný model.",
|
||||
"featurePickerLabel": "Funkce:",
|
||||
"minimumInputLabel": "Minimum:",
|
||||
"maximumInputLabel": "Maximum:",
|
||||
"stepInputLabel": "Kroky:",
|
||||
"loadingMessage": "Načítají se data...",
|
||||
"submitPrompt": "Pokud si chcete zobrazit graf ICE, odešlete rozsah.",
|
||||
"topLevelErrorMessage": "Chyba v parametru",
|
||||
"errorPrefix": "Došlo k chybě: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Načítání...",
|
||||
"perturbationLabel": "Perturbace:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Predikovaná hodnota: {0}",
|
||||
"predictedClassLabel": "Predikovaná třída: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Bez seskupení",
|
||||
"groupPredicted": "Predikované Y",
|
||||
"groupTrue": "Skutečné Y",
|
||||
"groupBy": "Seskupit podle"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Typ grafu:",
|
||||
"violinText": "Houslový",
|
||||
"barText": "Pruhový",
|
||||
"boxText": "Pole",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "Globální významnost atributu se počítá jako průměr absolutní hodnoty významnosti atributu ve všech bodech (normalizace L1). ",
|
||||
"multiclassImportanceAddendum": "Do výpočtu důležitosti příznaku pro všechny třídy se zahrnují všechny body, nepoužívá se žádné rozdílové vážení. Proto příznak, který má velkou zápornou důležitost pro mnoho bodů, pro které se predikovalo, že nejsou třída A, výrazně zvýší důležitost třídy A daného příznaku."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Rovno",
|
||||
"greaterThanComparison": "Větší než",
|
||||
"greaterThanEqualToComparison": "Je větší nebo rovno",
|
||||
"lessThanComparison": "Menší než",
|
||||
"lessThanEqualToComparison": "Je menší nebo rovno",
|
||||
"inTheRangeOf": "V rozsahu",
|
||||
"categoricalIncludeValues": "Zahrnuté hodnoty:",
|
||||
"numericValue": "Hodnota",
|
||||
"numericalComparison": "Operace",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Maximum",
|
||||
"min": "Minimum: {0}",
|
||||
"max": "Maximum: {0}",
|
||||
"uniqueValues": "Počet jedinečných hodnot: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Chyba regrese",
|
||||
"error": "Chyba",
|
||||
"classificationOutcome": "Výsledek klasifikace",
|
||||
"truePositive": "Pravdivě pozitivní",
|
||||
"trueNegative": "Pravdivě negativní",
|
||||
"falsePositive": "Falešně pozitivní",
|
||||
"falseNegative": "Falešně negativní",
|
||||
"dataset": "Datová sada",
|
||||
"predictedProbabilities": "Pravděpodobnosti predikce",
|
||||
"none": "Počet"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Zavřít",
|
||||
"defaultCustomRootName": "Kopie řádku {0}",
|
||||
"filterFeaturePlaceholder": "Vyhledat příznaky"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Kohorta",
|
||||
"defaultLabel": "Všechna data"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Prozkoumejte hlavních k důležitých příznaků, které mají vliv na celkové predikce modelu (tzv. globální vysvětlení). Pomocí posuvníku můžete zobrazit hodnoty důležitosti příznaků v sestupném pořadí. Pokud si chcete hodnoty důležitosti příznaků zobrazit vedle sebe, můžete vybrat až tři kohorty. Klikněte na kterýkoli z pruhů příznaků v grafu a zobrazí se míra vlivu vybraného příznaku na predikci modelu.",
|
||||
"topAtoB": "Hlavních {0}–{1} příznaků",
|
||||
"datasetCohorts": "Kohorty datové sady",
|
||||
"legendHelpText": "Kliknutím na položky legendy v grafu můžete zapínat nebo vypínat kohorty.",
|
||||
"sortBy": "Řadit podle",
|
||||
"viewDependencePlotFor": "Zobrazit graf závislostí pro:",
|
||||
"datasetCohortSelector": "Vyberte kohortu datové sady.",
|
||||
"aggregateFeatureImportance": "Důležitost agregovaného příznaku",
|
||||
"missingParameters": "Tato karta vyžaduje, aby se zadal parametr významnosti místního příznaku.",
|
||||
"weightOptions": "Váhy důležitosti tříd",
|
||||
"dependencePlotTitle": "Grafy závislostí",
|
||||
"dependencePlotHelperText": "Tento graf závislostí zobrazuje vztah mezi hodnotou příznaku a odpovídající důležitostí příznaku v celé kohortě.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Vyberte příznak.",
|
||||
"datasetRequired": "Grafy závislostí vyžadují vyhodnocovací datovou sadu a místní pole důležitostí příznaků."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Statistiky dat",
|
||||
"datapoints": "Počet datových bodů: {0}",
|
||||
"features": "Počet příznaků: {0}",
|
||||
"filters": "Počet filtrů: {0}",
|
||||
"binaryClassifier": "Binární klasifikátor",
|
||||
"regressor": "Regresor",
|
||||
"multiclassClassifier": "Klasifikátor pro více tříd",
|
||||
"datasetCohorts": "Kohorty datové sady",
|
||||
"editCohort": "Upravit kohortu",
|
||||
"duplicateCohort": "Duplikovat kohortu",
|
||||
"addCohort": "Přidat kohortu",
|
||||
"copy": " kopírovat"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Vyhodnoťte výkon modelu tím, že prozkoumáte distribuci hodnot predikce a hodnot metrik výkonu modelu. Podrobněji můžete model prozkoumat tak, že se podíváte na srovnávací analýzu výkonu mezi různými kohortami nebo podskupinami v datové sadě. Pro informace v různých dimenzích použijte filtry pro hodnoty x a y. Ozubené kolo v grafu umožňuje změnit typ grafu.",
|
||||
"modelStatistics": "Statistika modelu",
|
||||
"cohortPickerLabel": "Vyberte kohortu databáze, kterou chcete prozkoumat.",
|
||||
"missingParameters": "Tato karta vyžaduje, aby se zadalo pole predikovaných hodnot z modelu.",
|
||||
"missingTrueY": "Statistika výkonu modelu vyžaduje, aby se kromě predikovaných výsledků poskytly i skutečné výsledky."
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Hodnota Y",
|
||||
"numberOfDatapoints": "Počet datových bodů",
|
||||
"xValue": "Hodnota X",
|
||||
"rowIndex": "Index řádku",
|
||||
"featureImportance": "Důležitost příznaku",
|
||||
"countTooltipPrefix": "Počet: {0}",
|
||||
"count": "Počet",
|
||||
"featurePrefix": "Příznak",
|
||||
"importancePrefix": "Důležitost",
|
||||
"cohort": "Kohorta",
|
||||
"howToRead": "Jak přečíst tento graf"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Prozkoumejte statistiky datové sady tím, že vyberete různé filtry pro osy X, Y a barvy a rozdělíte data podle různých dimenzí. Vytvořte výše uvedené kohorty datové sady a analyzujte její statistiky pomocí filtrů, jako jsou predikované výsledky, příznaky datové sady a skupiny chyb. Pomocí ikony ozubeného kola v pravém horním rohu grafu můžete změnit typy grafů.",
|
||||
"colorValue": "Hodnota barvy",
|
||||
"individualDatapoints": "Jednotlivé datové body",
|
||||
"aggregatePlots": "Agregované grafy",
|
||||
"chartType": "Typ grafu",
|
||||
"missingParameters": "Tato karta vyžaduje, aby se zadala datová sada pro vyhodnocení.",
|
||||
"noColor": "Žádný"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Důležitost příznaku",
|
||||
"placeholder": "Pokud si chcete zobrazit graf závislostí příznaku, klikněte na příznak v pruhovém grafu výše."
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Když kliknete na bodový diagram, můžete vybrat datový bod, aby se zobrazily jeho místní hodnoty důležitosti příznaků (místní vysvětlení) a graf jednotlivých podmíněných očekávání (ICE) níže. Pomocí panelu napravo můžete perturbovat příznaky známého datového bodu a vytvořit hypotetický datový bod citlivostní analýzy. Hodnoty důležitosti příznaků se zakládají na mnoha odhadech, nepředstavují odůvodnění predikcí. Bez striktní matematické robustnosti běžného vyvozování nedoporučujeme uživatelům zakládat na tomto nástroji skutečná rozhodnutí.",
|
||||
"panelPlaceholder": "Aby bylo možné predikovat nové datové body, vyžaduje se model.",
|
||||
"cohortPickerLabel": "Vyberte kohortu databáze, kterou chcete prozkoumat.",
|
||||
"scatterLegendText": "Kliknutím na položky legendy v grafu můžete zapínat nebo vypínat datové body.",
|
||||
"realPoint": "Skutečné datové body",
|
||||
"noneSelectedYet": "Zatím se žádné nevybraly.",
|
||||
"whatIfDatapoints": "Datové body citlivostní analýzy",
|
||||
"noneCreatedYet": "Zatím se žádné nevytvořily.",
|
||||
"showLabel": "Zobrazit:",
|
||||
"featureImportancePlot": "Graf důležitosti příznaku",
|
||||
"icePlot": "Graf jednotlivých podmíněných očekávání (ICE)",
|
||||
"featureImportanceLackingParameters": "Pokud si chcete zobrazit, jak každý příznak ovlivňuje jednotlivé predikce, poskytněte významnosti místních příznaků.",
|
||||
"featureImportanceGetStartedText": "Pokud si chcete zobrazit důležitost příznaku, vyberte nějaký bod.",
|
||||
"iceLackingParameters": "Aby bylo možné provádět predikce hypotetických datových bodů, grafy ICE vyžadují zprovozněný model.",
|
||||
"IceGetStartedText": "Pokud si chcete zobrazit grafy ICE, vyberte bod nebo vytvořte bod citlivostní analýzy",
|
||||
"whatIfDatapoint": "Datový bod citlivostní analýzy",
|
||||
"whatIfHelpText": "Vyberte bod na grafu nebo ručně zadejte známý index datového bodu, který se má perturbovat, a uložte ho jako nový bod citlivostní analýzy.",
|
||||
"indexLabel": "Datový index pro perturbaci",
|
||||
"rowLabel": "Řádek {0}",
|
||||
"whatIfNameLabel": "Název datového bodu citlivostní analýzy",
|
||||
"featureValues": "Hodnoty příznaku",
|
||||
"predictedClass": "Predikovaná třída: ",
|
||||
"predictedValue": "Předpokládaná hodnota: ",
|
||||
"probability": "Pravděpodobnost: ",
|
||||
"trueClass": "Skutečná třída: ",
|
||||
"trueValue": "Skutečná hodnota: ",
|
||||
"trueValue.comment": "Předpona pro skutečný popisek pro regresi",
|
||||
"newPredictedClass": "Nová predikovaná třída: ",
|
||||
"newPredictedValue": "Nová predikovaná hodnota: ",
|
||||
"newProbability": "Nová pravděpodobnost: ",
|
||||
"saveAsNewPoint": "Uložit jako nový bod",
|
||||
"saveChanges": "Uložit změny",
|
||||
"loading": "Načítání...",
|
||||
"classLabel": "Třída: {0}",
|
||||
"minLabel": "Min.",
|
||||
"maxLabel": "Max.",
|
||||
"stepsLabel": "Kroky",
|
||||
"disclaimer": "Upozornění: Tato vysvětlení se zakládají na mnoha odhadech, nepředstavují odůvodnění predikcí. Bez striktní matematické robustnosti běžného vyvozování nedoporučujeme uživatelům zakládat na tomto nástroji skutečná rozhodnutí.",
|
||||
"missingParameters": "Tato karta vyžaduje, aby se zadala datová sada pro vyhodnocení.",
|
||||
"selectionLimit": "Maximálně 3 vybrané body",
|
||||
"classPickerLabel": "Třída",
|
||||
"tooltipTitleMany": "Hlavních {0} predikovaných tříd",
|
||||
"whatIfTooltipTitle": "Predikované třídy citlivostní analýzy",
|
||||
"tooltipTitleFew": "Predikované třídy",
|
||||
"probabilityLabel": "Pravděpodobnost",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "Hodnota by měla být číselná.",
|
||||
"icePlotHelperText": "Grafy ICE ukazují, jak se hodnoty predikcí vybraného datového bodu mění v rozsahu mezi maximální a minimální hodnotou příznaku."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Vybrat filtr",
|
||||
"TreatAsCategorical": "Považovat za kategorické",
|
||||
"addFilter": "Přidat filtr",
|
||||
"addedFilters": "Přidané filtry",
|
||||
"noAddedFilters": "Zatím se nepřidaly žádné filtry.",
|
||||
"defaultFilterState": "Pokud chcete do kohorty datové sady přidat parametry, vyberte filtr.",
|
||||
"cohortNameLabel": "Název kohorty datové sady",
|
||||
"cohortNamePlaceholder": "Pojmenujte svou kohortu.",
|
||||
"save": "Uložit",
|
||||
"delete": "Odstranit",
|
||||
"cancel": "Zrušit",
|
||||
"cohortNameError": "Chybí název kohorty",
|
||||
"placeholderName": "Kohorta {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Vybrat",
|
||||
"ditherLabel": "Použít dithering",
|
||||
"selectFilter": "Vyberte hodnotu osy.",
|
||||
"selectFeature": "Vyberte funkci.",
|
||||
"binLabel": "Použít pro data binning",
|
||||
"TreatAsCategorical": "Považovat za kategorické",
|
||||
"numOfBins": "Počet intervalů",
|
||||
"groupByCohort": "Seskupit podle kohorty",
|
||||
"selectClass": "Vybrat třídu",
|
||||
"countHelperText": "Histogram počtu bodů"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Predikovaná pravděpodobnost",
|
||||
"predictedY": "Predikované Y",
|
||||
"evalData": "Datová sada pro vyhodnocení",
|
||||
"localFeatureImportance": "Významnost místního příznaku",
|
||||
"inconsistentDimensions": "Nekonzistentní dimenze. {0} má dimenze {1}, očekávalo se {2}.",
|
||||
"notNonEmpty": "Vstup {0} není neprázdné pole.",
|
||||
"varyingLength": "Nekonzistentní dimenze. {0} má prvky různé délky.",
|
||||
"notArray": "{0} není pole. Očekávalo se pole dimenze {1}.",
|
||||
"errorHeader": "Některé vstupní parametry nebyly konzistentní a nepoužijí se: ",
|
||||
"datasizeWarning": "Datová sada pro vyhodnocení je příliš velká na to, aby se dala efektivně zobrazit v některých grafech. Přidejte prosím filtry, které kohortu zmenší. ",
|
||||
"datasizeError": "Vybraná kohorta je příliš velká, přidejte prosím filtry, které ji zmenší.",
|
||||
"addFilters": "Přidat filtry"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " zahrnuje {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} a další (celkem {1})"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "Spolehlivost R: {0}",
|
||||
"meanPrediction": "Střední predikce: {0}",
|
||||
"accuracy": "Úspěšnost: {0}",
|
||||
"precision": "Přesnost: {0}",
|
||||
"recall": "Úplnost: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Prozkoumejte hlavních k důležitých příznaků, které mají vliv na celkové predikce modelu. Pomocí posuvníku můžete zobrazit důležitosti příznaků v sestupném pořadí."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Co tato vysvětlení znamenají?",
|
||||
"clickHere": "Další informace",
|
||||
"shapTitle": "Hodnoty Shapley",
|
||||
"shapDescription": "Tento nástroj pro vysvětlení používá SHAP. To je přístup pro vysvětlování modelů založený na teorii her, u kterého se důležitost sad příznaků měří skrýváním těchto příznaků před modelem prostřednictvím marginalizace. Pro další informace klikněte na odkaz uvedený níže.",
|
||||
"limeTitle": "LIME (Local Interpretable Model-Agnostic Explanations)",
|
||||
"limeDescription": "Tento nástroj pro vysvětlení používá vysvětlení LIME, které nabízí lineární aproximaci modelu. Abychom získali vysvětlení, provádíme tyto kroky: perturbace instance, získání predikcí modelu a využití těchto predikcí jako popisky, abychom naučili zhuštěný, místně věrohodný lineární model. Váhy tohoto lineárního modelu se používají jako důležitosti příznaků. Pro další informace klikněte na odkaz uvedený níže.",
|
||||
"mimicTitle": "Napodobenina (globální náhradní vysvětlení)",
|
||||
"mimicDescription": "Tento nástroj pro vysvětlení se zakládá na myšlence trénování globálních náhradních modelů, které napodobují modely typu černá skříňka. Globální náhradní model je vnitřně interpretovatelný model, který je natrénovaný tak, aby co nejpřesněji aproximoval predikce jakéhokoli modelu typu černá skříňka. Hodnoty důležitosti příznaků se zakládají na modelu a jsou platné pro základní náhradní model (LightGBM, nebo lineární regrese, nebo metoda Stochastic Gradient Descent, nebo rozhodovací strom).",
|
||||
"pfiTitle": "Důležitost příznaků permutace (PFI)",
|
||||
"pfiDescription": "Tento nástroj pro vysvětlení náhodně přeuspořádá data pro celou datovou sadu po jednotlivých příznacích a vypočítá, jak moc se mění zkoumaná metrika výkonu (výchozí metriky výkonu: F1 pro binární klasifikaci, skóre F1 s mikroprůměrem pro klasifikaci s více třídami a střední absolutní chybu pro regresi). Čím větší bude změna, tím důležitější je příznak. Tento nástroj pro vysvětlení dokáže vysvětlit jen celkové chování základního modelu, nevysvětluje jednotlivé predikce. Hodnota důležitosti funkce příznaku představuje rozdíl ve výkonu modelu perturbací určitého příznaku."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Wählen Sie einen Punkt aus, um seine lokale Erklärung anzuzeigen.",
|
||||
"defaultClassNames": "Klasse \"{0}\"",
|
||||
"defaultFeatureNames": "Feature \"{0}\"",
|
||||
"absoluteAverage": "Durchschnitt des absoluten Werts",
|
||||
"predictedClass": "Vorhergesagte Klasse",
|
||||
"datasetExplorer": "Dataset-Explorer",
|
||||
"dataExploration": "Datasetuntersuchung",
|
||||
"aggregateFeatureImportance": "Aggregierte Featurerelevanz",
|
||||
"globalImportance": "Globale Relevanz",
|
||||
"explanationExploration": "Durchsuchen der Erklärung",
|
||||
"individualAndWhatIf": "Individuelle Featurerelevanz und Was-wäre-wenn-Analyse",
|
||||
"summaryImportance": "Zusammenfassung – Relevanz",
|
||||
"featureImportance": "Featurerelevanz",
|
||||
"featureImportanceOf": "Featurerelevanz von \"{0}\"",
|
||||
"perturbationExploration": "Durchsuchen des Störeinflusses",
|
||||
"localFeatureImportance": "Lokale Featurerelevanz",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Auswahl löschen",
|
||||
"feature": "Feature:",
|
||||
"intercept": "Abfangen",
|
||||
"modelPerformance": "Modellleistung",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Daten : {0}",
|
||||
"importanceLabel": "Relevanz : {0}",
|
||||
"predictedY": "Vorhersage für Y",
|
||||
"index": "Index",
|
||||
"dataGroupLabel": "Daten",
|
||||
"output": "Ausgabe",
|
||||
"probabilityLabel": "Wahrscheinlichkeit : {0}",
|
||||
"trueY": "TRUE Y",
|
||||
"class": "Klasse: ",
|
||||
"xValue": "X-Wert:",
|
||||
"yValue": "Y-Wert:",
|
||||
"colorValue": "Farbe:",
|
||||
"count": "Anzahl"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Klassenübergreifende Gewichtung:",
|
||||
"info": "Informationen zur klassenübergreifenden Berechnung",
|
||||
"overviewInfo": "Multiklassenmodelle generieren für jede Klasse einen unabhängigen Featurerelevanzvektor. Der Featurerelevanzvektor der einzelnen Klassen zeigt, welche Features eine Klasse wahrscheinlicher oder unwahrscheinlicher werden ließen. Sie können auswählen, wie die Gewichtungen der Featurerelevanzvektoren pro Klasse zu einem einzigen Wert zusammengefasst werden:",
|
||||
"absoluteValInfo": "Durchschnitt des absoluten Werts: Zeigt die Summe der Relevanz des Features für alle möglichen Klassen, geteilt durch die Klassenanzahl.",
|
||||
"predictedClassInfo": "Vorhergesagte Klasse: Zeigt den Wert der Featurerelevanz für die vorhergesagte Klasse eines bestimmten Punkts an.",
|
||||
"enumeratedClassInfo": "Aufgelistete Klassennamen: Zeigt nur die Featurerelevanz der angegebenen Klasse für alle Datenpunkte an.",
|
||||
"close": "Schließen",
|
||||
"crossClassWeights": "Klassenübergreifende Gewichtungen"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Featurewert auf Farbskala",
|
||||
"low": "Niedrig",
|
||||
"high": "Hoch",
|
||||
"featureLabel": "Feature: {0}",
|
||||
"valueLabel": "Featurewert: {0}",
|
||||
"importanceLabel": "Relevanz: {0}",
|
||||
"predictedClassTooltip": "Vorhergesagte Klasse: {0}",
|
||||
"trueClassTooltip": "TRUE-Klasse: {0}",
|
||||
"predictedOutputTooltip": "Vorhergesagte Ausgabe: {0}",
|
||||
"trueOutputTooltip": "TRUE-Ausgabe: {0}",
|
||||
"topKFeatures": "Wichtigste K Features:",
|
||||
"topKInfo": "Gibt an, wie die Top-K-Berechnung erfolgt.",
|
||||
"predictedValue": "Vorhergesagter Wert",
|
||||
"predictedClass": "Vorhergesagte Klasse",
|
||||
"trueValue": "True-Wert",
|
||||
"trueClass": "TRUE-Klasse",
|
||||
"noColor": "Keine",
|
||||
"tooManyRows": "Das angegebene Dataset ist größer als von diesem Diagramm unterstützt wird."
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Klasse: {0}",
|
||||
"sortBy": "Sortieren nach",
|
||||
"noData": "Keine Daten",
|
||||
"absoluteGlobal": "Absolut global",
|
||||
"absoluteLocal": "Absolut lokal",
|
||||
"calculatingExplanation": "Erklärung wird berechnet."
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Muss numerisch sein.",
|
||||
"integerError": "Muss eine ganze Zahl sein.",
|
||||
"prediction": "Vorhersage",
|
||||
"predictedProbability": "Vorhergesagte Wahrscheinlichkeit",
|
||||
"predictionLabel": "Vorhersage: {0}",
|
||||
"probabilityLabel": "Wahrscheinlichkeit: {0}",
|
||||
"noModelError": "Geben Sie ein operationalisiertes Modell an, um Vorhersagen in ICE-Plots zu untersuchen.",
|
||||
"featurePickerLabel": "Feature:",
|
||||
"minimumInputLabel": "Minimum:",
|
||||
"maximumInputLabel": "Maximum:",
|
||||
"stepInputLabel": "Schritte:",
|
||||
"loadingMessage": "Daten werden geladen...",
|
||||
"submitPrompt": "Übermitteln Sie einen Bereich, um einen ICE-Plot anzuzeigen.",
|
||||
"topLevelErrorMessage": "Fehler in Parameter.",
|
||||
"errorPrefix": "Fehler erkannt: {0} "
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Wird geladen...",
|
||||
"perturbationLabel": "Störeinfluss:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Vorhergesagter Wert : {0}",
|
||||
"predictedClassLabel": "Vorhergesagte Klasse : {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Keine Gruppierung",
|
||||
"groupPredicted": "Vorhersage für Y",
|
||||
"groupTrue": "TRUE Y",
|
||||
"groupBy": "Gruppieren nach"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Diagrammtyp:",
|
||||
"violinText": "Violine",
|
||||
"barText": "Balken",
|
||||
"boxText": "Feld",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "Die globale Featurerelevanz wird berechnet, indem der Durchschnitt des absoluten Werts der Featurerelevanz aller Punkte ermittelt wird (L1-Normalisierung). ",
|
||||
"multiclassImportanceAddendum": "Bei der Berechnung der Featurerelevanz für alle Klassen werden alle Punkte berücksichtigt, es wird keine differenzielle Gewichtung verwendet. Ein Feature mit hoher negativer Relevanz, bei dem viele Punkte in der Vorhersage nicht der \"Klasse A\" zugeordnet werden, führt daher zu einer erheblich erhöhten Featurerelevanz der \"Klasse A\"."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Gleich",
|
||||
"greaterThanComparison": "Größer als",
|
||||
"greaterThanEqualToComparison": "Größer oder gleich",
|
||||
"lessThanComparison": "Kleiner als",
|
||||
"lessThanEqualToComparison": "Kleiner oder gleich",
|
||||
"inTheRangeOf": "Im Bereich von",
|
||||
"categoricalIncludeValues": "Enthaltene Werte:",
|
||||
"numericValue": "Wert",
|
||||
"numericalComparison": "Vorgang",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Maximum",
|
||||
"min": "Mindestwert: {0}",
|
||||
"max": "Höchstwert: {0}",
|
||||
"uniqueValues": "Anzahl eindeutiger Werte: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Regressionsfehler",
|
||||
"error": "Fehler",
|
||||
"classificationOutcome": "Klassifizierungsergebnis",
|
||||
"truePositive": "True Positive",
|
||||
"trueNegative": "True Negative",
|
||||
"falsePositive": "False Positive",
|
||||
"falseNegative": "False Negative",
|
||||
"dataset": "Dataset",
|
||||
"predictedProbabilities": "Vorhersagewahrscheinlichkeiten",
|
||||
"none": "Anzahl"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Schließen",
|
||||
"defaultCustomRootName": "Kopie von Zeile {0}",
|
||||
"filterFeaturePlaceholder": "Features durchsuchen"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Kohorte",
|
||||
"defaultLabel": "Alle Daten"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Untersuchen Sie die wichtigsten k Funktionen, die sich auf die Gesamtvorhersagen Ihrer Modelle auswirken (globale Erklärungen). Verwenden Sie den Schieberegler, um absteigende Featurerelevanzwerte anzuzeigen. Wählen Sie bis zu drei Kohorten aus, um die zugehörigen Featurerelevanzwerte nebeneinander anzuzeigen. Klicken Sie auf einen der Featurebalken im Diagramm, um anzuzeigen, wie sich die Werte des ausgewählten Features auf die Modellvorhersage auswirken.",
|
||||
"topAtoB": "Erste {0} bis {1} Features",
|
||||
"datasetCohorts": "Datasetkohorten",
|
||||
"legendHelpText": "Aktivieren und deaktivieren Sie Kohorten im Plot, indem Sie auf die Legendenelemente klicken.",
|
||||
"sortBy": "Sortieren nach",
|
||||
"viewDependencePlotFor": "Abhängigkeitsplot anzeigen für:",
|
||||
"datasetCohortSelector": "Datasetkohorte auswählen",
|
||||
"aggregateFeatureImportance": "Aggregierte Featurerelevanz",
|
||||
"missingParameters": "Auf dieser Registerkarte muss der Parameter für die lokale Featurerelevanz angegeben werden.",
|
||||
"weightOptions": "Gewichtungen der Klassenrelevanz",
|
||||
"dependencePlotTitle": "Abhängigkeitsplots",
|
||||
"dependencePlotHelperText": "Dieser Abhängigkeitsplot zeigt die Beziehung zwischen dem Wert eines Features und der entsprechenden Relevanz des Features in einer Kohorte.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Feature auswählen",
|
||||
"datasetRequired": "Für Abhängigkeitsplots ist das Array aus Auswertungsdataset und lokaler Featurerelevanz erforderlich."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Datenstatistik",
|
||||
"datapoints": "{0} Datenpunkte",
|
||||
"features": "{0} Features",
|
||||
"filters": "{0} Filter",
|
||||
"binaryClassifier": "Binäre Klassifizierung",
|
||||
"regressor": "Regressor",
|
||||
"multiclassClassifier": "Klassifizierung für mehrere Klassen",
|
||||
"datasetCohorts": "Datasetkohorten",
|
||||
"editCohort": "Kohorte bearbeiten",
|
||||
"duplicateCohort": "Doppelte Kohorte",
|
||||
"addCohort": "Kohorte hinzufügen",
|
||||
"copy": " Kopie"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Bewerten Sie die Leistung Ihres Modells, indem Sie die Verteilung Ihrer Vorhersagewerte und die Werte der Leistungsmetriken Ihres Modells erkunden. Sie können Ihr Modell näher untersuchen, indem Sie sich eine vergleichende Analyse seiner Leistung in verschiedenen Kohorten oder Untergruppen Ihres Datasets ansehen. Wählen Sie Filter entlang des y-Werts und des x-Werts aus, um unterschiedliche Dimensionen einzubeziehen. Wählen Sie das Zahnrad im Diagramm aus, um den Diagrammtyp zu ändern.",
|
||||
"modelStatistics": "Modellstatistiken",
|
||||
"cohortPickerLabel": "Wählen Sie eine Datasetkohorte zur Untersuchung aus.",
|
||||
"missingParameters": "Auf dieser Registerkarte muss das Array mit vorhergesagten Werten aus dem Modell angegeben werden.",
|
||||
"missingTrueY": "Für die Modellleistungsstatistik müssen zusätzlich zu den vorhergesagten Ergebnissen die tatsächlichen Ergebnisse angegeben werden."
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y-Wert",
|
||||
"numberOfDatapoints": "Anzahl von Datenpunkten",
|
||||
"xValue": "X-Wert",
|
||||
"rowIndex": "Zeilenindex",
|
||||
"featureImportance": "Featurerelevanz",
|
||||
"countTooltipPrefix": "Anzahl: {0}",
|
||||
"count": "Anzahl",
|
||||
"featurePrefix": "Feature",
|
||||
"importancePrefix": "Relevanz",
|
||||
"cohort": "Kohorte",
|
||||
"howToRead": "Lesen dieses Diagramms"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Untersuchen Sie Ihre Datasetstatistik, indem Sie verschiedene Filter für die X-, Y- und Farbachse auswählen, um Slices Ihrer Daten anhand verschiedener Dimensionen zu erstellen. Erstellen Sie oben Datensatzkohorten, um Datensatzstatistiken mit Filtern wie Datensatzfeatures, Fehlergruppen und dem vorhergesagten Ergebnis zu analysieren. Verwenden Sie das Zahnradsymbol in der oberen rechten Ecke des Diagramms, um die Diagrammtypen zu ändern.",
|
||||
"colorValue": "Farbwert",
|
||||
"individualDatapoints": "Einzelne Datenpunkte",
|
||||
"aggregatePlots": "Aggregierte Plots",
|
||||
"chartType": "Diagrammtyp",
|
||||
"missingParameters": "Auf dieser Registerkarte muss ein Auswertungsdataset angegeben werden.",
|
||||
"noColor": "Keine"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Featurerelevanz von",
|
||||
"placeholder": "Klicken Sie im obigen Balkendiagramm auf ein Feature, um den zugehörigen Abhängigkeitsplot anzuzeigen."
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Sie können einen Datenpunkt auswählen, indem Sie auf das Streudiagramm klicken, um unten die zugehörigen lokalen Featurerelevanzwerte (lokale Erklärungen) und den ICE-Plot (Individual Conditional Expectation) anzuzeigen. Erstellen Sie einen hypothetischen Was-wäre-wenn-Datenpunkt, indem Sie den Bereich rechts verwenden, um Features eines bekannten Datenpunkts zu stören. Featurerelevanzwerte basieren auf zahlreichen Näherungen und sind nicht die \"Ursache\" von Vorhersagen. Ohne absolute mathematische Belastbarkeit der kausalen Rückschlüsse raten wir Benutzern davon ab, basierend auf diesem Tool Entscheidungen in der Praxis zu treffen.",
|
||||
"panelPlaceholder": "Für Vorhersagen für neue Datenpunkte ist ein Modell erforderlich.",
|
||||
"cohortPickerLabel": "Wählen Sie eine Datasetkohorte zur Untersuchung aus.",
|
||||
"scatterLegendText": "Aktivieren und deaktivieren Sie die Datenpunkte im Plot, indem Sie auf die Legendenelemente klicken.",
|
||||
"realPoint": "Reale Datenpunkte",
|
||||
"noneSelectedYet": "Noch keine ausgewählt.",
|
||||
"whatIfDatapoints": "Was-wäre-wenn-Datenpunkte",
|
||||
"noneCreatedYet": "Noch keine erstellt.",
|
||||
"showLabel": "Anzeigen:",
|
||||
"featureImportancePlot": "Featurerelevanzplot",
|
||||
"icePlot": "ICE-Plot (Individual Conditional Expectation)",
|
||||
"featureImportanceLackingParameters": "Geben Sie Werte für die lokale Featurerelevanz an, um festzustellen, wie sich jedes Feature auf die einzelnen Vorhersagen auswirkt.",
|
||||
"featureImportanceGetStartedText": "Wählen Sie einen Punkt für die Anzeige der Featurerelevanz aus.",
|
||||
"iceLackingParameters": "Für ICE-Plots ist ein operationalisiertes Modell erforderlich, um Vorhersagen für hypothetische Datenpunkte zu treffen.",
|
||||
"IceGetStartedText": "Wählen Sie einen Punkt aus, oder erstellen Sie einen Was-wäre-wenn-Punkt, um ICE-Plots anzuzeigen.",
|
||||
"whatIfDatapoint": "Was-wäre-wenn-Datenpunkt",
|
||||
"whatIfHelpText": "Wählen Sie einen Punkt im Plot aus, oder geben Sie einen bekannten Datenpunktindex manuell ein, um eine Störung hervorzurufen und ihn als neuen Was-wäre-wenn-Punkt zu speichern.",
|
||||
"indexLabel": "Datenindex für Störung",
|
||||
"rowLabel": "Zeile {0}",
|
||||
"whatIfNameLabel": "Name des Was-wäre-wenn-Datenpunkts",
|
||||
"featureValues": "Featurewerte",
|
||||
"predictedClass": "Vorhergesagte Klasse: ",
|
||||
"predictedValue": "Vorhergesagter Wert: ",
|
||||
"probability": "Wahrscheinlichkeit: ",
|
||||
"trueClass": "TRUE-Klasse: ",
|
||||
"trueValue": "TRUE-Wert: ",
|
||||
"trueValue.comment": "Präfix der tatsächlichen Beschriftung für Regression",
|
||||
"newPredictedClass": "Neue vorhergesagte Klasse: ",
|
||||
"newPredictedValue": "Neuer vorhergesagter Wert: ",
|
||||
"newProbability": "Neue Wahrscheinlichkeit: ",
|
||||
"saveAsNewPoint": "Als neuen Punkt speichern",
|
||||
"saveChanges": "Änderungen speichern",
|
||||
"loading": "Wird geladen...",
|
||||
"classLabel": "Klasse: {0}",
|
||||
"minLabel": "Min.",
|
||||
"maxLabel": "Max.",
|
||||
"stepsLabel": "Schritte",
|
||||
"disclaimer": "Haftungsausschluss: Diese Erläuterungen basieren auf zahlreichen Näherungswerten und stellen nicht die \"Ursache\" von Vorhersagen dar. Ohne absolute mathematische Belastbarkeit der kausalen Rückschlüsse raten wir Benutzern davon ab, basierend auf diesem Tool Entscheidungen in der Praxis zu treffen.",
|
||||
"missingParameters": "Auf dieser Registerkarte muss ein Auswertungsdataset angegeben werden.",
|
||||
"selectionLimit": "Maximal 3 ausgewählte Punkte",
|
||||
"classPickerLabel": "Klasse",
|
||||
"tooltipTitleMany": "Erste {0} vorhergesagte Klassen",
|
||||
"whatIfTooltipTitle": "Klassen mit Was-wäre-wenn-Vorhersage",
|
||||
"tooltipTitleFew": "Vorhergesagte Klassen",
|
||||
"probabilityLabel": "Wahrscheinlichkeit",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "Der Wert muss numerisch sein.",
|
||||
"icePlotHelperText": "ICE-Plots zeigen, wie sich die Vorhersagewerte des ausgewählten Datenpunkts entlang eines Bereichs von Featurewerten zwischen einem Mindest- und einem Höchstwert ändern."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Filter auswählen",
|
||||
"TreatAsCategorical": "Als kategorisch behandeln",
|
||||
"addFilter": "Filter hinzufügen",
|
||||
"addedFilters": "Hinzugefügte Filter",
|
||||
"noAddedFilters": "Noch keine Filter hinzugefügt.",
|
||||
"defaultFilterState": "Wählen Sie einen Filter aus, um Ihrer Datasetkohorte Parameter hinzuzufügen.",
|
||||
"cohortNameLabel": "Name der Datasetkohorte",
|
||||
"cohortNamePlaceholder": "Kohorte benennen",
|
||||
"save": "Speichern",
|
||||
"delete": "Löschen",
|
||||
"cancel": "Abbrechen",
|
||||
"cohortNameError": "Kohortenname fehlt.",
|
||||
"placeholderName": "Kohorte \"{0}\""
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Auswählen",
|
||||
"ditherLabel": "Dithern durchführen",
|
||||
"selectFilter": "Achsenwert auswählen",
|
||||
"selectFeature": "Feature auswählen",
|
||||
"binLabel": "Quantisierung auf Daten anwenden",
|
||||
"TreatAsCategorical": "Als kategorisch behandeln",
|
||||
"numOfBins": "Datengruppenanzahl",
|
||||
"groupByCohort": "Nach Kohorte gruppieren",
|
||||
"selectClass": "Klasse auswählen",
|
||||
"countHelperText": "Ein Histogramm der Punkteanzahl"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Vorhergesagte Wahrscheinlichkeit",
|
||||
"predictedY": "Vorhersage für Y",
|
||||
"evalData": "Auswertungsdataset",
|
||||
"localFeatureImportance": "Lokale Featurerelevanz",
|
||||
"inconsistentDimensions": "Inkonsistente Dimensionen. \"{0}\" weist die Dimensionen \"{1}\" auf, erwartet: {2}.",
|
||||
"notNonEmpty": "Die Eingabe \"{0}\" ist kein nicht leeres Array.",
|
||||
"varyingLength": "Inkonsistente Dimensionen. \"{0}\" weist Elemente unterschiedlicher Länge auf.",
|
||||
"notArray": "\"{0}\" ist kein Array. Es wird ein Array mit der Dimension \"{1}\" erwartet.",
|
||||
"errorHeader": "Einige Eingabeparameter waren inkonsistent und werden nicht verwendet: ",
|
||||
"datasizeWarning": "Das Auswertungsdataset ist zu groß, um in einigen Diagrammen effektiv angezeigt zu werden. Fügen Sie Filter hinzu, um die Größe der Kohorte zu verringern. ",
|
||||
"datasizeError": "Die ausgewählte Kohorte ist zu groß. Fügen Sie Filter hinzu, um die Größe der Kohorte zu verringern.",
|
||||
"addFilters": "Filter hinzufügen"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " enthält {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} und {1} andere"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "Mittlere quadratische Abweichung: {0}",
|
||||
"rSquared": "Bestimmtheitsmaß: {0}",
|
||||
"meanPrediction": "Mittlere Vorhersage: {0}",
|
||||
"accuracy": "Genauigkeit: {0}",
|
||||
"precision": "Genauigkeit: {0}",
|
||||
"recall": "Abruf: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Untersuchen Sie die Features mit Top-k-Relevanz, die sich auf Ihre Modellvorhersagen insgesamt auswirken. Verwenden Sie den Schieberegler, um Featurerelevanzen absteigend anzuzeigen."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Was bedeuten diese Erklärungen?",
|
||||
"clickHere": "Weitere Informationen",
|
||||
"shapTitle": "Shapley-Werte",
|
||||
"shapDescription": "Dieses Erklärmodul verwendet SHAP, einen spieletheoretischen Ansatz zur Erklärung von Modellen, bei dem die Relevanz von Featuresätzen gemessen wird, indem die betreffenden Features durch Marginalisierung vor dem Modell \"versteckt\" werden. Klicken Sie auf den Link unten, um mehr zu erfahren.",
|
||||
"limeTitle": "LIME (Local Interpretable Model-Agnostic Explanations)",
|
||||
"limeDescription": "Dieses Erklärmodul verwendet LIME, um eine lineare Näherung des Modells bereitzustellen. Gehen Sie wie folgt vor, um eine Erklärung zu erhalten: Fügen Sie Störungen in die Instanz ein, rufen Sie Modellvorhersagen ab, und verwenden Sie diese Vorhersagen als Beschriftungen, um ein lineares Modell geringer Datendichte zu trainieren, das lokal zuverlässig ist. Die Gewichtungen dieses linearen Modells werden als Featurerelevanz verwendet. Klicken Sie auf den Link unten, um weitere Informationen zu erhalten.",
|
||||
"mimicTitle": "Nachahmung (globales Ersatzmodell zur Erklärung)",
|
||||
"mimicDescription": "Dieses Erklärmodul basiert auf der Idee, globale Ersatzmodelle zur Nachahmung von Blackbox-Modellen zu trainieren. Ein globales Ersatzmodell ist ein intrinsisch interpretierbares Modell, das so trainiert wird, dass es sich so genau wie möglich an die Vorhersagen eines Black Box-Modells annähert. Die Werte der Featurerelevanz sind modellbasierte Featurerelevanzwerte des zugrunde liegenden Ersatzmodells (LightGBM, lineare Regression, stochastischer Gradientenabstieg oder Entscheidungsstruktur).",
|
||||
"pfiTitle": "Permutation Feature Importance (PFI)",
|
||||
"pfiDescription": "Dieses Erklärmodul ordnet die Daten für den gesamten Datensatz – jeweils ein Feature nach dem anderen – nach dem Zufallsprinzip an und berechnet, wie sich die relevante Leistungsmetrik ändert (standardmäßige Leistungsmetriken: F1 für binäre Klassifikation, F1-Bewertung mit Mikrodurchschnitt für Mehrklassenklassifikation und mittlerer absoluter Fehler für Regression). Je größer die Änderung, desto relevanter ist das Feature. Dieses Erklärmodul kann nur das Gesamtverhalten des zugrunde liegenden Modells erklären, nicht jedoch einzelne Vorhersagen. Der Wert der Featurerelevanz repräsentiert das Delta für die Modellleistung, indem dieses bestimmte Feature gestört wird."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,558 +0,0 @@
|
|||
{
|
||||
"calloutTitle": "Click for info",
|
||||
"selectPoint": "Select a point to see its local explanation",
|
||||
"_selectPoint.comment": "Prompts the user to select a point",
|
||||
"defaultClassNames": "Class {0}",
|
||||
"_defaultClassNames.comment": " models that output classes have this as the default class names",
|
||||
"defaultFeatureNames": "Feature {0}",
|
||||
"_defaultFeatureNames.comment": "the default column names",
|
||||
"absoluteAverage": "Average of absolute value",
|
||||
"_absoluteAverage.comment": "https://en.wikipedia.org/wiki/Norm_(mathematics) Absolute value norm",
|
||||
"predictedClass": "Predicted class",
|
||||
"_predictedClass.comment": "Norm based on the predicted class rather than absolute value as above",
|
||||
"datasetExplorer": "Dataset Explorer",
|
||||
"dataExploration": "Dataset Exploration",
|
||||
"_dataExploration.comment": "Label for tab showing scatterplot of dataset and predictions",
|
||||
"aggregateFeatureImportance": "Aggregate Feature Importance",
|
||||
"_aggregateFeatureImportance.comment": "tab label for view of aggregated (summed and averaged) feature importances",
|
||||
"globalImportance": "Global Importance",
|
||||
"_globalImportance.comment": "Label for tab showing bar chart of importance of features at a global level",
|
||||
"explanationExploration": "Explanation Exploration",
|
||||
"individualImportance": "Individual Feature Importance",
|
||||
"individualAndWhatIf": "Individual Feature Importance & What-If",
|
||||
"_individualAndWhatIf.comment": "tab label for feature importances of single rows, and 'what if?' which allows seeing hypothetical results",
|
||||
"_explanationExploration.comment": "Label for tab showing scatter plot of dataset and importance of features data",
|
||||
"summaryImportance": "Summary Importance",
|
||||
"_summmaryImportance.comment": "Label showing all importance of feature datapoints in scatter plot",
|
||||
"featureImportance": "Feature Importance",
|
||||
"_featureImportance.comment": "Label for feature importance ",
|
||||
"featureImportanceOf": "Feature importance of {0}",
|
||||
"perturbationExploration": "Perturbation Exploration",
|
||||
"_perturbationExploration.comment": "Label for local tab allowing the user to change parameters on a selected point",
|
||||
"localFeatureImportance": "Local Feature Importance",
|
||||
"_localFeatureImportance.comment": "Label for local tab showing feature importance of selected point",
|
||||
"ice": "ICE",
|
||||
"_ice.comment": "Label for tab showing https://christophm.github.io/interpretable-ml-book/ice.html",
|
||||
"clearSelection": "Clear selection",
|
||||
"feature": "Feature:",
|
||||
"intercept": "Intercept",
|
||||
"_intercept.comment": "The label for the linear intercept, the bias value",
|
||||
"modelPerformance": "Model Performance",
|
||||
"_modelPerformance.comment": "tab label to see how well a model predicted values when compared to true",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Data : {0}",
|
||||
"_dataLabel.comment": "prepend in front of column names",
|
||||
"importanceLabel": "Importance : {0}",
|
||||
"_importanceLabel.comment": "prepend in front of feature importance of column name",
|
||||
"predictedY": "Predicted Y",
|
||||
"_predictedY.comment": "predicted output",
|
||||
"index": "Index",
|
||||
"_index.comment": "the index value (an integer of the row number)",
|
||||
"dataGroupLabel": "Data",
|
||||
"output": "Output",
|
||||
"probabilityLabel": "Probability : {0}",
|
||||
"_probabilityLabel.comment": "Probability prefix for all classes in a multiclass problem",
|
||||
"trueY": "True Y",
|
||||
"_trueY.comment": "The true value to be predicted",
|
||||
"class": "class: ",
|
||||
"_class.comment": "label for predicted class",
|
||||
"xValue": "X value:",
|
||||
"_xValue.comment": "label for x value dropdown",
|
||||
"yValue": "Y value:",
|
||||
"_yValue.comment": "label for y value dropdown",
|
||||
"colorValue": "Color:",
|
||||
"_colorValue.comment": "label for selecting color value",
|
||||
"count": "Count",
|
||||
"_count.comment": "the default axis is the count of items"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Cross-class weighting:",
|
||||
"_label.comment": "label for dropdown allowing user to select how importance across multiple output classes is aggregated",
|
||||
"info": "Information on cross-class calculation",
|
||||
"_info.comment": "tooltip on hover",
|
||||
"overviewInfo": "Multiclass models generate an independent feature importance vector for each class. Each class's feature importance vector demonstrates which features made a class more likely or less likely. You can select how the weights of the per-class feature importance vectors are summarized into a single value:",
|
||||
"_overview.comment": "explains absolute value weights",
|
||||
"absoluteValInfo": "Average of absolute value: Shows the sum of the feature's importance across all possible classes, divided by number of classes",
|
||||
"predictedClassInfo": "Predicted class: Shows the feature importance value for a given point's predicted class",
|
||||
"_predictedClassInfo.comment": "explains predicted class weight",
|
||||
"enumeratedClassInfo": "Enumerated class names: Shows only the specified class's feature importance values across all data points.",
|
||||
"_enumeratedClassInfo.comment": "explains the weights for selecting a single class",
|
||||
"close": "Close",
|
||||
"crossClassWeights": "Cross class weights"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Scaled Feature Value",
|
||||
"_scaledFeatureValue.comment": "The chart shows all data in a color scale (normalized to 0 - 1). This is the label for the color bar",
|
||||
"low": "Low",
|
||||
"_low.comment": "The low end of the color bar",
|
||||
"high": "High",
|
||||
"_high.comment": "label for the high end of the color bar",
|
||||
"featureLabel": "Feature: {0}",
|
||||
"_featureLabel.comment": "Prefix to the feature name",
|
||||
"valueLabel": "Feature value: {0}",
|
||||
"_valueLabel.comment": "prefix to the feature value",
|
||||
"importanceLabel": "Importance: {0}",
|
||||
"_importanceLabel.comment": "prefix to the feature importance",
|
||||
"predictedClassTooltip": "Predicted Class: {0}",
|
||||
"_predictedClassTooltip.comment": "prefixed in front of the output class names predicted by the model",
|
||||
"trueClassTooltip": "True Class: {0}",
|
||||
"_trueClassTooltip.comment": "prefixed in front of the true class labels",
|
||||
"predictedOutputTooltip": "Predicted Output: {0}",
|
||||
"_predictedOutputTooltip.comment": "prefixed in front of the output in a regression model (numeric, no classes)",
|
||||
"trueOutputTooltip": "True Output: {0}",
|
||||
"_trueOutputTooltip.comment": "prefixed in front of the true value in a regression model (numeric, no classes)",
|
||||
"topKFeatures": "Top K Features:",
|
||||
"_topKFeatures.comment": "Label for slider to show only the top (k) most important features, where the slider is used to set the value of k",
|
||||
"topKInfo": "How top k is calculated",
|
||||
"predictedValue": "Predicted Value",
|
||||
"_predictedValue.comment": "Label for dropdown option, group data by the predicted value from the model (numeric values)",
|
||||
"predictedClass": "Predicted Class",
|
||||
"_predictedClass.comment": "Label for dropdown option, group data by predicted class from model",
|
||||
"trueValue": "True Value",
|
||||
"_trueValue.comment": "label for dropdown option, group data by true value (numeric)",
|
||||
"trueClass": "True Class",
|
||||
"_trueClass.comment": "label for dropdown, group data by true class",
|
||||
"noColor": "None",
|
||||
"_noColor.comment": "label for dropdown, do not apply any grouping",
|
||||
"tooManyRows": "The provided dataset is larger than this chart can support",
|
||||
"_tooManyRows.comment": "error message if the dataset is too large to visualize"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Class: {0}",
|
||||
"_classLabel.comment": "Prefix for class",
|
||||
"sortBy": "Sort by",
|
||||
"_sortBy.comment": "prompt for setting how values are sorted",
|
||||
"noData": "No Data",
|
||||
"_noData.comment": "Error message for no applicable data",
|
||||
"absoluteGlobal": "Absolute global",
|
||||
"_absoluteGlobal.comment": "sorting option, sort by the absolute value of the importance of all datapoints",
|
||||
"absoluteLocal": "Absolute local",
|
||||
"_absoluteLocal.comment": "sorting option, sort by the absolute value of the importance for the single selected point",
|
||||
"calculatingExplanation": "Calculating explanation",
|
||||
"_calculatingExplanation.comment": "loading message"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Must be numeric",
|
||||
"_numericError.comment": "error message if non-numeric characters typed",
|
||||
"integerError": "Must be an integer",
|
||||
"_integerError.comment": "error message if non-integer values typed by user",
|
||||
"prediction": "Prediction",
|
||||
"_prediction.comment": "Prediction label for y-axis",
|
||||
"predictedProbability": "Predicted probability",
|
||||
"_predictedProbability.comment": "predicted probability label for y-axis",
|
||||
"predictionLabel": "Prediction: {0}",
|
||||
"_predictionLabel.comment": "prediction hover prefix",
|
||||
"probabilityLabel": "Probability: {0}",
|
||||
"_probabilityLabel.comment": "probability hover prefix",
|
||||
"noModelError": "Please provide an operationalized model to explore predictions in ICE plots.",
|
||||
"_noModelError.comment": "error message for no model present",
|
||||
"featurePickerLabel": "Feature:",
|
||||
"_featurePicker.comment": "feature dropdown label",
|
||||
"minimumInputLabel": "Minimum:",
|
||||
"_minimumInputLabel.comment": "Set minimum bounds label",
|
||||
"maximumInputLabel": "Maximum:",
|
||||
"_maximumInputLabel.comment": "set maximum bounds label",
|
||||
"stepInputLabel": "Steps:",
|
||||
"_stepInputLabel.comment": "number of samples to include between minimum and maximum (integer)",
|
||||
"loadingMessage": "Loading data...",
|
||||
"_loadingMessage.comment": "loading message",
|
||||
"submitPrompt": "Submit a range to view an ICE plot",
|
||||
"_submitPrompt.comment": "prompt to user giving instructions to enter a numeric range",
|
||||
"topLevelErrorMessage": "Error in parameter",
|
||||
"_topLevelErrorMessage.comment": "error message for any parameter issue",
|
||||
"errorPrefix": "Error encountered: {0}",
|
||||
"_errorPrefix.comment": "prefix in front of external error"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Loading...",
|
||||
"perturbationLabel": "Perturbation:",
|
||||
"_perturbationLabel.comment": "Perturbation (ie. the user has made a set of small changes to the original data -- a perturbation. This is the label for the resulting prediction)"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Predicted value : {0}",
|
||||
"_predictionValueLabel.comment": "label for prediction of numeric values",
|
||||
"predictedClassLabel": "Predicted class : {0}",
|
||||
"_predictedClassLabel.comment": "label for prediction of class value"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "No grouping",
|
||||
"_groupName.comment": "Do not group data option",
|
||||
"groupPredicted": "Predicted Y",
|
||||
"_groupPredicted.comment": "option to group data by predicted class",
|
||||
"groupTrue": "True Y",
|
||||
"_groupTrue.comment": "option to group data by true class",
|
||||
"groupBy": "Group by",
|
||||
"_groupBy.comment": "Group by prompt for dropdown to select how data should be grouped"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Chart type:",
|
||||
"_chartType.comment": "label for dropdown to select chart format",
|
||||
"violinText": "Violin",
|
||||
"_violinText.comment": "a violin plot https://en.wikipedia.org/wiki/Violin_plot",
|
||||
"barText": "Bar",
|
||||
"_barText.comment": "a bar plot ",
|
||||
"boxText": "Box",
|
||||
"_boxText.comment": "a box plot https://en.wikipedia.org/wiki/Box_plot",
|
||||
"beehiveText": "Swarm",
|
||||
"_beehiveText.comment": "A swarm plot (its like a scatter plot with categorical x axis with dithering, see examples https://seaborn.pydata.org/generated/seaborn.swarmplot.html)",
|
||||
"globalImportanceExplanation": "Global feature importance is calculated by averaging the absolute value of the feature importance of all points (L1 normalization). ",
|
||||
"_globalImportanceExplanation.comment": "explains how global feature importance is calculated ",
|
||||
"multiclassImportanceAddendum": "All points are included in calculating a feature's importance for all classes, no differential weighting is used. So a feature that has large negative importance for many points predicted to not be of 'Class A' will greatly increase that feature's 'Class A' importance.",
|
||||
"_multiclassImportanceAddendum.comment": "explains how global importance is calculated for each class in a multiclass case."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Equal to",
|
||||
"_equalComparison.comment": "filter for rows that are exactly equal",
|
||||
"greaterThanComparison": "Greater than",
|
||||
"_greaterThanComparison.comment": "filter for rows that are greater than a value",
|
||||
"greaterThanEqualToComparison": "Greater than or equal to",
|
||||
"_greaterThanEqualToComparison.comment": "filter for rows that are greater than or equal to a value",
|
||||
"lessThanComparison": "Less than",
|
||||
"_lessThanComparison.comment": "filter for rows that are less than a value",
|
||||
"lessThanEqualToComparison": "Less than or equal to",
|
||||
"_lessThanEqualToComparison.comment": "filter for rows that are less than or equal to a value",
|
||||
"inTheRangeOf": "In the range of",
|
||||
"_inTheRangeOf.comment": "filter for rows that are between two values",
|
||||
"categoricalIncludeValues": "Included values:",
|
||||
"_categoricalIncludeValues.comment": "filter to selected categories",
|
||||
"numericValue": "Value",
|
||||
"_numericValue.comment": "the value to compare to in greater/less than or equal to filter",
|
||||
"numericalComparison": "Operation",
|
||||
"_numericalComparison.comment": "label for dropdown containing [greater than, less than, equal to]",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Maximum",
|
||||
"min": "Min: {0}",
|
||||
"max": "Max: {0}",
|
||||
"uniqueValues": "# of unique values: {0}",
|
||||
"_uniqueValues.comment": "the number of unique values for a selected categorical filter"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Regression error",
|
||||
"_regressionError.comment": "true value minus predicted value is regression error",
|
||||
"error": "Error",
|
||||
"classificationOutcome": "Classification outcome",
|
||||
"_classificationOutcome.comment": "Whether the prediction from the model matched the true value",
|
||||
"truePositive": "True positive",
|
||||
"trueNegative": "True negative",
|
||||
"falsePositive": "False positive",
|
||||
"falseNegative": "False negative",
|
||||
"dataset": "Dataset",
|
||||
"predictedProbabilities": "Prediction probabilities",
|
||||
"none": "Count",
|
||||
"_none.comment": "option to not have data on this axis, instead just counts number of points"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Close",
|
||||
"defaultCustomRootName": "Copy of row {0}",
|
||||
"_defaultCustomRootName.comment": "default prefix for a hypothetical point made by copying another point",
|
||||
"filterFeaturePlaceholder": "Search features",
|
||||
"_filterFeaturePlaceholder.comment": "placeholder in search box for searching for features by name"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Cohort",
|
||||
"_cohort.comment": "a subset of the data is called a cohort",
|
||||
"defaultLabel": "All data"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"_helperText.comment": "paragraph summarizing the view on this page and available actions",
|
||||
"helperText": "Explore the top-k important features that impact your overall model predictions (a.k.a. global explanation). Use the slider to show descending feature importance values. Select up to three cohorts to see their feature importance values side by side. Click on any of the feature bars in the graph to see how values of the selected feature impact model prediction.",
|
||||
"topAtoB": "Top {0}-{1} features",
|
||||
"_topAtoB.comment": "label on a slider, will tell user the index of the features they are currently seeing, like Top 5-10 features",
|
||||
"datasetCohorts": "Dataset cohorts",
|
||||
"_datasetCohorts.comment": "label for dropdown allowing users to select what cohorts to view",
|
||||
"legendHelpText": "Toggle cohorts on and off in the plot by clicking on the legend items.",
|
||||
"_legendHelperText.comment": "explanatory text on what actions can be done on a list of cohorts",
|
||||
"sortBy": "Sort by",
|
||||
"_sortBy.comment": "prompt for setting how values are sorted",
|
||||
"viewDependencePlotFor": "View dependence plot for:",
|
||||
"_viewDependencePlotFor.comment": "label for dropdown to select feature to be shown in a dependence plot (a kind of graph)",
|
||||
"datasetCohortSelector": "Select a dataset cohort",
|
||||
"_datasetCohortSelector.comment": "label for selecting single cohort",
|
||||
"aggregateFeatureImportance": "Aggregate Feature Importance",
|
||||
"_aggregateFeatureImportance.comment": "graph label for aggregated (summed and averaged) feature importances",
|
||||
"missingParameters": "This tab requires the local feature importance parameter be supplied.",
|
||||
"_missingParameters.comment": "Show a message if the required feature importance parameter is not provided",
|
||||
"weightOptions": "Class importance weights",
|
||||
"_weightOptions.comment": "Weight how importance values are averaged https://en.wikipedia.org/wiki/Weighted_arithmetic_mean",
|
||||
"dependencePlotTitle": "Dependence Plots",
|
||||
"dependencePlotHelperText": "This dependence plot shows the relationship between the value of a feature to the corresponding importance of the feature across a cohort.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Select feature",
|
||||
"datasetRequired": "Dependence plots require the evaluation dataset and local feature importance array."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Data Statistics",
|
||||
"_dataStatistics.comment": "label for section containing statistics about the dataset",
|
||||
"datapoints": "{0} datapoints",
|
||||
"_datapoints.comment": "formatted string of the number of datapoints in the dataset",
|
||||
"features": "{0} features",
|
||||
"_features.comment": "formatted string of the number of features (columns) in a dataset",
|
||||
"filters": "{0} filters",
|
||||
"_filters.comment": "the number of filters that define a cohort",
|
||||
"binaryClassifier": "Binary Classifier",
|
||||
"_binaryClassifier.comment": "a model that predicts true or false is a binary classifier, this is the label",
|
||||
"regressor": "Regressor",
|
||||
"_regressor.comment": "a class of models that output a numeric score, name is derived from statistical regression",
|
||||
"multiclassClassifier": "Multiclass Classifier",
|
||||
"_multiclassClassifier.comment": "models that output a category, with more than two categories",
|
||||
"datasetCohorts": "Dataset Cohorts",
|
||||
"_datasetCohorts.comment": "a subset of the original data, defined by filtering the data. This is the label for presenting all cohorts the user created",
|
||||
"editCohort": "Edit Cohort",
|
||||
"_editCohort.comment": "button text to edit the filters defining an existing cohort",
|
||||
"duplicateCohort": "Duplicate Cohort",
|
||||
"_duplicateCohort.comment": "button text to copy an existing cohort",
|
||||
"addCohort": "New Cohort",
|
||||
"_addCohort.comment": "button text to create a new cohort",
|
||||
"copy": " copy",
|
||||
"_copy.comment": "suffix attached to name of cohort created by copying other cohort, by default."
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Evaluate the performance of your model by exploring the distribution of your prediction values and the values of your model performance metrics. You can further investigate your model by looking at a comparative analysis of its performance across different cohorts or subgroups of your dataset. Select filters along y-value and x-value to cut across different dimensions. Select the gear in the graph to change graph type.",
|
||||
"_helperText.comment": "explains the view on this page and what actions can be taken",
|
||||
"modelStatistics": "Model Statistics",
|
||||
"_modelStatistics.comment": "label for area listing statistics about model prediction",
|
||||
"cohortPickerLabel": "Select a dataset cohort to explore",
|
||||
"_cohortPickerLabel.comment": "label for single-select dropdown to pick a cohort to view",
|
||||
"missingParameters": "This tab requires the array of predicted values from the model be supplied.",
|
||||
"_missingPArameters.comment": "Show a message if the required prediction array not provided",
|
||||
"missingTrueY": "Model performance statistics require the true outcomes be provided in addition to the predicted outcomes",
|
||||
"_missingTrueY.comment": "Show message if true Y values are not provided, since statistics require the true value"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y-value",
|
||||
"_yValue.comment": "label for y value button on chart",
|
||||
"numberOfDatapoints": "Number of datapoints",
|
||||
"_numberOfDatapoints.comment": "some charts will always show the count of the number of rows in a group, this is the axis label on the chart",
|
||||
"xValue": "X-value",
|
||||
"_xValue.comment": "label for x value button on chart",
|
||||
"rowIndex": "Row index",
|
||||
"_rowIndex.comment": "the index of a row in a dataset",
|
||||
"featureImportance": "Feature importance",
|
||||
"countTooltipPrefix": "Count: {0}",
|
||||
"_countTooltipPrefix.comment": "on hover, show the prefix followed by the total number of points",
|
||||
"count": "Count",
|
||||
"featurePrefix": "Feature",
|
||||
"_featurePrefix.comment": "label shown before feature name in tooltip",
|
||||
"importancePrefix": "Importance",
|
||||
"_importance.comment": "label shown before importance value in tooltip",
|
||||
"cohort": "Cohort",
|
||||
"howToRead": "How to read this chart"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Explore your dataset statistics by selecting different filters along the X, Y, and color axis to slice your data along different dimensions. Create dataset cohorts above to analyze dataset statistics with filters such as predicted outcome, dataset features and error groups. Use the gear icon in the upper right-hand corner of the graph to change graph types.",
|
||||
"_helperText.comment": "paragraph summarizing the view on this page and what actions the user can take",
|
||||
"colorValue": "Color value",
|
||||
"_colorValue.comment": "label on button to set how data is mapped to color in chart",
|
||||
"individualDatapoints": "Individual datapoints",
|
||||
"_individualDatapoints.comment": "configuration option to view chart with each individual point, opposed to chart summing/averaging points",
|
||||
"aggregatePlots": "Aggregate plots",
|
||||
"_aggregatePlots.comment": "configuration option to view plots that sum or average points, opposed to sowing each point on its own",
|
||||
"chartType": "Chart type",
|
||||
"_chartType.comment": "label for dropdown to select chart format",
|
||||
"missingParameters": "This tab requires an evaluation dataset be supplied.",
|
||||
"_missingPArameters.comment": "Show a message if the required dataset parameter is not provided",
|
||||
"noColor": "None",
|
||||
"_noColor.comment": "placeholder text when no color axis picked"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Feature importance of",
|
||||
"_featureImportanceOf.comment": "axis label on chart showing the importance of a selected feature (column)",
|
||||
"placeholder": "Click on a feature on the bar chart above to show its dependence plot",
|
||||
"_placeholder.comment": "placeholder text explaining how to activate this chart, by clicking on a neighboring chart."
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "You can select a datapoint by clicking on the scatterplot to view its local feature importance values (local explanation) and individual conditional expectation (ICE) plot below. Create a hypothetical what-if datapoint by using the panel on the right to perturb features of a known datapoint. Feature importance values are based on many approximations and are not the \"cause\" of predictions. Without strict mathematical robustness of causal inference, we do not advise users to make real-life decisions based on this tool.",
|
||||
"_helperText.comment": "explains what is shown in this tab and what actions are available",
|
||||
"panelPlaceholder": "A model is required to make predictions for new data points.",
|
||||
"_panelPlaceholder.comment": "message shown to user when they did not give a model as inputs",
|
||||
"cohortPickerLabel": "Select a dataset cohort to explore",
|
||||
"_cohortPickerLabel.comment": "label for single-select dropdown to pick a cohort to view",
|
||||
"scatterLegendText": "Toggle datapoints on and off in the plot by clicking on the legend items.",
|
||||
"_scatterLegendText.comment": "describes actions possible on the legend items in the chart",
|
||||
"realPoint": "Real datapoints",
|
||||
"_realPoint.comment": "label above the list of points from the dataset",
|
||||
"noneSelectedYet": "None selected yet",
|
||||
"_noneSelectedYet.comment": "placeholder if no points are selected, where list of points would go",
|
||||
"whatIfDatapoints": "What-If datapoints",
|
||||
"_whatIfDatapoints.comment": "label above the list of what if (hypothetical) points",
|
||||
"noneCreatedYet": "None created yet",
|
||||
"_noneCreatedYet.comment": "placeholder if no hypothetical points have been created, goes where list of points would be",
|
||||
"showLabel": "Show:",
|
||||
"_showLabel.comment": "label on button to pick what additional chart to show",
|
||||
"featureImportancePlot": "Feature importance plot",
|
||||
"_featureImportancePlot.comment": "name of a plot type showing the importance of each feature in making a decision",
|
||||
"icePlot": "Individual conditional expectation (ICE) plot",
|
||||
"_icePlot.comment": "name of a plot type named Individual Conditional Expectation (ICE)",
|
||||
"featureImportanceLackingParameters": "Provide local feature importances to see how each feature impacts individual predictions.",
|
||||
"_featureImportanceLackingParameters.comment": "placeholder if no local feature importances passed in",
|
||||
"featureImportanceGetStartedText": "Select a point to view feature importance",
|
||||
"_featureImportanceGetStartedText.comment": "placeholder to prompt user to click point in neighboring chart to view this chart",
|
||||
"iceLackingParameters": "ICE plots require an operationalized model to make predictions for hypothetical datapoints.",
|
||||
"_iceLackingParameters.comment": "placeholder text if a required parameter is not passed in, cannot show ICE plot if no model passed in",
|
||||
"IceGetStartedText": "Select a point or create a What-If point to view ICE plots",
|
||||
"_IceGetStartedText.comment": "placeholder to prompt user to click point in neighboring chart to see this chart",
|
||||
"whatIfDatapoint": "What-If datapoint",
|
||||
"_whatIfDatapoint.comment": "header text for area where user writes a hypothetical point's values",
|
||||
"whatIfHelpText": "Select a point on the plot or manually enter a known datapoint index to perturb and save as a new What-If point.",
|
||||
"_whatIfHelpText.comment": "describe how to create a what-if hypothetical row",
|
||||
"indexLabel": "Data index to perturb",
|
||||
"_indexLabel.comment": "label for dropdown for selecting a row by index value",
|
||||
"rowLabel": "Row {0}",
|
||||
"_rowLabel.comment": "the label for a selected row, with its index number appended",
|
||||
"whatIfNameLabel": "What-If datapoint name",
|
||||
"_whatIfNameLabel.comment": "label above text field where user can name their what-if hypothetical point",
|
||||
"featureValues": "Feature values",
|
||||
"_featureValues.comment": "header above the list of feature names (column names)",
|
||||
"predictedClass": "Predicted class: ",
|
||||
"_predictedClass.comment": "the predicted class for a row",
|
||||
"predictedValue": "Predicted value: ",
|
||||
"_predictedValue.comment": "the predicted value for a row",
|
||||
"probability": "Probability: ",
|
||||
"_probability.comment": "the probability that the predicted class is correct",
|
||||
"trueClass": "True class: ",
|
||||
"_trueClass.comment": "prefix to actual label",
|
||||
"trueValue": "True value: ",
|
||||
"trueValue.comment": "prefix to actual label for regression",
|
||||
"newPredictedClass": "New predicted class: ",
|
||||
"_newPredictedClass.comment": "the prediction after the user changed features",
|
||||
"newPredictedValue": "New predicted value: ",
|
||||
"_newPredictedValue.comment": "the prediction after the user changed features",
|
||||
"newProbability": "New probability: ",
|
||||
"_newProbability.comment": "the probability for the new prediction",
|
||||
"saveAsNewPoint": "Save as new point",
|
||||
"_saveAsNewPoint.comment": "button to save hypothetical point",
|
||||
"saveChanges": "Save changes",
|
||||
"_saveChanges.comment": "button text to save changes made to a row",
|
||||
"loading": "Loading...",
|
||||
"_loading.comment": "loading message while prediction is made",
|
||||
"classLabel": "Class: {0}",
|
||||
"_classLabel.comment": "Prefix for class",
|
||||
"minLabel": "Min",
|
||||
"_minLabel.comment": "minimum (small space available)",
|
||||
"maxLabel": "Max",
|
||||
"_maxLabel.comment": "maximum (small space available)",
|
||||
"stepsLabel": "Steps",
|
||||
"_stepsLabel.comment": "number of increments to use between minimum and maximum",
|
||||
"disclaimer": "Disclaimer: These are explanations based on many approximations and are not the \"cause\" of predictions. Without strict mathematical robustness of causal inference, we do not advise users to make real-life decisions based on this tool.",
|
||||
"_disclaimer.comment": "the tool should not be liable for any bad predictions",
|
||||
"missingParameters": "This tab requires an evaluation dataset be supplied.",
|
||||
"_missingParameters.comment": "Show a message if the required dataset parameter is not provided",
|
||||
"selectionLimit": "Maximum of 3 selected points",
|
||||
"_selectionLimit.comment": "A user can only select 3 points from a chart at a time, this message is displayed if they click a 4th",
|
||||
"classPickerLabel": "Class",
|
||||
"tooltipTitleMany": "Top {0} predicted classes",
|
||||
"_tooltipTitleMany.comment": "placeholder is the number of classes shown",
|
||||
"whatIfTooltipTitle": "What-If predicted classes",
|
||||
"tooltipTitleFew": "Predicted classes",
|
||||
"probabilityLabel": "Probability",
|
||||
"deltaLabel": "Delta",
|
||||
"_deltaLabel.comment": "represents the change in a value",
|
||||
"nonNumericValue": "Value should be numeric",
|
||||
"icePlotHelperText": "ICE plots demonstrate how the selected datapoint's prediction values change along a range of feature values between a minimum and maximum value."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Select Filter",
|
||||
"_selectFilter.comment": "prompt to select an attribute to filter on",
|
||||
"TreatAsCategorical": "Treat as categorical",
|
||||
"_TreatAsCategorical.comment": "a checkbox label to treat integers as categories instead of as numbers",
|
||||
"addFilter": "Add Filter",
|
||||
"_addFilter.comment": "button text to add the current settings as a new filter",
|
||||
"addedFilters": "Added Filters",
|
||||
"_addedFilters.comment": "header above the list of filters that have been saved",
|
||||
"noAddedFilters": "No filters added yet",
|
||||
"_noAddedFilters.comment": "placeholder text when no filters are included",
|
||||
"defaultFilterState": "Select a filter to add parameters to your dataset cohort.",
|
||||
"_defaultFilterState.comment": "placeholder text prompting user to start making a filter",
|
||||
"cohortNameLabel": "Dataset cohort name",
|
||||
"_cohortNameLabel.comment": "label for text filed where user adds name of a cohort (subset)",
|
||||
"cohortNamePlaceholder": "Name your cohort",
|
||||
"_cohortNamePlaceholder.comment": "placeholder for cohort name",
|
||||
"save": "Save",
|
||||
"delete": "Delete",
|
||||
"cancel": "Cancel",
|
||||
"cohortNameError": "Missing cohort name",
|
||||
"_cohortNameError.comment": "error message if required name is missing",
|
||||
"placeholderName": "Cohort {0}",
|
||||
"_placeholderName.comment": "starting name for a new cohort"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Select",
|
||||
"_select.comment": "label above dropdown to promp user to pick a feature",
|
||||
"ditherLabel": "Should dither",
|
||||
"_ditherLabel.comment": "checkbox label for if small random changes should be added to numbers to more easily see large clusters",
|
||||
"selectFilter": "Select your axis value",
|
||||
"_selectFilter.comment": "label on dropdown to pick value for charting",
|
||||
"selectFeature": "Select Feature",
|
||||
"_selectFeature.comment": "dropdown label to select feature (column) for charting",
|
||||
"binLabel": "Apply binning to data",
|
||||
"_binLabel.comment": "group all values into a fixed number of groups (bins)",
|
||||
"TreatAsCategorical": "Treat as categorical",
|
||||
"_TreatAsCategorical.comment": "a checkbox label to treat integers as categories instead of as numbers",
|
||||
"numOfBins": "Number of bins",
|
||||
"_numberOfBins.comment": "the number of groups (bins) to place all values into",
|
||||
"groupByCohort": "Group by cohort",
|
||||
"_groupByCohort.comment": "if user selects to group by cohort, no further parameters to set, just show a message to fill space",
|
||||
"selectClass": "Select class",
|
||||
"_selectClass.comment": "label for dropdown listing all classes",
|
||||
"countHelperText": "A histogram of the number of points"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Predicted probability",
|
||||
"predictedY": "Predicted Y",
|
||||
"evalData": "Evaluation dataset",
|
||||
"localFeatureImportance": "Local feature importance",
|
||||
"inconsistentDimensions": "Inconsistent dimensions. {0} has dimensions {1}, expected {2}",
|
||||
"_inconsistentDimensions.comment": "Raise warning if arguments passed in have different sizes, listing dimensions of both mismatching pieces.",
|
||||
"notNonEmpty": "{0} input not a non-empty array",
|
||||
"varyingLength": "Inconsistent dimensions. {0} has elements of varying length",
|
||||
"notArray": "{0} not an array. Expected array of dimension {1}",
|
||||
"errorHeader": "Some input parameters were inconsistent and will not be used: ",
|
||||
"datasizeWarning": "The evaluation dataset is too large to be effectively displayed in some charts, please add filters to decrease the size of the cohort. ",
|
||||
"datasizeError": "The selected cohort is too large, please add filters to decrease the size of the cohort.",
|
||||
"addFilters": "Add filters"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " includes {0} ",
|
||||
"_includes.comment": "tooltip label for a filter with included values",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} and {1} others",
|
||||
"_overflowFilterArgs.comment": "first placeholder is the first one or two items in a long list, the second placeholder is the count of remaining items"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"_mse.comment": "the mean squared error, see https://en.wikipedia.org/wiki/Mean_squared_error",
|
||||
"rSquared": "R-squared: {0}",
|
||||
"_rSquared.comment": "the coefficient of determination, see https://en.wikipedia.org/wiki/Coefficient_of_determination",
|
||||
"meanPrediction": "Mean prediction {0}",
|
||||
"_meanPrediction.comment": "the average of all the predictions",
|
||||
"accuracy": "Accuracy: {0}",
|
||||
"_accuracy.comment": "computed accuracy of model on a subgroup, see https://en.wikipedia.org/wiki/Evaluation_of_binary_classifiers",
|
||||
"precision": "Precision: {0}",
|
||||
"_precision.comment": "computed precision of model, see https://en.wikipedia.org/wiki/Evaluation_of_binary_classifiers",
|
||||
"recall": "Recall: {0}",
|
||||
"_recall.comment": "computed recall of model, see https://en.wikipedia.org/wiki/Evaluation_of_binary_classifiers",
|
||||
"fpr": "FPR: {0}",
|
||||
"_fpr.comment": "False positive rate, see https://en.wikipedia.org/wiki/Evaluation_of_binary_classifiers",
|
||||
"fnr": "FNR: {0}",
|
||||
"_fnr.comment": "False negative rate, see https://en.wikipedia.org/wiki/Evaluation_of_binary_classifiers"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Explore the top k important features that impact your overall model predictions. Use the slider to show descending feature importances."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "What do these explanations mean?",
|
||||
"clickHere": "Learn more",
|
||||
"shapTitle": "Shapley values",
|
||||
"shapDescription": "This explainer uses SHAP, which is a game theoretic approach to explaining models, where the importance of features sets is measured by \"hiding\" those features from the model through marginalization. Click the link below to learn more.",
|
||||
"limeTitle": "LIME (Local Interpretable Model-Agnostic Explanations)",
|
||||
"limeDescription": "This explainer uses LIME, which provides a linear approximation of the model. To get an explanation, we do the following: perturb the instance, get model predictions, and use these predictions as labels to learn a sparse linear model that is locally faithful. The weights of this linear model are used as 'feature importances'. Click the link below to learn more.",
|
||||
"mimicTitle": "Mimic (Global Surrogate Explanations)",
|
||||
"mimicDescription": "This explainer is based on the idea of training global surrogate models to mimic blackbox models. A global surrogate model is an intrinsically interpretable model that is trained to approximate the predictions of any black box model as accurately as possible. Feature importance values are model-based feature importance values of your underlying surrogate model (LightGBM, or Linear Regression, or Stochastic Gradient Descent, or Decision Tree)",
|
||||
"pfiTitle": "Permutation Feature Importance (PFI)",
|
||||
"pfiDescription": "This explainer randomly shuffles data one feature at a time for the entire dataset and calculates how much the performance metric of interest changes (default performance metrics: F1 for binary classification, F1 Score with micro average for multiclass classification and mean absolute error for regression). The larger the change, the more important that feature is. This explainer can only explain the overall behavior of the underlying model but does not explain individual predictions. Feature importance value of a feature represents the delta in the performance of the model by perturbing that particular feature."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Seleccione un punto para ver su explicación local.",
|
||||
"defaultClassNames": "Clase {0}",
|
||||
"defaultFeatureNames": "{0} de la característica",
|
||||
"absoluteAverage": "Promedio del valor absoluto",
|
||||
"predictedClass": "Clase predicha",
|
||||
"datasetExplorer": "Explorador de conjuntos de datos",
|
||||
"dataExploration": "Exploración de conjuntos de datos",
|
||||
"aggregateFeatureImportance": "Importancia de características agregadas",
|
||||
"globalImportance": "Importancia global",
|
||||
"explanationExploration": "Exploración de explicación",
|
||||
"individualAndWhatIf": "Importancia de características individuales e hipótesis",
|
||||
"summaryImportance": "Importancia resumida",
|
||||
"featureImportance": "Importancia de la característica",
|
||||
"featureImportanceOf": "Importancia de la característica: {0}",
|
||||
"perturbationExploration": "Exploración de la perturbación",
|
||||
"localFeatureImportance": "Importancia de la característica local",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Borrar selección",
|
||||
"feature": "Característica:",
|
||||
"intercept": "Interceptar",
|
||||
"modelPerformance": "Rendimiento del modelo",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Datos: {0}",
|
||||
"importanceLabel": "Importancia: {0}",
|
||||
"predictedY": "Eje Y predicho",
|
||||
"index": "Índice",
|
||||
"dataGroupLabel": "Datos",
|
||||
"output": "Salida",
|
||||
"probabilityLabel": "Probabilidad: {0}",
|
||||
"trueY": "Y verdadero",
|
||||
"class": "clase: ",
|
||||
"xValue": "Valor de X:",
|
||||
"yValue": "Valor de Y:",
|
||||
"colorValue": "Color:",
|
||||
"count": "Recuento"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Ponderación entre clases:",
|
||||
"info": "Información sobre el cálculo entre clases",
|
||||
"overviewInfo": "Los modelos multiclase generan un vector de importancia de característica independiente para cada clase. El vector de importancia de la característica de cada clase muestra qué características han hecho que una clase sea más o menos probable. Puede seleccionar el modo en que la ponderación de los vectores de importancia de la característica por clase se resume en un único valor:",
|
||||
"absoluteValInfo": "Promedio del valor absoluto: muestra la suma de la importancia de la característica en todas las clases posibles, dividida entre el número de clases.",
|
||||
"predictedClassInfo": "Clase predicha: muestra el valor de importancia de la característica para una clase predicha de un punto determinado.",
|
||||
"enumeratedClassInfo": "Nombres de la clase enumerada: solo muestra los valores de importancia de la característica de la clase especificada en todos los puntos de datos.",
|
||||
"close": "Cerrar",
|
||||
"crossClassWeights": "Niveles de importancia de varias clases"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Valor de característica escalado",
|
||||
"low": "Baja",
|
||||
"high": "Alta",
|
||||
"featureLabel": "Característica: {0}",
|
||||
"valueLabel": "Valor de la característica: {0}",
|
||||
"importanceLabel": "Importancia: {0}",
|
||||
"predictedClassTooltip": "Clase predicha: {0}",
|
||||
"trueClassTooltip": "Clase verdadera: {0}",
|
||||
"predictedOutputTooltip": "Resultado predicho: {0}",
|
||||
"trueOutputTooltip": "Resultado verdadero: {0}",
|
||||
"topKFeatures": "Características principales de K:",
|
||||
"topKInfo": "Procedimiento para calcular los principales valores K",
|
||||
"predictedValue": "Valor de predicción",
|
||||
"predictedClass": "Clase predicha",
|
||||
"trueValue": "Valor verdadero",
|
||||
"trueClass": "Clase verdadera",
|
||||
"noColor": "Ninguno",
|
||||
"tooManyRows": "El conjunto de datos proporcionado es mayor de lo que admite este gráfico."
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Clase: {0}",
|
||||
"sortBy": "Ordenar por",
|
||||
"noData": "Sin datos",
|
||||
"absoluteGlobal": "Global absoluto",
|
||||
"absoluteLocal": "Local absoluto",
|
||||
"calculatingExplanation": "Calculando explicación"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Debe ser un valor numérico.",
|
||||
"integerError": "Debe ser un entero.",
|
||||
"prediction": "Predicción",
|
||||
"predictedProbability": "Probabilidad predicha",
|
||||
"predictionLabel": "Predicción: {0}",
|
||||
"probabilityLabel": "Probabilidad: {0}",
|
||||
"noModelError": "Proporcione un modelo de en el que se hayan ejecutado operaciones para explorar las predicciones en los trazados de ICE.",
|
||||
"featurePickerLabel": "Característica:",
|
||||
"minimumInputLabel": "Mínimo:",
|
||||
"maximumInputLabel": "Máximo:",
|
||||
"stepInputLabel": "Pasos:",
|
||||
"loadingMessage": "Cargando datos...",
|
||||
"submitPrompt": "Envíe un rango para ver un trazado de ICE.",
|
||||
"topLevelErrorMessage": "Error en el parámetro",
|
||||
"errorPrefix": "Se detectó un error: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Cargando...",
|
||||
"perturbationLabel": "Perturbación:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Valor predicho: {0}",
|
||||
"predictedClassLabel": "Clase predicha: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Sin agrupar",
|
||||
"groupPredicted": "Y predicho",
|
||||
"groupTrue": "Y verdadero",
|
||||
"groupBy": "Agrupar por"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Tipo de gráfico:",
|
||||
"violinText": "Violín",
|
||||
"barText": "Barra",
|
||||
"boxText": "Cuadro",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "La importancia de la característica global se calcula mediante el promedio del valor absoluto de la importancia de la característica de todos los puntos (normalización L1). ",
|
||||
"multiclassImportanceAddendum": "Para el cálculo de la importancia de una característica para todas las clases se incluyen todos los puntos y no se utiliza ninguna ponderación diferencial. Por tanto, una característica con una gran importancia negativa para muchos puntos con una predicción de que no será de \"clase A\" aumentará considerablemente la importancia de la \"clase A\" de la característica."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Igual que",
|
||||
"greaterThanComparison": "Mayor que",
|
||||
"greaterThanEqualToComparison": "Es mayor o igual que",
|
||||
"lessThanComparison": "Menor que",
|
||||
"lessThanEqualToComparison": "Es menor o igual que",
|
||||
"inTheRangeOf": "En el intervalo:",
|
||||
"categoricalIncludeValues": "Valores incluidos:",
|
||||
"numericValue": "Valor",
|
||||
"numericalComparison": "Operación",
|
||||
"minimum": "Mínimo",
|
||||
"maximum": "Máximo",
|
||||
"min": "Mín.: {0}",
|
||||
"max": "Máx.: {0}",
|
||||
"uniqueValues": "número de valores únicos: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Error de regresión",
|
||||
"error": "Error",
|
||||
"classificationOutcome": "Resultado de la clasificación",
|
||||
"truePositive": "Verdadero positivo",
|
||||
"trueNegative": "Verdadero negativo",
|
||||
"falsePositive": "Falso positivo",
|
||||
"falseNegative": "Falso negativo",
|
||||
"dataset": "Conjunto de datos",
|
||||
"predictedProbabilities": "Probabilidades de predicción",
|
||||
"none": "Recuento"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Cerrar",
|
||||
"defaultCustomRootName": "Copia de la fila {0}",
|
||||
"filterFeaturePlaceholder": "Buscar características"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Cohorte",
|
||||
"defaultLabel": "Todos los datos"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Explore las principales características importantes K que afectan a las predicciones generales de su modelo (explicación global). Puede usar el control deslizante para mostrar los valores de la importancia de las características en orden descendente, así como seleccionar hasta tres cohortes para consultar los valores de importancia de las características en paralelo. Asimismo, puede hacer clic en cualquiera de las barras de características del gráfico para consultar el modo en el que los valores de la característica seleccionada afectan al modelo de predicción.",
|
||||
"topAtoB": "Principales características: {0}-{1}",
|
||||
"datasetCohorts": "Cohortes de conjunto de datos",
|
||||
"legendHelpText": "Para activar o desactivar los cohortes en el trazado, haga clic en los elementos de la leyenda.",
|
||||
"sortBy": "Ordenar por",
|
||||
"viewDependencePlotFor": "Ver trazado de dependencias para:",
|
||||
"datasetCohortSelector": "Seleccionar cohorte de un conjunto de datos",
|
||||
"aggregateFeatureImportance": "Importancia de características agregadas",
|
||||
"missingParameters": "Esta pestaña requiere que se proporcione el parámetro de importancia de la característica local.",
|
||||
"weightOptions": "Niveles de importancia de las clases",
|
||||
"dependencePlotTitle": "Trazados de dependencia",
|
||||
"dependencePlotHelperText": "Este trazado de dependencia muestra la relación entre el valor de una característica y la importancia correspondiente de la característica en un cohorte.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Seleccionar característica",
|
||||
"datasetRequired": "Los trazados de dependencia requieren el conjunto de datos de evaluación y la matriz de importancia de características locales."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Estadísticas de datos",
|
||||
"datapoints": "{0} puntos de datos",
|
||||
"features": "{0} características",
|
||||
"filters": "{0} filtros",
|
||||
"binaryClassifier": "Clasificador de elementos binarios",
|
||||
"regressor": "Regresor",
|
||||
"multiclassClassifier": "Clasificador multiclase",
|
||||
"datasetCohorts": "Cohortes de conjunto de datos",
|
||||
"editCohort": "Editar cohorte",
|
||||
"duplicateCohort": "Duplicar cohorte",
|
||||
"addCohort": "Agregar cohorte",
|
||||
"copy": " copia"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Evalúe el rendimiento de su modelo. Para ello, explore la distribución de los valores de predicción y los de las métricas de rendimiento del modelo. Asimismo, si quiere investigar el modelo con todavía mayor profundidad, consulte un análisis comparativo de su rendimiento en varios cohortes o subgrupos de su conjunto de datos. Seleccione los filtros pertinentes, junto con los valores X e Y, para segmentar dimensiones distintas. También puede seleccionar el icono de engranaje para cambiar el tipo de gráfico.",
|
||||
"modelStatistics": "Estadísticas del modelo",
|
||||
"cohortPickerLabel": "Seleccionar cohorte de un conjunto de datos para explorar",
|
||||
"missingParameters": "Esta pestaña requiere que se proporcione la matriz de valores de predicción del modelo.",
|
||||
"missingTrueY": "Las estadísticas de rendimiento del modelo requieren que se proporcionen los resultados verdaderos además de los previstos."
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Valor Y",
|
||||
"numberOfDatapoints": "Número de puntos de datos",
|
||||
"xValue": "Valor X",
|
||||
"rowIndex": "Índice de fila",
|
||||
"featureImportance": "Importancia de característica",
|
||||
"countTooltipPrefix": "Recuento: {0}",
|
||||
"count": "Recuento",
|
||||
"featurePrefix": "Característica",
|
||||
"importancePrefix": "Importancia",
|
||||
"cohort": "Cohorte",
|
||||
"howToRead": "Cómo leer este gráfico"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Explore las estadísticas del conjunto de datos seleccionando distintos filtros a lo largo de los ejes X, Y y de color para segmentar los datos en distintas dimensiones. Cree cohortes de conjunto de datos más arriba para analizar las estadísticas de conjunto de datos con filtros como los de resultados previstos, características de conjunto de datos y grupos de errores. Use el icono de engranaje situado en la esquina superior derecha del gráfico para cambiar los tipos de gráfico.",
|
||||
"colorValue": "Valor de color",
|
||||
"individualDatapoints": "Puntos de datos individuales",
|
||||
"aggregatePlots": "Trazados agregados",
|
||||
"chartType": "Tipo de gráfico",
|
||||
"missingParameters": "Esta pestaña requiere que se proporcione un conjunto de datos de evaluación.",
|
||||
"noColor": "Ninguno"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Importancia de la característica:",
|
||||
"placeholder": "Haga clic en una característica del gráfico de barras para mostrar su trazado de dependencias."
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Puede seleccionar un punto de datos haciendo clic en el trazado disperso para ver sus valores de importancia de característica local (explicación local) y el trazado de expectativa condicional individual (ICE) a continuación. Cree un punto de datos \"What If\" hipotético con el panel de la derecha para perturbar las características de un punto de datos conocido. Los valores de importancia de característica se basan en varias aproximaciones y no son la causa de las predicciones. Sin la solidez matemática estricta de la inferencia causal, no se recomienda a los usuarios que tomen decisiones reales basadas en esta herramienta.",
|
||||
"panelPlaceholder": "Se necesita un modelo para efectuar predicciones para nuevos puntos de datos.",
|
||||
"cohortPickerLabel": "Seleccionar cohorte de un conjunto de datos para explorar",
|
||||
"scatterLegendText": "Para activar o desactivar los puntos de datos en el trazado, haga clic en los elementos de la leyenda.",
|
||||
"realPoint": "Puntos de datos reales",
|
||||
"noneSelectedYet": "Todavía no se ha seleccionado ninguno.",
|
||||
"whatIfDatapoints": "Puntos de datos hipotéticos",
|
||||
"noneCreatedYet": "Todavía no se ha creado ninguno.",
|
||||
"showLabel": "Mostrar:",
|
||||
"featureImportancePlot": "Trazado de la importancia de la característica",
|
||||
"icePlot": "Trazado de expectativa condicional individual (ICE)",
|
||||
"featureImportanceLackingParameters": "Especifique la importancia de las características locales para ver cómo afecta cada característica a las predicciones individuales.",
|
||||
"featureImportanceGetStartedText": "Seleccione un punto para consultar la importancia de la característica.",
|
||||
"iceLackingParameters": "Los trazados de ICE requieren un modelo de operaciones para poder realizar predicciones para los puntos de datos hipotéticos.",
|
||||
"IceGetStartedText": "Seleccione un punto o cree uno hipotético para consultar los trazados de ICE.",
|
||||
"whatIfDatapoint": "Punto de datos hipotético",
|
||||
"whatIfHelpText": "Seleccione un punto en el trazado o indique manualmente un índice de puntos de datos conocido para alterarlo y guardarlo como hipotético.",
|
||||
"indexLabel": "Índice de datos para alterar",
|
||||
"rowLabel": "Fila {0}",
|
||||
"whatIfNameLabel": "Nombre de punto de datos hipotético",
|
||||
"featureValues": "Valores de característica",
|
||||
"predictedClass": "Clase prevista: ",
|
||||
"predictedValue": "Valor previsto: ",
|
||||
"probability": "Probabilidad: ",
|
||||
"trueClass": "Clase verdadera: ",
|
||||
"trueValue": "Valor verdadero: ",
|
||||
"trueValue.comment": "prefijo para la etiqueta real para la regresión",
|
||||
"newPredictedClass": "Nueva clase prevista: ",
|
||||
"newPredictedValue": "Nuevo valor previsto: ",
|
||||
"newProbability": "Nueva probabilidad: ",
|
||||
"saveAsNewPoint": "Guardar como nuevo punto",
|
||||
"saveChanges": "Guardar cambios",
|
||||
"loading": "Cargando...",
|
||||
"classLabel": "Clase: {0}",
|
||||
"minLabel": "Mín.",
|
||||
"maxLabel": "Máx.",
|
||||
"stepsLabel": "Pasos",
|
||||
"disclaimer": "Declinación de responsabilidades: Estas explicaciones se basan en una gran variedad de enfoques y no son la \"causa\" de las predicciones. Sin una solidez matemática estricta de la inferencia causal, recomendamos a los usuarios que no utilicen esta herramienta para tomar decisiones que afecten a la vida real.",
|
||||
"missingParameters": "Esta pestaña requiere que se proporcione un conjunto de datos de evaluación.",
|
||||
"selectionLimit": "Máximo de 3 puntos seleccionados",
|
||||
"classPickerLabel": "Clase",
|
||||
"tooltipTitleMany": "Principales {0} clases previstas",
|
||||
"whatIfTooltipTitle": "Clases de hipótesis previstas",
|
||||
"tooltipTitleFew": "Clases previstas",
|
||||
"probabilityLabel": "Probabilidad",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "El valor debe ser numérico.",
|
||||
"icePlotHelperText": "Los trazados de ICE muestran cómo cambian los valores de predicción del punto de datos seleccionado en un rango de valores de característica entre un valor mínimo y uno máximo."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Seleccionar filtro",
|
||||
"TreatAsCategorical": "Tratar como valor categórico",
|
||||
"addFilter": "Agregar filtro",
|
||||
"addedFilters": "Filtros agregados",
|
||||
"noAddedFilters": "Todavía no se ha agregado ningún filtro.",
|
||||
"defaultFilterState": "Seleccione un filtro para agregar parámetros al cohorte de su conjunto de datos.",
|
||||
"cohortNameLabel": "Nombre de cohorte de conjunto de datos",
|
||||
"cohortNamePlaceholder": "Asignar nombre a cohorte",
|
||||
"save": "Guardar",
|
||||
"delete": "Eliminar",
|
||||
"cancel": "Cancelar",
|
||||
"cohortNameError": "Falta el nombre de cohorte.",
|
||||
"placeholderName": "Cohorte {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Seleccionar",
|
||||
"ditherLabel": "Debe interpolar",
|
||||
"selectFilter": "Seleccionar valor del eje",
|
||||
"selectFeature": "Seleccionar característica",
|
||||
"binLabel": "Aplicar discretización a los datos",
|
||||
"TreatAsCategorical": "Tratar como valor categórico",
|
||||
"numOfBins": "Número de discretizaciones",
|
||||
"groupByCohort": "Agrupar por cohorte",
|
||||
"selectClass": "Seleccionar clase",
|
||||
"countHelperText": "Histograma del número de puntos"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Probabilidad prevista",
|
||||
"predictedY": "Eje Y previsto",
|
||||
"evalData": "Conjunto de datos de evaluación",
|
||||
"localFeatureImportance": "Importancia de la característica local",
|
||||
"inconsistentDimensions": "Las dimensiones son incoherentes. Dimensiones de {0}: {1} dimensiones esperadas: {2}.",
|
||||
"notNonEmpty": "La entrada de {0} no es una matriz no vacía.",
|
||||
"varyingLength": "Las dimensiones son incoherentes. {0} tiene elementos de longitud variable.",
|
||||
"notArray": "{0} no es una matriz. Se esperaba una matriz de dimensión {1}.",
|
||||
"errorHeader": "Algunos parámetros de entrada no son coherentes y no se usarán: ",
|
||||
"datasizeWarning": "El conjunto de datos de evaluación es demasiado grande para mostrarse correctamente en algunos gráficos. Agregue filtros para reducir el tamaño de la cohorte. ",
|
||||
"datasizeError": "La cohorte seleccionada es demasiado grande. Agregue filtros para reducir su tamaño.",
|
||||
"addFilters": "Agregar filtros"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " incluye {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} y {1} más"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "Error cuadrático medio: {0}",
|
||||
"rSquared": "Coeficiente de determinación: {0}",
|
||||
"meanPrediction": "Predicción media: {0}",
|
||||
"accuracy": "Precisión: {0}",
|
||||
"precision": "Precisión: {0}",
|
||||
"recall": "Recuperación: {0}",
|
||||
"fpr": "Tasa de falsos positivos: {0}",
|
||||
"fnr": "Tasa de falsos negativos: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Explore las características k más importantes que afectan a las predicciones generales de modelos. Use el control deslizante para mostrar la importancia de las características en orden descendente."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "¿Qué significan estas explicaciones?",
|
||||
"clickHere": "Más información",
|
||||
"shapTitle": "Valores de Shapley",
|
||||
"shapDescription": "Este explicador usa SHAP, que es un enfoque teórico de juego para explicar modelos en los que la importancia de los conjuntos de características se mide \"ocultando\" en el modelo las características en cuestión a través de la marginalización. Haga clic en el vínculo a continuación para obtener más información.",
|
||||
"limeTitle": "LIME (explicaciones independientes del modelo interpretables locales)",
|
||||
"limeDescription": "Este explicador utiliza LIME, que proporciona una aproximación lineal del modelo. Para obtener una explicación, debe hacerse lo siguiente: perturbar la instancia, obtener predicciones de modelo y usar las predicciones como etiquetas para obtener un modelo lineal disperso que sea fiel localmente. Las ponderaciones de este modelo lineal se usan como \"importancias de características\". Haga clic en el vínculo a continuación para obtener más información.",
|
||||
"mimicTitle": "Imitación (explicaciones subrogadas globales)",
|
||||
"mimicDescription": "Este explicador se basa en la idea de los modelos subrogados globales de entrenamiento para imitar los modelos de una caja negra. Un modelo subrogado global es un modelo interpretable de forma intrínseca que se ha entrenado para aproximar las predicciones de cualquier modelo de una caja negra de la forma más precisa posible. Los valores de importancia de característica son valores de importancia de característica basados en modelos del modelo subrogado subyacente (LightGBM, regresión lineal, descenso de gradiente estocástico o árbol de decisión).",
|
||||
"pfiTitle": "Importancia de la característica de permutación (PFI)",
|
||||
"pfiDescription": "Este explicador mezcla aleatoriamente datos de característica en característica para todo el conjunto de datos y calcula el grado de cambio de una métrica de rendimiento en cuestión (métricas de rendimiento predeterminadas: F1 para clasificación binaria, puntuación F1 con micropromedio para clasificación multiclase y error absoluto medio para regresión). Cuanto mayor sea el cambio, más importante será la característica. Este explicador solo puede explicar el comportamiento general del modelo subyacente, por lo que no explica predicciones individuales. El valor de importancia de una característica representa el valor delta del rendimiento del modelo perturbando la característica correspondiente."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Sélectionner un point pour voir son explication locale",
|
||||
"defaultClassNames": "Classe {0}",
|
||||
"defaultFeatureNames": "{0} de la caractéristique",
|
||||
"absoluteAverage": "Moyenne de la valeur absolue",
|
||||
"predictedClass": "Classe prédite",
|
||||
"datasetExplorer": "Explorateur de jeux de données",
|
||||
"dataExploration": "Exploration de jeu de données",
|
||||
"aggregateFeatureImportance": "Agréger l'importance des caractéristiques",
|
||||
"globalImportance": "Importance globale",
|
||||
"explanationExploration": "Exploration de l'explication",
|
||||
"individualAndWhatIf": "Importance d'une caractéristique individuelle et simulation",
|
||||
"summaryImportance": "Importance combinée",
|
||||
"featureImportance": "Importance de la caractéristique",
|
||||
"featureImportanceOf": "Importance de la caractéristique {0}",
|
||||
"perturbationExploration": "Exploration de la perturbation",
|
||||
"localFeatureImportance": "Importance de la caractéristique locale",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Effacer la sélection",
|
||||
"feature": "Fonctionnalité :",
|
||||
"intercept": "Intercepter",
|
||||
"modelPerformance": "Performances du modèle",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Données : {0}",
|
||||
"importanceLabel": "Importance : {0}",
|
||||
"predictedY": "Y prédit",
|
||||
"index": "Index",
|
||||
"dataGroupLabel": "Données",
|
||||
"output": "Sortie",
|
||||
"probabilityLabel": "Probabilité : {0}",
|
||||
"trueY": "Y réel",
|
||||
"class": "classe : ",
|
||||
"xValue": "Valeur X :",
|
||||
"yValue": "Valeur Y :",
|
||||
"colorValue": "Couleur :",
|
||||
"count": "Nombre"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Pondération interclasse :",
|
||||
"info": "Informations sur le calcul interclasse",
|
||||
"overviewInfo": "Les modèles multiclasses génèrent, pour chaque classe, un vecteur d'importance de caractéristique indépendant qui montre les caractéristiques affectant le degré de probabilité d'une classe. Vous pouvez sélectionner la façon dont les poids des vecteurs d'importance de caractéristique par classe sont combinés en une seule valeur :",
|
||||
"absoluteValInfo": "Moyenne de la valeur absolue : montre la somme des valeurs d'importance de la caractéristique dans toutes les classes possibles, divisée par le nombre de classes",
|
||||
"predictedClassInfo": "Classe prédite : montre la valeur d'importance de la caractéristique pour la classe prédite d'un point donné",
|
||||
"enumeratedClassInfo": "Noms de classe énumérés : montre uniquement les valeurs d'importance de la caractéristique pour la classe spécifiée sur tous les points de données.",
|
||||
"close": "Fermer",
|
||||
"crossClassWeights": "Pondérations multiclasses"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Valeur de caractéristique mise à l'échelle",
|
||||
"low": "Basse",
|
||||
"high": "Élevée",
|
||||
"featureLabel": "Caractéristique : {0}",
|
||||
"valueLabel": "Valeur de caractéristique : {0}",
|
||||
"importanceLabel": "Importance : {0}",
|
||||
"predictedClassTooltip": "Classe prédite : {0}",
|
||||
"trueClassTooltip": "Classe réelle : {0}",
|
||||
"predictedOutputTooltip": "Sortie prédite : {0}",
|
||||
"trueOutputTooltip": "Sortie réelle : {0}",
|
||||
"topKFeatures": "Top K des caractéristiques :",
|
||||
"topKInfo": "Mode de calcul du top k",
|
||||
"predictedValue": "Valeur prédite",
|
||||
"predictedClass": "Classe prédite",
|
||||
"trueValue": "Valeur vraie",
|
||||
"trueClass": "Classe réelle",
|
||||
"noColor": "Aucun",
|
||||
"tooManyRows": "Ce graphique ne peut pas prendre en charge le jeu de données fourni en raison de sa taille trop importante"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Classe : {0}",
|
||||
"sortBy": "Trier par",
|
||||
"noData": "Aucune donnée",
|
||||
"absoluteGlobal": "Valeur absolue globale",
|
||||
"absoluteLocal": "Valeur absolue locale",
|
||||
"calculatingExplanation": "Calcul de l'explication"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Doit être numérique",
|
||||
"integerError": "Doit être un entier",
|
||||
"prediction": "Prédiction",
|
||||
"predictedProbability": "Probabilité prédite",
|
||||
"predictionLabel": "Prédiction : {0}",
|
||||
"probabilityLabel": "Probabilité : {0}",
|
||||
"noModelError": "Indiquez un modèle opérationnalisé pour explorer les prédictions dans les tracés ICE.",
|
||||
"featurePickerLabel": "Fonctionnalité :",
|
||||
"minimumInputLabel": "Minimum :",
|
||||
"maximumInputLabel": "Maximum :",
|
||||
"stepInputLabel": "Étapes :",
|
||||
"loadingMessage": "Chargement des données...",
|
||||
"submitPrompt": "Envoyer une plage pour voir un tracé ICE",
|
||||
"topLevelErrorMessage": "Erreur dans le paramètre",
|
||||
"errorPrefix": "Erreur rencontrée : {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Chargement...",
|
||||
"perturbationLabel": "Perturbation :"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Valeur prédite : {0}",
|
||||
"predictedClassLabel": "Classe prédite : {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Aucun regroupement",
|
||||
"groupPredicted": "Y prédit",
|
||||
"groupTrue": "Y réel",
|
||||
"groupBy": "Regrouper par"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Type de graphique :",
|
||||
"violinText": "Violon",
|
||||
"barText": "Barre",
|
||||
"boxText": "Zone",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "L'importance d'une caractéristique globale est déterminée en calculant la moyenne de la valeur absolue de l'importance de la caractéristique sur tous les points (normalisation L1). ",
|
||||
"multiclassImportanceAddendum": "Tous les points sont inclus dans le calcul de l'importance d'une caractéristique pour toutes les classes, sans aucune pondération différentielle. Ainsi, une caractéristique avec une grande importance négative pour de nombreux points qui, par prédiction, ne sont pas de « Classe A », augmente considérablement l'importance de cette classe."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Égal à",
|
||||
"greaterThanComparison": "Supérieur à",
|
||||
"greaterThanEqualToComparison": "Supérieur ou égal à",
|
||||
"lessThanComparison": "Inférieur à",
|
||||
"lessThanEqualToComparison": "Inférieur ou égal à",
|
||||
"inTheRangeOf": "Dans la plage",
|
||||
"categoricalIncludeValues": "Valeurs incluses :",
|
||||
"numericValue": "Valeur",
|
||||
"numericalComparison": "Opération",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Maximum",
|
||||
"min": "Min : {0}",
|
||||
"max": "Max : {0}",
|
||||
"uniqueValues": "Nb de valeurs uniques : {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Erreur de régression",
|
||||
"error": "Erreur",
|
||||
"classificationOutcome": "Résultat de la classification",
|
||||
"truePositive": "Vrai positif",
|
||||
"trueNegative": "Vrai négatif",
|
||||
"falsePositive": "Faux positif",
|
||||
"falseNegative": "Faux négatif",
|
||||
"dataset": "Jeu de données",
|
||||
"predictedProbabilities": "Probabilités de prédiction",
|
||||
"none": "Nombre"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Fermer",
|
||||
"defaultCustomRootName": "Copie de la ligne {0}",
|
||||
"filterFeaturePlaceholder": "Rechercher des caractéristiques"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Cohorte",
|
||||
"defaultLabel": "Toutes les données"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Explorez les k principales caractéristiques importantes qui impactent vos prédictions de modèle générales (c'est-à-dire, l'explication globale). Utilisez le curseur pour afficher les valeurs d'importance de caractéristique par ordre décroissant. Sélectionnez jusqu'à trois cohortes pour voir leurs valeurs d'importance de caractéristique côte à côte. Cliquez sur les barres de caractéristique dans le graphe pour voir comment les valeurs de la caractéristique sélectionnée impactent le modèle de prédiction.",
|
||||
"topAtoB": "{0}-{1} principales caractéristiques",
|
||||
"datasetCohorts": "Cohortes de jeu de données",
|
||||
"legendHelpText": "Activez ou désactivez les cohortes dans le tracé en cliquant sur les éléments de la légende.",
|
||||
"sortBy": "Trier par",
|
||||
"viewDependencePlotFor": "Voir le tracé des dépendances pour :",
|
||||
"datasetCohortSelector": "Sélectionner une cohorte de jeu de données",
|
||||
"aggregateFeatureImportance": "Agréger l'importance des caractéristiques",
|
||||
"missingParameters": "Cet onglet nécessite la spécification du paramètre d'importance de la caractéristique locale.",
|
||||
"weightOptions": "Pondérations de l'importance de classe",
|
||||
"dependencePlotTitle": "Tracés de dépendance",
|
||||
"dependencePlotHelperText": "Ce tracé de dépendance montre la relation entre la valeur d'une caractéristique et son importance dans une cohorte.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Sélectionner une caractéristique",
|
||||
"datasetRequired": "Les tracés de dépendance nécessitent le jeu de données d'évaluation et le tableau des importances de caractéristique locale."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Statistiques des données",
|
||||
"datapoints": "{0} points de données",
|
||||
"features": "{0} caractéristiques",
|
||||
"filters": "{0} filtres",
|
||||
"binaryClassifier": "Classifieur binaire",
|
||||
"regressor": "Régresseur",
|
||||
"multiclassClassifier": "Classifieur multiclasse",
|
||||
"datasetCohorts": "Cohortes de jeu de données",
|
||||
"editCohort": "Modifier la cohorte",
|
||||
"duplicateCohort": "Dupliquer la cohorte",
|
||||
"addCohort": "Ajouter une cohorte",
|
||||
"copy": " copie"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Évaluez les performances de votre modèle en explorant la distribution de vos valeurs de prédiction et les valeurs de vos métriques de performances de modèle. Vous pouvez examiner plus en détail votre modèle en étudiant une analyse comparative de ses performances sur différentes cohortes ou sous-groupes de votre jeu de données. Sélectionnez des filtres avec une valeur X et une valeur Y pour les croiser sur différentes dimensions. Sélectionnez l'engrenage dans le graphique pour changer le type de graphique.",
|
||||
"modelStatistics": "Statistiques du modèle",
|
||||
"cohortPickerLabel": "Sélectionner une cohorte de jeu de données à explorer",
|
||||
"missingParameters": "Cet onglet nécessite la spécification du tableau des valeurs prédites du modèle.",
|
||||
"missingTrueY": "Les statistiques de performance du modèle nécessitent la spécification des résultats réels en plus des résultats prédits"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Valeur Y",
|
||||
"numberOfDatapoints": "Nombre de points de données",
|
||||
"xValue": "Valeur X",
|
||||
"rowIndex": "Index de ligne",
|
||||
"featureImportance": "Importance de la caractéristique",
|
||||
"countTooltipPrefix": "Nombre : {0}",
|
||||
"count": "Nombre",
|
||||
"featurePrefix": "Caractéristique",
|
||||
"importancePrefix": "Importance",
|
||||
"cohort": "Cohorte",
|
||||
"howToRead": "Comment lire ce graphique"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Explorez les statistiques de votre jeu de données en sélectionnant différents filtres sur les axes X, Y et de couleur pour découper vos données selon différentes dimensions. Créez en plus des cohortes de jeux de données pour analyser les statistiques de jeu de données avec des filtres, comme les résultats prédits, les caractéristiques de jeu de données et les groupes d'erreurs. Utilisez l'icône d'engrenage en haut à droite du graphe pour changer les types de graphe.",
|
||||
"colorValue": "Valeur de couleur",
|
||||
"individualDatapoints": "Points de données individuels",
|
||||
"aggregatePlots": "Agréger les tracés",
|
||||
"chartType": "Type de graphique",
|
||||
"missingParameters": "Cet onglet nécessite la spécification d'un jeu de données d'évaluation.",
|
||||
"noColor": "Aucun"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Importance de la caractéristique",
|
||||
"placeholder": "Cliquer sur une caractéristique dans l'histogramme ci-dessus pour afficher le tracé de ses dépendances"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Vous pouvez sélectionner un point de données en cliquant sur le nuage de points pour voir ses valeurs d'importance de caractéristique locale (explication locale) et le tracé d'espérance conditionnelle individuelle (ICE) ci-dessous. Créez un point de données « Et si » hypothétique en utilisant le panneau de droite pour perturber les caractéristiques d'un point de données connu. Les valeurs d'importance de caractéristique sont basées sur de nombreuses approximations et ne sont pas la « cause » des prédictions. Sans une robustesse mathématique stricte de l'inférence causale, nous ne conseillons pas aux utilisateurs de prendre des décisions concrètes basées sur cet outil.",
|
||||
"panelPlaceholder": "Un modèle est nécessaire pour effectuer des prédictions pour les nouveaux points de données.",
|
||||
"cohortPickerLabel": "Sélectionner une cohorte de jeu de données à explorer",
|
||||
"scatterLegendText": "Activez ou désactivez les points de données dans le tracé en cliquant sur les éléments de la légende.",
|
||||
"realPoint": "Points de données réels",
|
||||
"noneSelectedYet": "Aucun sélectionné pour l'instant",
|
||||
"whatIfDatapoints": "Points de données de simulation",
|
||||
"noneCreatedYet": "Aucun créé pour l'instant",
|
||||
"showLabel": "Afficher :",
|
||||
"featureImportancePlot": "Tracé de l'importance des caractéristiques",
|
||||
"icePlot": "Tracé ICE (Individual Conditional Expectation)",
|
||||
"featureImportanceLackingParameters": "Spécifiez l'importance des caractéristiques locales pour voir comment chaque caractéristique impacte les prédictions individuelles.",
|
||||
"featureImportanceGetStartedText": "Sélectionner un point pour voir l'importance de la caractéristique",
|
||||
"iceLackingParameters": "Les tracés ICE nécessitent un modèle opérationnalisé afin de faire des prédictions pour des points de données hypothétiques.",
|
||||
"IceGetStartedText": "Sélectionner un point ou créer un point de simulation pour voir les tracés ICE",
|
||||
"whatIfDatapoint": "Point de données de simulation",
|
||||
"whatIfHelpText": "Sélectionnez un point sur le tracé ou entrez manuellement un index de point de données connu afin de créer une perturbation et l'enregistrer comme nouveau point de simulation.",
|
||||
"indexLabel": "Index des données à perturber",
|
||||
"rowLabel": "Ligne {0}",
|
||||
"whatIfNameLabel": "Nom de point de données de simulation",
|
||||
"featureValues": "Valeurs de caractéristique",
|
||||
"predictedClass": "Classe prédite : ",
|
||||
"predictedValue": "Valeur prédite : ",
|
||||
"probability": "Probabilité : ",
|
||||
"trueClass": "Classe réelle : ",
|
||||
"trueValue": "Valeur réelle : ",
|
||||
"trueValue.comment": "préfixe de l'étiquette réelle pour la régression",
|
||||
"newPredictedClass": "Nouvelle classe prédite : ",
|
||||
"newPredictedValue": "Nouvelle valeur prédite : ",
|
||||
"newProbability": "Nouvelle probabilité : ",
|
||||
"saveAsNewPoint": "Enregistrer comme nouveau point",
|
||||
"saveChanges": "Enregistrer les changements",
|
||||
"loading": "Chargement...",
|
||||
"classLabel": "Classe : {0}",
|
||||
"minLabel": "Min",
|
||||
"maxLabel": "Max",
|
||||
"stepsLabel": "Étapes",
|
||||
"disclaimer": "Exclusion : Ces explications sont basées sur de nombreuses approximations et ne sont pas la « cause » des prédictions. Sans une robustesse mathématique stricte de l'inférence causale, nous ne conseillons pas aux utilisateurs de prendre des décisions réelles basées sur cet outil.",
|
||||
"missingParameters": "Cet onglet nécessite la spécification d'un jeu de données d'évaluation.",
|
||||
"selectionLimit": "3 points maximum sélectionnés",
|
||||
"classPickerLabel": "Classe",
|
||||
"tooltipTitleMany": "Top {0} des classes prédites",
|
||||
"whatIfTooltipTitle": "Classes prédites des scénarios",
|
||||
"tooltipTitleFew": "Classes prédites",
|
||||
"probabilityLabel": "Probabilité",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "La valeur doit être numérique",
|
||||
"icePlotHelperText": "Les tracés ICE montrent comment les valeurs de prédiction du point de données sélectionné varient pour plusieurs valeurs de caractéristique entre la valeur minimale et la valeur maximale."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Sélectionner un filtre",
|
||||
"TreatAsCategorical": "Considérer comme catégorique",
|
||||
"addFilter": "Ajouter un filtre",
|
||||
"addedFilters": "Filtres ajoutés",
|
||||
"noAddedFilters": "Aucun filtre ajouté pour le moment",
|
||||
"defaultFilterState": "Sélectionnez un filtre pour ajouter des paramètres à votre cohorte de jeu de données.",
|
||||
"cohortNameLabel": "Nom de cohorte de jeu de données",
|
||||
"cohortNamePlaceholder": "Nommer votre cohorte",
|
||||
"save": "Enregistrer",
|
||||
"delete": "Supprimer",
|
||||
"cancel": "Annuler",
|
||||
"cohortNameError": "Nom de cohorte manquant",
|
||||
"placeholderName": "Cohorte {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Sélectionner",
|
||||
"ditherLabel": "Doit tramer",
|
||||
"selectFilter": "Sélectionner votre valeur d'axe",
|
||||
"selectFeature": "Sélectionner une caractéristique",
|
||||
"binLabel": "Appliquer le binning aux données",
|
||||
"TreatAsCategorical": "Considérer comme catégorique",
|
||||
"numOfBins": "Nombre de compartiments",
|
||||
"groupByCohort": "Regrouper par cohorte",
|
||||
"selectClass": "Sélectionner une classe",
|
||||
"countHelperText": "Histogramme du nombre de points"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Probabilité prédite",
|
||||
"predictedY": "Valeur Y prédite",
|
||||
"evalData": "Jeu de données d'évaluation",
|
||||
"localFeatureImportance": "Importance de la caractéristique locale",
|
||||
"inconsistentDimensions": "Dimensions incohérentes. {0} a les dimensions {1}, {2} était attendu",
|
||||
"notNonEmpty": "L'entrée {0} n'est pas un tableau non vide",
|
||||
"varyingLength": "Dimensions incohérentes. {0} a des éléments de longueur variable",
|
||||
"notArray": "{0} n'est pas un tableau. Tableau de dimension {1} attendu",
|
||||
"errorHeader": "Certains paramètres d'entrée étaient incohérents et ne sont pas utilisés : ",
|
||||
"datasizeWarning": "Le jeu de données d'évaluation est trop grand pour être affiché correctement dans certains graphiques, ajoutez des filtres pour réduire la taille de la cohorte. ",
|
||||
"datasizeError": "La cohorte sélectionnée est trop grande, ajoutez des filtres pour en réduire la taille.",
|
||||
"addFilters": "Ajouter des filtres"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " inclut {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} et {1} autres"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE : {0}",
|
||||
"rSquared": "Coefficient de détermination : {0}",
|
||||
"meanPrediction": "Prédiction moyenne {0}",
|
||||
"accuracy": "Justesse : {0}",
|
||||
"precision": "Précision : {0}",
|
||||
"recall": "Rappel : {0}",
|
||||
"fpr": "FPR : {0}",
|
||||
"fnr": "FNR : {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Explorez le top k des caractéristiques les plus importantes qui impactent vos prédictions de modèle globales. Utilisez le curseur pour afficher les caractéristiques par importance décroissante."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Que signifient ces explications ?",
|
||||
"clickHere": "En savoir plus",
|
||||
"shapTitle": "Valeurs Shapley",
|
||||
"shapDescription": "Cet explicateur utilise SHAP, qui est une approche utilisant la théorie des jeux pour expliquer les modèles, où l'importance d'ensembles de caractéristiques est mesurée en « masquant » ces caractéristiques dans le modèle par le biais de la marginalisation. Cliquez sur le lien ci-dessous pour en savoir plus.",
|
||||
"limeTitle": "LIME (Local Interpretable Model-Agnostic Explanations)",
|
||||
"limeDescription": "Cet explicateur utilise LIME, qui fournit une approximation linéaire du modèle. Pour obtenir une explication, nous effectuons ce qui suit : perturber l'instance, obtenir des prédictions de modèle et utiliser ces prédictions comme des étiquettes pour apprendre un modèle linéaire creux qui est fidèle localement. Les pondérations de ce modèle linéaire sont utilisées sous forme d'« importances de caractéristique ». Cliquez sur le lien ci-dessous pour en savoir plus.",
|
||||
"mimicTitle": "Imiter (explications de substitution globales)",
|
||||
"mimicDescription": "Cet explicateur s'appuie sur l'idée d'entraîner des modèles de substitution globaux pour imiter les modèles de boîte noire. Un modèle de substitution global est un modèle interprétable de manière intrinsèque qui est entraîné pour se rapprocher des prédictions d'un modèle de boîte noire de manière aussi juste que possible. Les valeurs d'importance de caractéristique sont les valeurs basées sur votre modèle de substitution sous-jacent (LightGBM ou régression linéaire ou algorithme du gradient stochastique ou arbre de décision)",
|
||||
"pfiTitle": "Importance de caractéristique de permutation (PFI)",
|
||||
"pfiDescription": "Cet explicateur lit les données de manière aléatoire caractéristique par caractéristique pour l'ensemble du jeu de données et calcule la variation de la métrique de performances concernée (métriques de performances par défaut : F1 pour la classification binaire, score F1 avec moyenne micro pour la classification multiclasse et erreur absolue moyenne pour la régression). Plus la variation est grande, plus cette caractéristique est importante. Cet explicateur peut uniquement expliquer le comportement général du modèle sous-jacent, mais n'explique pas les prédictions individuelles. La valeur d'importance d'une caractéristique représente le delta des performances du modèle en perturbant cette caractéristique particulière."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "A helyi magyarázat megjelenítéséhez válassza ki a kívánt pontot",
|
||||
"defaultClassNames": "{0} osztály",
|
||||
"defaultFeatureNames": "{0} funkció",
|
||||
"absoluteAverage": "Abszolút érték átlaga",
|
||||
"predictedClass": "Előre jelzett osztály",
|
||||
"datasetExplorer": "Adathalmaz-böngésző",
|
||||
"dataExploration": "Adathalmaz-felderítés",
|
||||
"aggregateFeatureImportance": "Aggregált jellemzőfontosság",
|
||||
"globalImportance": "Globális fontosság",
|
||||
"explanationExploration": "Magyarázatvizsgálat",
|
||||
"individualAndWhatIf": "Egyéni jellemzőfontosság és lehetőségelemzés",
|
||||
"summaryImportance": "Összegzés fontossága",
|
||||
"featureImportance": "Jellemzőfontosság",
|
||||
"featureImportanceOf": "A(z) {0} jellemzőfontossága",
|
||||
"perturbationExploration": "Perturbációvizsgálat",
|
||||
"localFeatureImportance": "Helyi jellemzőfontosság",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Kijelölés törlése",
|
||||
"feature": "Funkciófrissítés:",
|
||||
"intercept": "Metszéspont",
|
||||
"modelPerformance": "Modell teljesítménye",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Adatok: {0}",
|
||||
"importanceLabel": "Fontosság: {0}",
|
||||
"predictedY": "Előre jelzett Y",
|
||||
"index": "Index",
|
||||
"dataGroupLabel": "Adatok",
|
||||
"output": "Kimenet",
|
||||
"probabilityLabel": "Valószínűség: {0}",
|
||||
"trueY": "Igaz Y",
|
||||
"class": "osztály: ",
|
||||
"xValue": "X érték",
|
||||
"yValue": "Y érték",
|
||||
"colorValue": "Szín:",
|
||||
"count": "Mennyiség"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Többosztályos súlyozás:",
|
||||
"info": "Az osztályok közötti számítással kapcsolatos információk",
|
||||
"overviewInfo": "A többosztályos modellek minden osztály esetében létrehoznak egy független jellemzőfontossági vektort. Az egyes osztályok jellemzőfontossági vektora azt mutatja, hogy mely jellemzők növelték vagy csökkentették egy adott osztály valószínűségét. Megadhatja, hogy az osztályonkénti jellemzőfontossági vektorok súlyai miként legyenek összegezve egyetlen értékbe:",
|
||||
"absoluteValInfo": "Az abszolút érték átlaga: a funkció fontosságának összege az összes lehetséges osztályra vonatkoztatva, osztva az osztályok számával",
|
||||
"predictedClassInfo": "Előre jelzett osztály: a jellemzőfontossági érték megjelenítése egy adott pont előre jelzett osztálya esetében",
|
||||
"enumeratedClassInfo": "Enumerált osztálynevek: csak a megadott osztály jellemzőfontossági értékeinek megjelenítése az összes adatpontra vonatkozóan.",
|
||||
"close": "Bezárás",
|
||||
"crossClassWeights": "Több osztályra vonatkozó súlyozás"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Skálázott funkció értéke",
|
||||
"low": "Kicsi",
|
||||
"high": "Magas",
|
||||
"featureLabel": "Funkció: {0}",
|
||||
"valueLabel": "Funkció értéke: {0}",
|
||||
"importanceLabel": "Fontosság: {0}",
|
||||
"predictedClassTooltip": "Előre jelzett osztály: {0}",
|
||||
"trueClassTooltip": "Igaz osztály: {0}",
|
||||
"predictedOutputTooltip": "Előre jelzett kimenet: {0}",
|
||||
"trueOutputTooltip": "Igaz kimenet: {0}",
|
||||
"topKFeatures": "Első K funkció:",
|
||||
"topKInfo": "A k legfontosabb jellemző kiszámítása",
|
||||
"predictedValue": "Előre jelzett érték",
|
||||
"predictedClass": "Előre jelzett osztály",
|
||||
"trueValue": "Igaz érték",
|
||||
"trueClass": "Igaz osztály",
|
||||
"noColor": "Nincs",
|
||||
"tooManyRows": "A megadott adathalmaz nagyobb a diagram által támogatott méretnél"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Osztály: {0}",
|
||||
"sortBy": "Rendezési szempont",
|
||||
"noData": "Nincs adat",
|
||||
"absoluteGlobal": "Abszolút globális",
|
||||
"absoluteLocal": "Abszolút helyi",
|
||||
"calculatingExplanation": "Magyarázat kiszámítása"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Numerikus karakternek kell lennie",
|
||||
"integerError": "Egész számnak kell lennie",
|
||||
"prediction": "Előrejelzés",
|
||||
"predictedProbability": "Előre jelzett valószínűség",
|
||||
"predictionLabel": "Előrejelzés: {0}",
|
||||
"probabilityLabel": "Valószínűség: {0}",
|
||||
"noModelError": "Adjon meg egy üzembe helyezett modellt az ICE-diagramokban lévő előrejelzések vizsgálatához.",
|
||||
"featurePickerLabel": "Funkciófrissítés:",
|
||||
"minimumInputLabel": "Minimum:",
|
||||
"maximumInputLabel": "Maximum:",
|
||||
"stepInputLabel": "Lépések:",
|
||||
"loadingMessage": "Adatok betöltése...",
|
||||
"submitPrompt": "Adja meg az ICE-diagramon megtekinteni kívánt tartományt",
|
||||
"topLevelErrorMessage": "Hiba a paraméterben",
|
||||
"errorPrefix": "Hiba történt: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Betöltés...",
|
||||
"perturbationLabel": "Perturbáció:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Előre jelzett érték: {0}",
|
||||
"predictedClassLabel": "Előre jelzett osztály: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Nincs csoportosítás",
|
||||
"groupPredicted": "Előre jelzett Y",
|
||||
"groupTrue": "Igaz Y",
|
||||
"groupBy": "Csoportosítási szempont"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Diagram típusa:",
|
||||
"violinText": "Hegedű",
|
||||
"barText": "Sáv",
|
||||
"boxText": "Doboz",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "A globális jellemzőfontosság az összes pont (L1 normalizálás) abszolút jellemzőfontossági értékének átlagolásával számítható ki. ",
|
||||
"multiclassImportanceAddendum": "A funkciófontosság összes osztályra vonatkozóan történő kiszámítása minden pontot magában foglal, tehát nem használ különbségi súlyozást. Így egy olyan funkció, amelynek nagy negatív fontossága van számos nem „A osztályúként” előrejelzett pont esetében, nagy mértékben növelni fogja a funkció „A osztályú” fontosságát."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Egyenlő",
|
||||
"greaterThanComparison": "Nagyobb, mint",
|
||||
"greaterThanEqualToComparison": "Nagyobb vagy egyenlő",
|
||||
"lessThanComparison": "Kevesebb, mint",
|
||||
"lessThanEqualToComparison": "Kisebb vagy egyenlő",
|
||||
"inTheRangeOf": "A következő tartományban:",
|
||||
"categoricalIncludeValues": "Belefoglalt értékek:",
|
||||
"numericValue": "Érték",
|
||||
"numericalComparison": "Művelet",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Maximum",
|
||||
"min": "Minimum: {0}",
|
||||
"max": "Maximum: {0}",
|
||||
"uniqueValues": "egyedi értékek száma: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Regressziós hiba",
|
||||
"error": "Hiba",
|
||||
"classificationOutcome": "Besorolás eredménye",
|
||||
"truePositive": "Valós pozitív",
|
||||
"trueNegative": "Valós negatív",
|
||||
"falsePositive": "Vakriasztás",
|
||||
"falseNegative": "Álnegatív",
|
||||
"dataset": "Adathalmaz",
|
||||
"predictedProbabilities": "Előrejelzési valószínűségek",
|
||||
"none": "Mennyiség"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Bezárás",
|
||||
"defaultCustomRootName": "{0}. sor másolata",
|
||||
"filterFeaturePlaceholder": "Jellemzők keresése"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Kohorsz",
|
||||
"defaultLabel": "Minden adat"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Megtekintheti azokat a top-k típusú fontos jellemzőket, amelyek hatással vannak az általános modell-előrejelzésekre (más néven globális értelmezésre). A jellemzőfontossági értékeket a csúszka segítségével jelenítheti meg csökkenő sorrendben. Legfeljebb három kohorszot kiválasztva egymás mellet jelenítheti meg azok jellemzőfontosságát. A diagram bármely jellemzősávjára kattintva megtekintheti, hogy a kiválasztott jellemző értékei miként befolyásolják a modell előrejelzését.",
|
||||
"topAtoB": "Első {0}-{1} jellemző",
|
||||
"datasetCohorts": "Adathalmazi kohorszok",
|
||||
"legendHelpText": "A jelmagyarázat elemeire kattintva be- és kikapcsolhatja a kohorszokat a diagramon.",
|
||||
"sortBy": "Rendezés szempontja",
|
||||
"viewDependencePlotFor": "Függőségi terület megtekintése:",
|
||||
"datasetCohortSelector": "Adathalmazkohorsz kiválasztása",
|
||||
"aggregateFeatureImportance": "Aggregált jellemzőfontosság",
|
||||
"missingParameters": "Ezen a lapon meg kell adni a helyi jellemzőfontossági paramétert.",
|
||||
"weightOptions": "Osztály fontossági súlyozása",
|
||||
"dependencePlotTitle": "Függőségi ábrák",
|
||||
"dependencePlotHelperText": "Ez a függőségi ábra a funkció értéke és a funkció kohorszban betöltött fontossága közötti kapcsolatot mutatja meg.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Jellemző kiválasztása",
|
||||
"datasetRequired": "A függőségi ábrák megkövetelik a kiértékelési adathalmazt és a helyi jellemzőfontossági tömböt."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Adatstatisztikák",
|
||||
"datapoints": "{0} adatpont",
|
||||
"features": "{0} jellemző",
|
||||
"filters": "{0} szűrő",
|
||||
"binaryClassifier": "Bináris osztályozó",
|
||||
"regressor": "Magyarázó változó",
|
||||
"multiclassClassifier": "Többosztályos osztályozó",
|
||||
"datasetCohorts": "Adathalmazi kohorszok",
|
||||
"editCohort": "Kohorsz szerkesztése",
|
||||
"duplicateCohort": "Ismétlődő kohorsz",
|
||||
"addCohort": "Kohorsz hozzáadása",
|
||||
"copy": " másolás"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Megvizsgálhatja a modell teljesítményét a becslési értékek eloszlásának és a modell teljesítménymetrika-értékeinek feltárásával. A modell további vizsgálatával áttekintheti a modellnek az adathalmaz különböző kohorszaiban vagy alcsoportjában mutatott teljesítményét összehasonlító elemzést. Kiválaszthatja az y-értékhez és az x-értékhez a szűrőket a különböző dimenziók kivágásához. A diagram típusát a fogaskerék kiválasztásával módosíthatja a grafikonon.",
|
||||
"modelStatistics": "Modell statisztikái",
|
||||
"cohortPickerLabel": "Válassza ki a böngészni kívánt adathalmazkohorszt",
|
||||
"missingParameters": "Ezen a lapon meg kell adni a modell előrejelzett értékeinek tömbjét.",
|
||||
"missingTrueY": "A modell teljesítménystatisztikáihoz arra van szükség, hogy az előrejelzett kimenetek mellett a tényleges kimenetek is meg legyenek adva"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y-érték",
|
||||
"numberOfDatapoints": "Adatpontok száma",
|
||||
"xValue": "X-érték",
|
||||
"rowIndex": "Sorindex",
|
||||
"featureImportance": "Funkció fontossága",
|
||||
"countTooltipPrefix": "Számláló: {0}",
|
||||
"count": "Mennyiség",
|
||||
"featurePrefix": "Funkció:",
|
||||
"importancePrefix": "Fontosság:",
|
||||
"cohort": "Kohorsz",
|
||||
"howToRead": "A diagram értelmezése"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Ha szeretné megtekinteni az adathalmaz-statisztikákat, adjon meg szűrőket az X, az Y és a színtengelyen az adatok különböző dimenziók szerint történő szeleteléséhez. Fent adathalmaz-kohorszokat hozhat létre, ha olyan szűrőkkel szeretné elemezni az adathalmaz-statisztikákat, mint például a becsült eredmény, az adathalmaz-funkciók vagy a hibacsoportok. A diagram jobb felső sarkában lévő fogaskerékikonra kattintva módosíthatja a diagramtípusokat.",
|
||||
"colorValue": "Színérték",
|
||||
"individualDatapoints": "Egyéni adatpontok",
|
||||
"aggregatePlots": "Aggregált diagramok",
|
||||
"chartType": "Diagram típusa",
|
||||
"missingParameters": "Ehhez a laphoz meg kell adni egy kiértékelési adathalmazt.",
|
||||
"noColor": "Nincs"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "A következő jellemzőfontossága:",
|
||||
"placeholder": "Kattintson a fenti oszlopdiagram egyik funkciójára a függőségi diagram megjelenítéséhez"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "A pontdiagram kívánt adatpontjára kattintva megtekintheti az adatpont helyi jellemzőfontossági értékeit (helyi értelmezését) és ICE-diagramját. Egy ismert adatpont jellemzőinek perturbálásához hozzon létre egy hipotetikus „mi lenne, ha” adatpontot a jobb oldalon lévő panelen. A jellemzőfontossági értékek nem az előrejelzések „okán”, hanem számos approximáción alapulnak. Az ok–okozati következtetés szilárd matematikai megalapozottsága nélkül nem tanácsoljuk a felhasználóknak, hogy a valós helyzetekben ezen eszköz alapján hozzanak döntéseket.",
|
||||
"panelPlaceholder": "Az új adatpontokra vonatkozó előrejelzések készítéséhez modell szükséges.",
|
||||
"cohortPickerLabel": "Válassza ki a böngészni kívánt adathalmazkohorszt",
|
||||
"scatterLegendText": "A jelmagyarázat elemeire kattintva be- és kikapcsolhatja a adatpontokat a diagramon.",
|
||||
"realPoint": "Valós adatpontok",
|
||||
"noneSelectedYet": "Még nincs kiválasztott",
|
||||
"whatIfDatapoints": "Lehetőségelemzési adatpontok",
|
||||
"noneCreatedYet": "Még nincs létrehozott",
|
||||
"showLabel": "Megjelenítés:",
|
||||
"featureImportancePlot": "Jellemzőfontosság ábrázolása",
|
||||
"icePlot": "ICE-diagram",
|
||||
"featureImportanceLackingParameters": "Adja meg a helyi jellemzőfontosságokat, ha meg szeretné tekinteni, hogy az egyes jellemzők hogyan befolyásolják a különálló előrejelzéseket.",
|
||||
"featureImportanceGetStartedText": "Válasszon ki egy pontot a jellemzőfontosság megtekintéséhez",
|
||||
"iceLackingParameters": "Az ICE-ábrákhoz arra van szükség, hogy egy operacionalizált modell előrejelzéseket készítsen feltételes adatpontokkal kapcsolatban.",
|
||||
"IceGetStartedText": "Válasszon ki egy pontot vagy hozzon létre egy lehetőségelemzési pontot az ICE-diagramok megtekintéséhez",
|
||||
"whatIfDatapoint": "Lehetőségelemzési adatpont",
|
||||
"whatIfHelpText": "Válasszon ki egy pontot a diagramon, vagy adjon meg egy ismert adatpontindexet manuálisan a perturbáláshoz és lehetőségelemzési pontként való mentéshez.",
|
||||
"indexLabel": "Adatindex perturbáláshoz",
|
||||
"rowLabel": "Rekord: {0}",
|
||||
"whatIfNameLabel": "Lehetőségelemzési adatpont neve",
|
||||
"featureValues": "Jellemző értékei",
|
||||
"predictedClass": "Előrejelzett osztály: ",
|
||||
"predictedValue": "Előrejelzett érték: ",
|
||||
"probability": "Valószínűség: ",
|
||||
"trueClass": "Igaz osztály: ",
|
||||
"trueValue": "Igaz érték: ",
|
||||
"trueValue.comment": "a tényleges regressziós címke előtagja",
|
||||
"newPredictedClass": "Új előrejelzett osztály: ",
|
||||
"newPredictedValue": "Új előrejelzett érték: ",
|
||||
"newProbability": "Új valószínűség: ",
|
||||
"saveAsNewPoint": "Mentés új pontként",
|
||||
"saveChanges": "Módosítások mentése",
|
||||
"loading": "Betöltés...",
|
||||
"classLabel": "Osztály: {0}",
|
||||
"minLabel": "Min",
|
||||
"maxLabel": "Max",
|
||||
"stepsLabel": "Lépések",
|
||||
"disclaimer": "Jogi nyilatkozat: ezek a magyarázatok számos közelítésen alapulnak, és nem az előrejelzések okai. Az ok-okozati megállapítás szigorú matematikai szilárdsága nélkül nem tanácsoljuk a felhasználóknak, hogy az eszközre alapozva hozzanak valós döntéseket.",
|
||||
"missingParameters": "Ehhez a laphoz meg kell adni egy kiértékelési adathalmazt.",
|
||||
"selectionLimit": "Legfeljebb 3 kiválasztott pont",
|
||||
"classPickerLabel": "Osztály",
|
||||
"tooltipTitleMany": "{0} leggyakoribb előrejelzett osztály",
|
||||
"whatIfTooltipTitle": "Lehetőségelemzés által előrejelzett osztályok",
|
||||
"tooltipTitleFew": "Előrejelzett osztályok",
|
||||
"probabilityLabel": "Valószínűség",
|
||||
"deltaLabel": "Különbözet",
|
||||
"nonNumericValue": "Numerikus értéknek kell lennie",
|
||||
"icePlotHelperText": "Az ICE-diagramok azt mutatják be, hogy a kiválasztott adatpont előrejelzési értékei miként változnak egy adott jellemzőérték-tartomány minimális és a maximális értéke között."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Válasszon szűrőt",
|
||||
"TreatAsCategorical": "Kezelés kategorikusként",
|
||||
"addFilter": "Szűrő hozzáadása",
|
||||
"addedFilters": "Hozzáadott szűrők",
|
||||
"noAddedFilters": "Még nincsenek hozzáadva szűrők",
|
||||
"defaultFilterState": "Válasszon ki egy szűrőt a paraméterek az adatkészlet kohorszához való hozzáadásához.",
|
||||
"cohortNameLabel": "Adatkészlet kohorszának neve",
|
||||
"cohortNamePlaceholder": "Adja meg a kohorsz nevét",
|
||||
"save": "Mentés",
|
||||
"delete": "Törlés",
|
||||
"cancel": "Mégse",
|
||||
"cohortNameError": "Hiányzik a kohorsznév",
|
||||
"placeholderName": "{0} kohorsz"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Kiválasztás",
|
||||
"ditherLabel": "Árnyalás",
|
||||
"selectFilter": "Tengely értékének kiválasztása",
|
||||
"selectFeature": "Jellemző kiválasztása",
|
||||
"binLabel": "Dobozolás alkalmazása az adatokra",
|
||||
"TreatAsCategorical": "Kezelés kategorikusként",
|
||||
"numOfBins": "Dobozok száma",
|
||||
"groupByCohort": "Csoportosítás kohorsz szerint",
|
||||
"selectClass": "Osztály kijelölése",
|
||||
"countHelperText": "A pontok számának hisztogramja"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Előrejelzett valószínűség",
|
||||
"predictedY": "Előrejelzett Y",
|
||||
"evalData": "Értékelési adathalmaz",
|
||||
"localFeatureImportance": "Helyi jellemzőfontosság",
|
||||
"inconsistentDimensions": "Inkonzisztens dimenziók. A(z) {0} dimenziói: {1}, a várt érték: {2}",
|
||||
"notNonEmpty": "A(z) {0} bemenet nem egy nem üres tömb",
|
||||
"varyingLength": "Inkonzisztens dimenziók. A(z) {0} eltérő hosszúságú elemekkel rendelkezik",
|
||||
"notArray": "A(z) {0} nem tömb. A tömb várt dimenziója: {1}",
|
||||
"errorHeader": "Néhány bemeneti paraméter nem volt konzisztens, és nem lesz használva: ",
|
||||
"datasizeWarning": "A kiértékelési adathalmaz túl nagy ahhoz, hogy néhány diagramon hatékonyan megjelenjen. Adjon hozzá szűrőket a kohorsz csökkentéséhez.",
|
||||
"datasizeError": "A kiválasztott kohorsz túl nagy. Adjon hozzá szűrőket a kohorsz csökkentéséhez.",
|
||||
"addFilters": "Szűrők hozzáadása"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " tartalmazza: {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} és {1} további"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R-négyzet: {0}",
|
||||
"meanPrediction": "Előrejelzési átlag: {0}",
|
||||
"accuracy": "Pontosság: {0}",
|
||||
"precision": "Pontosság: {0}",
|
||||
"recall": "Felidézés: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Áttekintheti a modell egészére kiható top k számú legfontosabb jellemzőt. A csúszka segítségével megtekintheti a jellemzők fontosságát csökkenő sorrendben."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Mit jelentenek ezek az értelmezések?",
|
||||
"clickHere": "További információ",
|
||||
"shapTitle": "Shapley-értékek",
|
||||
"shapDescription": "Ez az értelmező az SHAP módszert használja, amely egy olyan játékelméleti modellértelmezési megközelítés, amely a jellemzőhalmazok fontosságát úgy méri, hogy a jellemzőket marginalizálással „elrejti” a modell elől. További információért kattintson az alábbi hivatkozásra.",
|
||||
"limeTitle": "LIME (helyi értelmezhető modellagnosztikus magyarázatok)",
|
||||
"limeDescription": "Ez az értelmező a LIME módszert használja, amely a modell lineáris approximációján alapul. Az értelmezést a következőképpen kaphatjuk meg: perturbáljuk a példányt, beolvassuk a modell előrejelzéseit, és ezeket címkeként használjuk egy olyan ritka lineáris modell betanítására, amely helyileg hű. E lineáris modell súlyozásai jellemzőfontosságokként használhatók. További információért kattintson az alábbi hivatkozásra.",
|
||||
"mimicTitle": "Utánzás (globális helyettesítő értelmezések)",
|
||||
"mimicDescription": "Ez az értelmezés a globális helyettesítő modellek feketedoboz-modellek utánzására történő betanításának ötletén alapul. A globális helyettesítő modell olyan belsőleg értelmezhető modell, amely a feketedoboz-modellek predikcióinak lehető legpontosabb megközelítésére van betanítva. A jellemzőfontossági értékek a mögöttes helyettesítő modell (LightGBM, lineáris regresszió, sztochasztikus gradiensesés vagy döntési fa) modellalapú jellemzőfontossági értékei",
|
||||
"pfiTitle": "Permutációs jellemzőfontosság (PFI)",
|
||||
"pfiDescription": "Ez az értelmező a teljes adathalmaz esetében jellemzőnként összekeveri az adatokat, és kiszámítja, hogy a fontosság teljesítménymetrikája milyen mértékben változik (alapértelmezett teljesítménymetrikák: F1 bináris besorolás esetén, F1-mérték átlagos mikro pontossággal többosztályos besorolás esetén és átlagos abszolút hiba regresszió esetén). Minél nagyobb mértékű a változás, annál fontosabb a jellemző. Ez az értelmező csak az alapul szolgáló modell általános viselkedését képes értelmezni, az egyes előrejelzéseket nem. A jellemző fontossági értéke a modell teljesítményének az adott jellemző perturbálásával kiszámított változását jelzi."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Selezionare un punto per visualizzarne la spiegazione locale",
|
||||
"defaultClassNames": "Classe {0}",
|
||||
"defaultFeatureNames": "Funzionalità {0}",
|
||||
"absoluteAverage": "Media del valore assoluto",
|
||||
"predictedClass": "Classe stimata",
|
||||
"datasetExplorer": "Esplora set di dati",
|
||||
"dataExploration": "Esplorazione del set di dati",
|
||||
"aggregateFeatureImportance": "Importanza delle caratteristiche aggregate",
|
||||
"globalImportance": "Importanza globale",
|
||||
"explanationExploration": "Esplorazione spiegazioni",
|
||||
"individualAndWhatIf": "Importanza della singola caratteristica e simulazione",
|
||||
"summaryImportance": "Riepilogo importanza",
|
||||
"featureImportance": "Importanza della caratteristica",
|
||||
"featureImportanceOf": "Importanza della caratteristica di {0}",
|
||||
"perturbationExploration": "Esplorazione perturbazioni",
|
||||
"localFeatureImportance": "Importanza della caratteristica locale",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Cancella selezione",
|
||||
"feature": "Aggiornamento delle funzionalità:",
|
||||
"intercept": "Intercetta",
|
||||
"modelPerformance": "Prestazioni del modello",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Dati: {0}",
|
||||
"importanceLabel": "Importanza: {0}",
|
||||
"predictedY": "Y stimato",
|
||||
"index": "Indice",
|
||||
"dataGroupLabel": "Dati",
|
||||
"output": "Output",
|
||||
"probabilityLabel": "Probabilità: {0}",
|
||||
"trueY": "True Y",
|
||||
"class": "classe: ",
|
||||
"xValue": "Valore X:",
|
||||
"yValue": "Valore Y:",
|
||||
"colorValue": "Colore:",
|
||||
"count": "Conteggio"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Ponderazione tra classi:",
|
||||
"info": "Informazioni sul calcolo tra classi",
|
||||
"overviewInfo": "I modelli multiclasse generano un vettore di importanza della caratteristica indipendente per ogni classe. Il vettore di importanza della caratteristica di ogni classe mostra le caratteristiche che hanno reso una classe più probabile o meno probabile. È possibile selezionare il modo in cui i pesi dei vettori di importanza della caratteristica per classe vengono riepilogati in un singolo valore:",
|
||||
"absoluteValInfo": "Media del valore assoluto: mostra la somma dell'importanza della funzionalità in tutte le classi possibili, divisa per numero di classi",
|
||||
"predictedClassInfo": "Classe stimata: mostra il valore di importanza della caratteristica per la classe stimata di un determinato punto",
|
||||
"enumeratedClassInfo": "Nomi di classi enumerate: mostra solo i valori di importanza della caratteristica della classe specificata in tutti i punti dati.",
|
||||
"close": "Chiudi",
|
||||
"crossClassWeights": "Pesi tra classi"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Valore della funzionalità in scala",
|
||||
"low": "Minimo",
|
||||
"high": "Massimo",
|
||||
"featureLabel": "Funzionalità: {0}",
|
||||
"valueLabel": "Valore della funzionalità: {0}",
|
||||
"importanceLabel": "Importanza: {0}",
|
||||
"predictedClassTooltip": "Classe stimata: {0}",
|
||||
"trueClassTooltip": "Classe true: {0}",
|
||||
"predictedOutputTooltip": "Output stimato: {0}",
|
||||
"trueOutputTooltip": "Output true: {0}",
|
||||
"topKFeatures": "Prime K funzionalità:",
|
||||
"topKInfo": "Come viene calcolato top k",
|
||||
"predictedValue": "Valore stimato",
|
||||
"predictedClass": "Classe stimata",
|
||||
"trueValue": "Valore True",
|
||||
"trueClass": "Classe true",
|
||||
"noColor": "Nessuno",
|
||||
"tooManyRows": "Il set di dati fornito è più grande di quello che questo grafico può supportare"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Classe: {0}",
|
||||
"sortBy": "Ordina per",
|
||||
"noData": "Dati non disponibili",
|
||||
"absoluteGlobal": "Assoluto globale",
|
||||
"absoluteLocal": "Assoluto locale",
|
||||
"calculatingExplanation": "Calcolo della spiegazione"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Deve essere un carattere numerico",
|
||||
"integerError": "Deve essere un numero intero",
|
||||
"prediction": "Stima",
|
||||
"predictedProbability": "Probabilità stimata",
|
||||
"predictionLabel": "Stima: {0}",
|
||||
"probabilityLabel": "Probabilità: {0}",
|
||||
"noModelError": "Fornire un modello operativo per esplorare le stime nei tracciati ICE.",
|
||||
"featurePickerLabel": "Aggiornamento delle funzionalità:",
|
||||
"minimumInputLabel": "Minimo:",
|
||||
"maximumInputLabel": "Massimo:",
|
||||
"stepInputLabel": "Passi:",
|
||||
"loadingMessage": "Caricamento dei dati...",
|
||||
"submitPrompt": "Inviare un intervallo per visualizzare un tracciato ICE",
|
||||
"topLevelErrorMessage": "Errore nel parametro",
|
||||
"errorPrefix": "Errore riscontrato: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Caricamento...",
|
||||
"perturbationLabel": "Perturbazione:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Valore stimato: {0}",
|
||||
"predictedClassLabel": "Classe stimata: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Nessun raggruppamento",
|
||||
"groupPredicted": "Y stimato",
|
||||
"groupTrue": "True Y",
|
||||
"groupBy": "Raggruppa per"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Tipo di grafico:",
|
||||
"violinText": "Violino",
|
||||
"barText": "Barre",
|
||||
"boxText": "Casella",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "L'importanza della caratteristica globale viene calcolata facendo la media del valore assoluto dell'importanza della caratteristica di tutti i punti (normalizzazione L1). ",
|
||||
"multiclassImportanceAddendum": "Tutti i punti sono inclusi nel calcolo dell'importanza di una funzionalità per tutte le classi, non viene usata alcuna ponderazione differenziale. Quindi, una funzionalità con grande importanza negativa per molti punti che si prevede non siano di 'Classe A' aumenterà notevolmente l'importanza 'Classe A' di tale funzionalità."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Uguale a",
|
||||
"greaterThanComparison": "Maggiore di",
|
||||
"greaterThanEqualToComparison": "Maggiore o uguale a",
|
||||
"lessThanComparison": "Minore di",
|
||||
"lessThanEqualToComparison": "Minore o uguale a",
|
||||
"inTheRangeOf": "Nell'intervallo",
|
||||
"categoricalIncludeValues": "Valori inclusi:",
|
||||
"numericValue": "Valore",
|
||||
"numericalComparison": "Operazione",
|
||||
"minimum": "Minimo",
|
||||
"maximum": "Massimo",
|
||||
"min": "Min: {0}",
|
||||
"max": "Max: {0}",
|
||||
"uniqueValues": "Numero di valori univoci: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Errore di regressione",
|
||||
"error": "Errore",
|
||||
"classificationOutcome": "Risultato della classificazione",
|
||||
"truePositive": "Vero positivo",
|
||||
"trueNegative": "Vero negativo",
|
||||
"falsePositive": "Falso positivo",
|
||||
"falseNegative": "Falso negativo",
|
||||
"dataset": "Set di dati",
|
||||
"predictedProbabilities": "Probabilità di stima",
|
||||
"none": "Conteggio"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Chiudi",
|
||||
"defaultCustomRootName": "Copia della riga {0}",
|
||||
"filterFeaturePlaceholder": "Cerca funzionalità"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Coorte",
|
||||
"defaultLabel": "Tutti i dati"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Esplorare le caratteristiche importanti top - k che influiscono sulle stime generali del modello (ovvero spiegazione globale). Usare il dispositivo di scorrimento per visualizzare i valori di importanza della caratteristica in ordine decrescente. Selezionare un massimo di tre coorti per visualizzare i valori di importanza della caratteristica affiancati. Fare clic su una delle barre delle caratteristiche nel grafo per visualizzare l'impatto dei valori della caratteristica selezionata sulla previsione del modello.",
|
||||
"topAtoB": "Prime {0}-{1} caratteristiche",
|
||||
"datasetCohorts": "Coorti di set di dati",
|
||||
"legendHelpText": "Attivare e disattivare le coorti nel tracciato facendo clic sugli elementi della legenda.",
|
||||
"sortBy": "Ordina per",
|
||||
"viewDependencePlotFor": "Visualizza tracciato delle dipendenze per:",
|
||||
"datasetCohortSelector": "Seleziona una coorte di set di dati",
|
||||
"aggregateFeatureImportance": "Importanza delle caratteristiche aggregate",
|
||||
"missingParameters": "Questa scheda richiede che sia fornito il parametro di importanza della caratteristica locale.",
|
||||
"weightOptions": "Pesi di importanza della classe",
|
||||
"dependencePlotTitle": "Tracciati delle dipendenze",
|
||||
"dependencePlotHelperText": "Questo tracciato delle dipendenze mostra la relazione tra il valore di una caratteristica e l'importanza corrispondente della caratteristica in una coorte.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Seleziona caratteristica",
|
||||
"datasetRequired": "I tracciati delle dipendenze richiedono il set di dati di valutazione e la matrice di importanza delle caratteristiche locali."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Statistiche dati",
|
||||
"datapoints": "{0} punti dati",
|
||||
"features": "{0} funzionalità",
|
||||
"filters": "{0} filtri",
|
||||
"binaryClassifier": "Classificatore binario",
|
||||
"regressor": "Regressore",
|
||||
"multiclassClassifier": "Classificatore multiclasse",
|
||||
"datasetCohorts": "Coorti di set di dati",
|
||||
"editCohort": "Modifica coorte",
|
||||
"duplicateCohort": "Duplica coorte",
|
||||
"addCohort": "Aggiungi coorte",
|
||||
"copy": " copia"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Valutare le prestazioni del modello esplorando la distribuzione dei valori di stima e i valori delle metriche delle prestazioni del modello. È possibile esaminare ulteriormente il modello osservando un'analisi comparativa delle relative prestazioni in diverse coorti o sottogruppi del set di dati. Selezionare i filtri con il valore y e il valore x per intersecare dimensioni diverse. Selezionare l'icona a forma di ingranaggio nel grafico per modificare il tipo di grafico.",
|
||||
"modelStatistics": "Statistiche modello",
|
||||
"cohortPickerLabel": "Seleziona una coorte di set di dati da esplorare",
|
||||
"missingParameters": "Questa scheda richiede che sia fornita la matrice di valori stimati del modello.",
|
||||
"missingTrueY": "Le statistiche sulle prestazioni del modello richiedono che vengano forniti i risultati reali oltre ai risultati previsti"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Valore Y",
|
||||
"numberOfDatapoints": "Numero di punti dati",
|
||||
"xValue": "Valore X",
|
||||
"rowIndex": "Indice di riga",
|
||||
"featureImportance": "Importanza della caratteristica",
|
||||
"countTooltipPrefix": "Conteggio: {0}",
|
||||
"count": "Conteggio",
|
||||
"featurePrefix": "Caratteristica",
|
||||
"importancePrefix": "Importanza",
|
||||
"cohort": "Coorte",
|
||||
"howToRead": "Come leggere questo grafico"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Esplorare le statistiche dei set di dati selezionando diversi filtri lungo l'asse X, Y e colore per suddividere i dati in dimensioni diverse. Creare le coorti di set di dati sopra per analizzare le statistiche dei set di dati con filtri, ad esempio risultati stimati, funzionalità del set di dati e gruppi di errori. Usare l'icona dell'ingranaggio nell'angolo in alto a destra del grafo per modificare i tipi di grafo.",
|
||||
"colorValue": "Valore colore",
|
||||
"individualDatapoints": "Singoli punti dati",
|
||||
"aggregatePlots": "Aggrega tracciati",
|
||||
"chartType": "Tipo di grafico",
|
||||
"missingParameters": "Questa scheda richiede che sia fornito un set di dati di valutazione.",
|
||||
"noColor": "Nessuno"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Importanza della caratteristica di",
|
||||
"placeholder": "Fare clic su una caratteristica nel grafico a barre sopra per mostrarne il tracciato delle dipendenze"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "È possibile selezionare un punto dati facendo clic sul diagramma di dispersione per visualizzarne i valori di importanza della caratteristica locali (spiegazione locale) e il tracciato Aspettativa condizionale individuale (ICE, Individual Conditional Expectation) seguente. Creare un punto dati What If ipotetico usando il pannello sulla destra per perturbare le caratteristiche di un punto dati noto. I valori di importanza della caratteristica si basano su molte approssimazioni e non sono la \"causa\" delle stime. Senza una rigorosa affidabilità matematica dell'inferenza causale, non si consiglia agli utenti di prendere decisioni reali basate su questo strumento.",
|
||||
"panelPlaceholder": "Per effettuare stime per i nuovi punti dati, è necessario un modello.",
|
||||
"cohortPickerLabel": "Seleziona una coorte di set di dati da esplorare",
|
||||
"scatterLegendText": "Attivare e disattivare i punti dati nel tracciato facendo clic sugli elementi della legenda.",
|
||||
"realPoint": "Punti dati reali",
|
||||
"noneSelectedYet": "Non è stato ancora selezionato alcun punto",
|
||||
"whatIfDatapoints": "Punti dati di simulazione",
|
||||
"noneCreatedYet": "Non è stato ancora creato alcun punto",
|
||||
"showLabel": "Mostra:",
|
||||
"featureImportancePlot": "Tracciato dell'importanza della caratteristica",
|
||||
"icePlot": "Tracciato Aspettativa condizionale individuale (ICE, Individual Conditional Expectation)",
|
||||
"featureImportanceLackingParameters": "Fornire l'importanza delle caratteristiche locali per vedere in che modo ogni caratteristica influisce sulle singole stime.",
|
||||
"featureImportanceGetStartedText": "Seleziona un punto per visualizzare l'importanza della caratteristica",
|
||||
"iceLackingParameters": "I tracciati ICE richiedono un modello operativo per fare stime per punti dati ipotetici.",
|
||||
"IceGetStartedText": "Selezionare un punto o creare un punto di simulazione per visualizzare i tracciati ICE",
|
||||
"whatIfDatapoint": "Punto dati di simulazione",
|
||||
"whatIfHelpText": "Selezionare un punto nel tracciato o immettere manualmente un indice del punto dati noto per perturbare e salvare come nuovo punto di simulazione.",
|
||||
"indexLabel": "Indice dati da perturbare",
|
||||
"rowLabel": "Riga {0}",
|
||||
"whatIfNameLabel": "Nome del punto dati di simulazione",
|
||||
"featureValues": "Valori delle caratteristiche",
|
||||
"predictedClass": "Classe stimata: ",
|
||||
"predictedValue": "Valore stimato: ",
|
||||
"probability": "Probabilità: ",
|
||||
"trueClass": "Classe true: ",
|
||||
"trueValue": "Valore true: ",
|
||||
"trueValue.comment": "prefisso per l'etichetta effettiva per la regressione",
|
||||
"newPredictedClass": "Nuova classe stimata: ",
|
||||
"newPredictedValue": "Nuovo valore stimato: ",
|
||||
"newProbability": "Nuova probabilità: ",
|
||||
"saveAsNewPoint": "Salva come nuovo punto",
|
||||
"saveChanges": "Salva modifiche",
|
||||
"loading": "Caricamento...",
|
||||
"classLabel": "Classe: {0}",
|
||||
"minLabel": "Min",
|
||||
"maxLabel": "Max",
|
||||
"stepsLabel": "Passi",
|
||||
"disclaimer": "Dichiarazione di non responsabilità: queste sono spiegazioni basate su molte approssimazioni e non sono la \"causa\" delle stime. Senza una rigorosa affidabilità matematica dell'inferenza causale, non si consiglia agli utenti di prendere decisioni reali basate su questo strumento.",
|
||||
"missingParameters": "Questa scheda richiede che sia fornito un set di dati di valutazione.",
|
||||
"selectionLimit": "Massimo di 3 punti selezionati",
|
||||
"classPickerLabel": "Classe",
|
||||
"tooltipTitleMany": "Prime {0} classi stimate",
|
||||
"whatIfTooltipTitle": "Classi stimate di simulazione",
|
||||
"tooltipTitleFew": "Classi stimate",
|
||||
"probabilityLabel": "Probabilità",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "Il valore deve essere un valore numerico",
|
||||
"icePlotHelperText": "I tracciati ICE illustrano la variazione dei valori di stima del punto dati selezionato in un intervallo di valori di caratteristica compreso tra un valore minimo e un valore massimo."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Seleziona filtro",
|
||||
"TreatAsCategorical": "Gestisci come categorie",
|
||||
"addFilter": "Aggiungi filtro",
|
||||
"addedFilters": "Filtri aggiunti",
|
||||
"noAddedFilters": "Non sono stati ancora aggiunti filtri",
|
||||
"defaultFilterState": "Selezionare un filtro per aggiungere i parametri alla coorte del set di dati.",
|
||||
"cohortNameLabel": "Nome coorte del set di dati",
|
||||
"cohortNamePlaceholder": "Assegnare un nome alla coorte",
|
||||
"save": "Salva",
|
||||
"delete": "Elimina",
|
||||
"cancel": "Annulla",
|
||||
"cohortNameError": "Nome coorte mancante",
|
||||
"placeholderName": "Coorte {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Seleziona",
|
||||
"ditherLabel": "Applica il dithering",
|
||||
"selectFilter": "Seleziona il valore dell'asse",
|
||||
"selectFeature": "Seleziona funzionalità",
|
||||
"binLabel": "Raggruppa i dati in contenitori",
|
||||
"TreatAsCategorical": "Gestisci come categorie",
|
||||
"numOfBins": "Numero di contenitori",
|
||||
"groupByCohort": "Raggruppa per coorte",
|
||||
"selectClass": "Seleziona classe",
|
||||
"countHelperText": "Istogramma del numero di punti"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Probabilità stimata",
|
||||
"predictedY": "Y stimato",
|
||||
"evalData": "Set di dati di valutazione",
|
||||
"localFeatureImportance": "Importanza della caratteristica locale",
|
||||
"inconsistentDimensions": "Dimensioni incoerenti. {0} contiene dimensioni {1}, previsto {2}",
|
||||
"notNonEmpty": "Input {0} non contiene una matrice non vuota",
|
||||
"varyingLength": "Dimensioni incoerenti. {0} contiene elementi di lunghezza variabile",
|
||||
"notArray": "{0} non contiene una matrice. È prevista una matrice di dimensione {1}",
|
||||
"errorHeader": "Alcuni parametri di input sono incoerenti e non verranno usati: ",
|
||||
"datasizeWarning": "Il set di dati di valutazione è troppo grande per essere visualizzato in modo efficace in alcuni grafici, aggiungere i filtri per ridurre le dimensioni della coorte. ",
|
||||
"datasizeError": "La coorte selezionata è troppo grande, aggiungere i filtri per ridurre le dimensioni della coorte.",
|
||||
"addFilters": "Aggiungi filtri"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " include {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} e altri {1}"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R-squared: {0}",
|
||||
"meanPrediction": "Stima media {0}",
|
||||
"accuracy": "Accuratezza: {0}",
|
||||
"precision": "Precisione: {0}",
|
||||
"recall": "Richiamo: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Esplorare le caratteristiche importanti top k che influiscono sulle stime generali del modello. Usare il dispositivo di scorrimento per visualizzare l'importanza delle caratteristiche in ordine decrescente."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Che cosa indicano queste spiegazioni?",
|
||||
"clickHere": "Altre informazioni",
|
||||
"shapTitle": "Valori di Shapley",
|
||||
"shapDescription": "Questo explainer usa SHAP, ovvero un approccio basato sulla teoria dei giochi per spiegare i modelli, in cui l'importanza dei set di caratteristiche viene misurata \"nascondendo\" tali caratteristiche nel modello tramite la marginalizzazione. Per altre informazioni, fare clic sul collegamento seguente.",
|
||||
"limeTitle": "LIME (Local Interpretable Model-Agnostic Explanations)",
|
||||
"limeDescription": "Questo explainer usa LIME, che fornisce un'approssimazione lineare del modello. Per ottenere una spiegazione, seguire questa procedura: perturbare l'istanza, ottenere le stime del modello e usarle come etichette per apprendere un modello lineare di tipo sparse fedele a livello locale. I pesi di questo modello lineare vengono usati come 'importanze della caratteristica'. Per altre informazioni, fare clic sul collegamento seguente.",
|
||||
"mimicTitle": "Simulazione (Global Surrogate Explanations)",
|
||||
"mimicDescription": "Questo explainer si basa sul concetto di training dei modelli di surrogato globale per simulare modelli black box. Per modello di surrogato globale si intende un modello intrinsecamente interpretabile di cui viene eseguito il training per approssimare le stime di qualsiasi modello black box nel modo più accurato possibile. I valori di importanza della caratteristica sono valori di importanza della caratteristica basati sul modello del modello surrogato sottostante (LightGBM, regressione lineare, discesa stocastica del gradiente o albero delle decisioni)",
|
||||
"pfiTitle": "Permutation Feature Importance (PFI)",
|
||||
"pfiDescription": "Questo explainer seleziona in ordine casuale i dati una caratteristica alla volta per l'intero set di dati e calcola la variazione della metrica delle prestazioni di interesse (metriche delle prestazioni predefinite: F1 per la classificazione binaria, punteggio F1 con media micro per la classificazione multiclasse e errore assoluto medio per la regressione). Maggiore è la variazione, più importante è la caratteristica. Questo explainer può solo spiegare il comportamento complessivo del modello sottostante, ma non spiega le singole stime. Il valore dell'importanza di una caratteristica rappresenta il delta nelle prestazioni del modello e causa la perturbazione di tale caratteristica specifica."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "ローカルの説明を表示する点を選択します",
|
||||
"defaultClassNames": "クラス {0}",
|
||||
"defaultFeatureNames": "機能 {0}",
|
||||
"absoluteAverage": "絶対値の平均",
|
||||
"predictedClass": "予測クラス",
|
||||
"datasetExplorer": "データセット エクスプローラー",
|
||||
"dataExploration": "データセットの探索",
|
||||
"aggregateFeatureImportance": "特徴量の重要度の集計",
|
||||
"globalImportance": "グローバルの重要度",
|
||||
"explanationExploration": "説明の探索",
|
||||
"individualAndWhatIf": "個々の特徴量の重要度と What-If",
|
||||
"summaryImportance": "概要の重要度",
|
||||
"featureImportance": "特徴量の重要度",
|
||||
"featureImportanceOf": "{0} の特徴量の重要度",
|
||||
"perturbationExploration": "補正の探索",
|
||||
"localFeatureImportance": "ローカル特徴量の重要度",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "選択範囲のクリア",
|
||||
"feature": "機能:",
|
||||
"intercept": "切片",
|
||||
"modelPerformance": "モデルのパフォーマンス",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "データ: {0}",
|
||||
"importanceLabel": "重要度: {0}",
|
||||
"predictedY": "予測 Y",
|
||||
"index": "インデックス",
|
||||
"dataGroupLabel": "データ",
|
||||
"output": "出力",
|
||||
"probabilityLabel": "確率: {0}",
|
||||
"trueY": "True Y",
|
||||
"class": "クラス:",
|
||||
"xValue": "X 値:",
|
||||
"yValue": "Y 値:",
|
||||
"colorValue": "色:",
|
||||
"count": "カウント"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "クロスクラスの重み付け:",
|
||||
"info": "クラス間の計算に関する情報",
|
||||
"overviewInfo": "マルチクラス モデルにより、クラスごとに独立した特徴量の重要度ベクターが生成されます。各クラスの特徴量の重要度ベクターは、クラスの可能性を高めたか低くした機能を示しています。クラスごとの特徴量の重要度ベクターの重みを 1 つの値に集計する方法を選択できます。",
|
||||
"absoluteValInfo": "絶対値の平均: 指定できるすべてのクラスにおける機能の重要性の合計をクラスの数で割ったものを表示します。",
|
||||
"predictedClassInfo": "予測クラス: 指定した点の予測クラスについて特徴量の重要度の値を表示します",
|
||||
"enumeratedClassInfo": "列挙クラス名: 指定されたクラスの特徴量の重要度の値のみをすべてのデータ ポイントに表示します。",
|
||||
"close": "閉じる",
|
||||
"crossClassWeights": "クロス クラスの重み"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "スケーリングされた機能の値",
|
||||
"low": "安値",
|
||||
"high": "高値",
|
||||
"featureLabel": "機能: {0}",
|
||||
"valueLabel": "機能の値: {0}",
|
||||
"importanceLabel": "重要度: {0}",
|
||||
"predictedClassTooltip": "予測クラス: {0}",
|
||||
"trueClassTooltip": "真のクラス: {0}",
|
||||
"predictedOutputTooltip": "予測出力: {0}",
|
||||
"trueOutputTooltip": "真の出力: {0}",
|
||||
"topKFeatures": "上位の K 個の機能:",
|
||||
"topKInfo": "上位 k 個を計算する方法",
|
||||
"predictedValue": "予測された値",
|
||||
"predictedClass": "予測クラス",
|
||||
"trueValue": "真の値",
|
||||
"trueClass": "真のクラス",
|
||||
"noColor": "なし",
|
||||
"tooManyRows": "指定されたデータセットは、このグラフでサポートできるサイズを超えています"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "クラス: {0}",
|
||||
"sortBy": "並べ替え",
|
||||
"noData": "データなし",
|
||||
"absoluteGlobal": "絶対グローバル",
|
||||
"absoluteLocal": "絶対ローカル",
|
||||
"calculatingExplanation": "説明を計算しています"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "数値を指定する必要があります",
|
||||
"integerError": "整数にする必要があります",
|
||||
"prediction": "予測",
|
||||
"predictedProbability": "予測確率",
|
||||
"predictionLabel": "予測: {0}",
|
||||
"probabilityLabel": "確率: {0}",
|
||||
"noModelError": "ICE プロットで予測を調べるには、操作可能なモデルを指定してください。",
|
||||
"featurePickerLabel": "機能:",
|
||||
"minimumInputLabel": "最小:",
|
||||
"maximumInputLabel": "最大:",
|
||||
"stepInputLabel": "ステップ:",
|
||||
"loadingMessage": "データを読み込んでいます...",
|
||||
"submitPrompt": "ICE プロットを表示するには範囲を送信します",
|
||||
"topLevelErrorMessage": "パラメーターでエラーが発生しました",
|
||||
"errorPrefix": "エラーが発生しました: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "読み込んでいます...",
|
||||
"perturbationLabel": "摂動:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "予測値: {0}",
|
||||
"predictedClassLabel": "予測クラス: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "グループ化なし",
|
||||
"groupPredicted": "予測 Y",
|
||||
"groupTrue": "True Y",
|
||||
"groupBy": "グループ化"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "グラフの種類:",
|
||||
"violinText": "バイオリン",
|
||||
"barText": "横棒",
|
||||
"boxText": "ボックス",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "すべての点の特徴量の重要度の絶対値を平均化 (L1 正規化) して、グローバル特徴量の重要度を計算します。",
|
||||
"multiclassImportanceAddendum": "すべてのクラスについて機能の重要度を計算するときにすべての点が含まれます。差分の重み付けは使用されません。したがって、'クラス A' ではないと予測される多くの点に対して大きな負の重要度をもつ機能は、その機能の 'クラス A' の重要度を大幅に増加させます。"
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "次の値に等しい",
|
||||
"greaterThanComparison": "次の値より大きい",
|
||||
"greaterThanEqualToComparison": "次の値より大きいか等しい",
|
||||
"lessThanComparison": "次の値より小さい",
|
||||
"lessThanEqualToComparison": "次の値より小さいか等しい",
|
||||
"inTheRangeOf": "次の範囲",
|
||||
"categoricalIncludeValues": "含める値:",
|
||||
"numericValue": "値",
|
||||
"numericalComparison": "操作",
|
||||
"minimum": "最小値",
|
||||
"maximum": "最大値",
|
||||
"min": "最小: {0}",
|
||||
"max": "最大: {0}",
|
||||
"uniqueValues": "一意の値の数: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "回帰エラー",
|
||||
"error": "エラー",
|
||||
"classificationOutcome": "分類の結果",
|
||||
"truePositive": "真陽性",
|
||||
"trueNegative": "真陰性",
|
||||
"falsePositive": "偽陽性",
|
||||
"falseNegative": "偽陰性",
|
||||
"dataset": "データセット",
|
||||
"predictedProbabilities": "予測確率",
|
||||
"none": "カウント"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "閉じる",
|
||||
"defaultCustomRootName": "行 {0} のコピー",
|
||||
"filterFeaturePlaceholder": "特徴量を検索してください"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "コーホート",
|
||||
"defaultLabel": "すべてのデータ"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "モデル全体の予測に影響を与える上位 k 個の重要な特徴量を探索します (グローバル説明としても知られます)。スライダーを使用して、特徴量の重要度の値を降順で表示します。最大 3 個のコーホートを選択して、特徴量の重要度を左右に並べて表示します。グラフ内のいずれかの特徴量バーをクリックすると、選択した特徴量の値がモデル予測に与える影響が表示されます。",
|
||||
"topAtoB": "上位 {0} から {1} の特徴量",
|
||||
"datasetCohorts": "データセットのコーホート",
|
||||
"legendHelpText": "凡例項目をクリックして、プロットのコーホートのオンとオフを切り替えます。",
|
||||
"sortBy": "並べ替え",
|
||||
"viewDependencePlotFor": "従属プロットの表示:",
|
||||
"datasetCohortSelector": "データセットのコーホートを選択",
|
||||
"aggregateFeatureImportance": "特徴量の重要度の集計",
|
||||
"missingParameters": "このタブでは、ローカル特徴量の重要度パラメーターを指定する必要があります。",
|
||||
"weightOptions": "クラスの重要度の重み",
|
||||
"dependencePlotTitle": "従属プロット",
|
||||
"dependencePlotHelperText": "この従属プロットは、コーホートでの特徴量の値と対応する特徴量の重要度の関係を示しています。",
|
||||
"dependencePlotFeatureSelectPlaceholder": "特徴量を選択してください",
|
||||
"datasetRequired": "従属プロットには、評価データセットとローカルの特徴量の重要度配列が必要です。"
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "データの統計情報",
|
||||
"datapoints": "{0} 個のデータポイント",
|
||||
"features": "{0} 個の特徴量",
|
||||
"filters": "{0} 個のフィルター",
|
||||
"binaryClassifier": "2 項分類子",
|
||||
"regressor": "リグレッサー",
|
||||
"multiclassClassifier": "マルチクラスの分類子",
|
||||
"datasetCohorts": "データセットのコーホート",
|
||||
"editCohort": "コーホートの編集",
|
||||
"duplicateCohort": "コーホートの複製",
|
||||
"addCohort": "コーホートの追加",
|
||||
"copy": " コピー"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "モデルのパフォーマンスを評価するには、得られた予測値とご使用のモデルのパフォーマンス メトリックの値の分布を調べます。データセットのさまざまなコーホートまたはサブグループにおけるパフォーマンスの比較分析を確認して、モデルを詳しく調べることができます。さまざまなディメンションを横断するよう、y 値と x 値に沿ったフィルターを選択します。グラフの種類を変更するには、グラフの歯車を選択します。",
|
||||
"modelStatistics": "モデルの統計情報",
|
||||
"cohortPickerLabel": "探索するデータセットのコーホートの選択",
|
||||
"missingParameters": "このタブでは、モデルから予測される値の配列を指定する必要があります。",
|
||||
"missingTrueY": "モデル パフォーマンス統計では、予測された結果に加えて、実際の結果を提供する必要があります"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y 値",
|
||||
"numberOfDatapoints": "データポイントの数",
|
||||
"xValue": "X 値",
|
||||
"rowIndex": "行インデックス",
|
||||
"featureImportance": "特徴量の重要度",
|
||||
"countTooltipPrefix": "個数: {0}",
|
||||
"count": "カウント",
|
||||
"featurePrefix": "特徴量",
|
||||
"importancePrefix": "重要度",
|
||||
"cohort": "コーホート",
|
||||
"howToRead": "このグラフの読み取り方法"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "X 軸、Y 軸、および色軸に沿ってさまざまなフィルターを選択してデータセットの統計を探索し、異なるディメンションに沿ってデータをスライスします。上のデータセット コーホートを作成して、予測結果、データセットの特徴量、エラー グループなどのフィルターを使用してデータセット統計を分析します。グラフの種類を変更するには、グラフの右上隅にある歯車のアイコンを使用してください。",
|
||||
"colorValue": "色の値",
|
||||
"individualDatapoints": "個々のデータポイント",
|
||||
"aggregatePlots": "集計プロット",
|
||||
"chartType": "グラフの種類",
|
||||
"missingParameters": "このタブでは、評価データセットを指定する必要があります。",
|
||||
"noColor": "なし"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "特徴量の重要度:",
|
||||
"placeholder": "上の横棒グラフの特徴量をクリックして、その従属プロットを表示してください"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "散布図をクリックしてデータポイントを選択することで、その局所的な特徴量の重要度の値 (局所的な説明) と以下の個別条件付き期待値 (ICE) プロットを表示することができます。右側のパネルを使用して仮定の what-if データポイントを作成し、既知のデータポイントの特徴量に摂動を与えてください。特徴量の重要度の値は多数の近似値に基づいており、予測の \"原因\" ではありません。因果推論の厳密な数学的堅牢性がなければ、このツールに基づいて実際の決定を行うことはお勧めしません。",
|
||||
"panelPlaceholder": "新しいデータ ポイントの予測を行うには、モデルが必要です。",
|
||||
"cohortPickerLabel": "探索するデータセットのコーホートの選択",
|
||||
"scatterLegendText": "凡例項目をクリックして、プロットのデータポイントのオンとオフを切り替えます。",
|
||||
"realPoint": "実際のデータポイント",
|
||||
"noneSelectedYet": "まだ何も選択されていません",
|
||||
"whatIfDatapoints": "What-If データポイント",
|
||||
"noneCreatedYet": "まだ何も作成されていません",
|
||||
"showLabel": "表示:",
|
||||
"featureImportancePlot": "特徴量の重要度プロット",
|
||||
"icePlot": "個別条件付き期待値 (ICE) プロット",
|
||||
"featureImportanceLackingParameters": "それぞれの特徴量が個々の予測に与える影響を確認するため、ローカルの特徴量の重要度を指定してください。",
|
||||
"featureImportanceGetStartedText": "特徴量の重要度を表示するポイントを選択してください",
|
||||
"iceLackingParameters": "ICE プロットには、仮定のデータ ポイントを予測するための運用可能なモデルが必要です。",
|
||||
"IceGetStartedText": "ICE プロットを表示するには、ポイントを選択するか、What-If ポイントを作成してください",
|
||||
"whatIfDatapoint": "What-If データポイント",
|
||||
"whatIfHelpText": "プロット上のポイントを選択するか、摂動する既知のデータポイント インデックスを手動で入力して、新しい What-If ポイントとして保存します。",
|
||||
"indexLabel": "摂動するデータ インデックス",
|
||||
"rowLabel": "行 {0}",
|
||||
"whatIfNameLabel": "What-If データポイント名",
|
||||
"featureValues": "特徴量の値",
|
||||
"predictedClass": "予測クラス: ",
|
||||
"predictedValue": "予測された値: ",
|
||||
"probability": "確率: ",
|
||||
"trueClass": "真のクラス: ",
|
||||
"trueValue": "真の値: ",
|
||||
"trueValue.comment": "回帰の実際のラベルにプレフィックスを付ける",
|
||||
"newPredictedClass": "新しい予測クラス: ",
|
||||
"newPredictedValue": "新しい予測値: ",
|
||||
"newProbability": "新しい確率: ",
|
||||
"saveAsNewPoint": "新しいポイントとして保存",
|
||||
"saveChanges": "変更の保存",
|
||||
"loading": "読み込み中...",
|
||||
"classLabel": "クラス: {0}",
|
||||
"minLabel": "最小",
|
||||
"maxLabel": "最大",
|
||||
"stepsLabel": "ステップ",
|
||||
"disclaimer": "免責事項: これらは多くの概算に基づいた説明であり、予測の \"原因\" ではありません。因果推論の厳密な数学的頑健性がない場合、このツールに基づいて実際の意思決定を行わないようユーザーにお勧めします。",
|
||||
"missingParameters": "このタブでは、評価データセットを指定する必要があります。",
|
||||
"selectionLimit": "最大数である 3 つのポイントが選択済みです",
|
||||
"classPickerLabel": "クラス",
|
||||
"tooltipTitleMany": "上位 {0} 件の予測クラス",
|
||||
"whatIfTooltipTitle": "What-If 予測クラス",
|
||||
"tooltipTitleFew": "予測クラス",
|
||||
"probabilityLabel": "確率",
|
||||
"deltaLabel": "差分",
|
||||
"nonNumericValue": "値は数値でなければなりません",
|
||||
"icePlotHelperText": "ICE プロットは、選択したデータ ポイントの予測値が、最小値と最大値の間のある範囲の特徴量の値に沿って変化する様子を示しています。"
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "フィルターを選択してください",
|
||||
"TreatAsCategorical": "カテゴリとして扱う",
|
||||
"addFilter": "フィルターの追加",
|
||||
"addedFilters": "追加されたフィルター",
|
||||
"noAddedFilters": "まだフィルターは追加されていません",
|
||||
"defaultFilterState": "データセットのコーホートにパラメーターを追加するには、フィルターを選択してください。",
|
||||
"cohortNameLabel": "データセットのコーホート名",
|
||||
"cohortNamePlaceholder": "コーホートの名前を指定してください",
|
||||
"save": "保存",
|
||||
"delete": "削除",
|
||||
"cancel": "キャンセル",
|
||||
"cohortNameError": "コーホート名がありません",
|
||||
"placeholderName": "コーホート {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "選択",
|
||||
"ditherLabel": "ディザーが必要",
|
||||
"selectFilter": "軸の値の選択",
|
||||
"selectFeature": "特徴量の選択",
|
||||
"binLabel": "データにビン分割を適用する",
|
||||
"TreatAsCategorical": "カテゴリとして扱う",
|
||||
"numOfBins": "ビンの数",
|
||||
"groupByCohort": "コーホートでグループ化",
|
||||
"selectClass": "クラスの選択",
|
||||
"countHelperText": "ポイント数のヒストグラム"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "予測確率",
|
||||
"predictedY": "予測 Y",
|
||||
"evalData": "評価データセット",
|
||||
"localFeatureImportance": "ローカル特徴量の重要度",
|
||||
"inconsistentDimensions": "ディメンションが矛盾しています。{0} のディメンションは {1} ですが、必要なのは {2} です",
|
||||
"notNonEmpty": "{0} 入力は、空でない配列ではありません",
|
||||
"varyingLength": "ディメンションが矛盾しています。{0} には、長さの異なる要素があります",
|
||||
"notArray": "{0} は配列ではありません。ディメンション {1} の配列が必要です",
|
||||
"errorHeader": "一部の入力パラメーターは、矛盾しているため、使用されません: ",
|
||||
"datasizeWarning": "評価データセットが大きすぎて、一部のグラフで効果的に表示できません。フィルターを追加して、コーホートのサイズを小さくしてください。",
|
||||
"datasizeError": "選択したコーホートが大きすぎます。フィルターを追加して、コーホートのサイズを小さくしてください。",
|
||||
"addFilters": "フィルターの追加"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " {0} を含む",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} およびその他の {1} 人"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R-2 乗: {0}",
|
||||
"meanPrediction": "平均予測 {0}",
|
||||
"accuracy": "正確性: {0}",
|
||||
"precision": "精度: {0}",
|
||||
"recall": "リコール: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "モデル全体の予測に影響を与える上位 k の特徴量の重要度を探索します。スライダーを使用して、特徴量の重要度を降順で表示します。"
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "これらの説明の詳細",
|
||||
"clickHere": "詳細情報",
|
||||
"shapTitle": "Shapley 値",
|
||||
"shapDescription": "この解説では、説明しているモデルへのゲーム論的アプローチである SHAP を使用しており、特徴セットの重要度は、モデルからマージナライゼイションによってそれらの特徴を「隠す」ことで測定されます。詳細については、下のリンクをクリックしてください。",
|
||||
"limeTitle": "LIME (ローカルの解釈可能なモデルにとらわれない説明)",
|
||||
"limeDescription": "この解説では、モデルの線形近似を提供する LIME を使用します。説明するために、次の操作を行います: インスタンスに摂動を与え、モデル予測を取得し、これらの予測をラベルとして使用して、局所的に正確な疎線形モデルを学習します。この線形モデルの重みは '特徴量の重要度' として使用されます。詳細については、下のリンクをクリックしてください。",
|
||||
"mimicTitle": "Mimic (グローバル サロゲートの説明)",
|
||||
"mimicDescription": "この解説は、ブラックボックス モデルの模倣のためにグローバル サロゲート モデルをトレーニングするという概念に基づいています。グローバル サロゲート モデルは、ブラックボックス モデルの予測を可能な限り正確に近似するようにトレーニングされた、本質的に解釈可能なモデルです。特徴量の重要度の値は、基になるサロゲート モデル (LightGBM、線形回帰、確率的勾配降下法、またはデシジョン ツリー) のモデルベースの特徴量の重要度の値です",
|
||||
"pfiTitle": "順列の特徴量の重要度 (PFI)",
|
||||
"pfiDescription": "この解説では、データセット全体に対して一度に 1 つの特徴量でデータをランダムにシャッフルし、重要なパフォーマンス メトリックがどのくらい変化するかを算出します (既定のパフォーマンス メトリック: バイナリ分類の F1、マルチクラス分類のミクロ平均を含む F1 スコア、および回帰の平均絶対誤差)。変化が大きいほど、その特徴量は重要になります。この解説で説明できるのは、基になるモデルの全体的な動作のみで、個々の予測については説明していません。特徴量の重要度の値は、特定の特徴量に摂動を与えることで、モデルのパフォーマンスのデルタを表します。"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "로컬 설명을 보려면 요소를 선택하세요.",
|
||||
"defaultClassNames": "클래스 {0}",
|
||||
"defaultFeatureNames": "기능 {0}",
|
||||
"absoluteAverage": "절대값의 평균",
|
||||
"predictedClass": "예측 클래스",
|
||||
"datasetExplorer": "데이터 세트 탐색기",
|
||||
"dataExploration": "데이터 세트 탐색",
|
||||
"aggregateFeatureImportance": "기능 중요도 집계",
|
||||
"globalImportance": "글로벌 중요도",
|
||||
"explanationExploration": "설명 탐색",
|
||||
"individualAndWhatIf": "개별 기능 중요도 및 가상",
|
||||
"summaryImportance": "요약 중요도",
|
||||
"featureImportance": "기능 중요도",
|
||||
"featureImportanceOf": "{0}의 기능 중요도",
|
||||
"perturbationExploration": "작은 변경 탐색",
|
||||
"localFeatureImportance": "로컬 기능 중요도",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "선택 영역 지우기",
|
||||
"feature": "기능:",
|
||||
"intercept": "가로채기",
|
||||
"modelPerformance": "모델 성능",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "데이터: {0}",
|
||||
"importanceLabel": "중요도: {0}",
|
||||
"predictedY": "예상 Y",
|
||||
"index": "인덱스",
|
||||
"dataGroupLabel": "데이터",
|
||||
"output": "출력",
|
||||
"probabilityLabel": "가능성: {0}",
|
||||
"trueY": "True Y",
|
||||
"class": "클래스: ",
|
||||
"xValue": "X 값:",
|
||||
"yValue": "Y 값:",
|
||||
"colorValue": "색:",
|
||||
"count": "개수"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "클래스 간 가중치:",
|
||||
"info": "클래스 간 계산에 대한 정보",
|
||||
"overviewInfo": "다중 클래스 모델은 각 클래스의 독립 기능 중요도 벡터를 생성합니다. 각 클래스의 기능 중요도 벡터는 클래스의 가능성을 높인 기능 또는 그 반대에 해당하는 기능을 보여 줍니다. 클래스별 기능 중요도 벡터의 가중치가 단일 값으로 요약되는 방법을 선택할 수 있습니다.",
|
||||
"absoluteValInfo": "절대값의 평균: 가능한 모든 클래스에서 기능의 중요도 합계를 클래스 수로 나눈 값을 표시합니다.",
|
||||
"predictedClassInfo": "예측 클래스: 지정된 요소의 예측 클래스에 해당하는 기능 중요도 값을 보여 줍니다.",
|
||||
"enumeratedClassInfo": "열거된 클래스 이름: 모든 데이터 요소에서 지정된 클래스의 기능 중요도 값만 표시합니다.",
|
||||
"close": "닫기",
|
||||
"crossClassWeights": "클래스 간 가중치"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "확장된 기능 값",
|
||||
"low": "저가",
|
||||
"high": "고가",
|
||||
"featureLabel": "기능: {0}",
|
||||
"valueLabel": "기능 값: {0}",
|
||||
"importanceLabel": "중요도: {0}",
|
||||
"predictedClassTooltip": "예측 클래스: {0}",
|
||||
"trueClassTooltip": "True 클래스: {0}",
|
||||
"predictedOutputTooltip": "예측 출력: {0}",
|
||||
"trueOutputTooltip": "True 출력: {0}",
|
||||
"topKFeatures": "상위 K개 기능:",
|
||||
"topKInfo": "상위 k개를 계산하는 방법",
|
||||
"predictedValue": "예측 값",
|
||||
"predictedClass": "예측 클래스",
|
||||
"trueValue": "True 값",
|
||||
"trueClass": "True 클래스",
|
||||
"noColor": "없음",
|
||||
"tooManyRows": "제공된 데이터 세트가 이 차트에서 지원되는 크기보다 큽니다."
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "클래스: {0}",
|
||||
"sortBy": "정렬 기준",
|
||||
"noData": "데이터 없음",
|
||||
"absoluteGlobal": "절대 글로벌",
|
||||
"absoluteLocal": "절대 로컬",
|
||||
"calculatingExplanation": "설명을 계산하는 중"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "숫자여야 합니다.",
|
||||
"integerError": "정수여야 합니다.",
|
||||
"prediction": "예측",
|
||||
"predictedProbability": "예측 가능성",
|
||||
"predictionLabel": "예측: {0}",
|
||||
"probabilityLabel": "가능성: {0}",
|
||||
"noModelError": "ICE 플롯에서 예측을 탐색하려면 조작 가능한 모델을 제공하세요.",
|
||||
"featurePickerLabel": "기능:",
|
||||
"minimumInputLabel": "최소:",
|
||||
"maximumInputLabel": "최대:",
|
||||
"stepInputLabel": "단계:",
|
||||
"loadingMessage": "데이터 로드 중...",
|
||||
"submitPrompt": "ICE 플롯을 보려면 범위를 제출하세요.",
|
||||
"topLevelErrorMessage": "매개 변수 오류",
|
||||
"errorPrefix": "오류 발생: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "로드 중...",
|
||||
"perturbationLabel": "작은 변경:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "예상 값: {0}",
|
||||
"predictedClassLabel": "예측 클래스: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "그룹화 안 함",
|
||||
"groupPredicted": "예상 Y",
|
||||
"groupTrue": "True Y",
|
||||
"groupBy": "그룹화 방법"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "차트 종류:",
|
||||
"violinText": "바이올린",
|
||||
"barText": "가로 막대형",
|
||||
"boxText": "상자",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "글로벌 기능 중요도는 모든 요소의 기능 중요도 절대값을 평균하여 계산됩니다(L1 정규화). ",
|
||||
"multiclassImportanceAddendum": "모든 요소는 모든 클래스의 기능 중요도를 계산하는 데 포함되며 차등 가중치가 사용되지 않습니다. 따라서 '클래스 A'가 아닌 것으로 예측된 많은 요소의 큰 음수 중요도가 있는 기능은 해당 기능의 '클래스 A' 중요도를 크게 높입니다."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "같음",
|
||||
"greaterThanComparison": "보다 큼",
|
||||
"greaterThanEqualToComparison": "크거나 같음",
|
||||
"lessThanComparison": "보다 작음",
|
||||
"lessThanEqualToComparison": "작거나 같음",
|
||||
"inTheRangeOf": "다음 범위 내에 있음",
|
||||
"categoricalIncludeValues": "포함된 값:",
|
||||
"numericValue": "값",
|
||||
"numericalComparison": "연산",
|
||||
"minimum": "최솟값",
|
||||
"maximum": "최댓값",
|
||||
"min": "최소: {0}",
|
||||
"max": "최대: {0}",
|
||||
"uniqueValues": "고유 값 개수: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "회귀 오류",
|
||||
"error": "오류",
|
||||
"classificationOutcome": "분류 결과",
|
||||
"truePositive": "진양성",
|
||||
"trueNegative": "참 부정",
|
||||
"falsePositive": "가양성",
|
||||
"falseNegative": "거짓 부정",
|
||||
"dataset": "데이터 세트",
|
||||
"predictedProbabilities": "예측 가능성",
|
||||
"none": "개수"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "닫기",
|
||||
"defaultCustomRootName": "행 {0}의 복사본",
|
||||
"filterFeaturePlaceholder": "기능 검색"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "코호트",
|
||||
"defaultLabel": "모든 데이터"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "전체 모델 예측에 영향을 주는 상위 k개의 중요한 기능을 살펴봅니다(예: 전역 설명). 내림차순 기능 중요도 값을 표시하려면 슬라이더를 사용합니다. 최대 3개의 코호트를 선택하여 해당 기능 중요도 값을 나란히 표시할 수 있습니다. 그래프에서 기능 막대를 클릭하여 선택한 기능의 값이 모델 예측에 영향을 주는 방식을 확인하세요.",
|
||||
"topAtoB": "상위 {0}~{1}개 기능",
|
||||
"datasetCohorts": "데이터 세트 코호트",
|
||||
"legendHelpText": "범례 항목을 클릭하여 플롯에서 코호트를 설정/해제합니다.",
|
||||
"sortBy": "정렬 기준",
|
||||
"viewDependencePlotFor": "다음의 종속 플롯 보기:",
|
||||
"datasetCohortSelector": "데이터 세트 코호트 선택",
|
||||
"aggregateFeatureImportance": "기능 중요도 집계",
|
||||
"missingParameters": "이 탭에서 로컬 기능 중요도 매개 변수를 제공해야 합니다.",
|
||||
"weightOptions": "클래스 중요도 가중치",
|
||||
"dependencePlotTitle": "종속 플롯",
|
||||
"dependencePlotHelperText": "이 종속 플롯은 코호트에서 기능 값과 해당 기능 중요도 간의 관계를 보여 줍니다.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "기능 선택",
|
||||
"datasetRequired": "종속 플롯에는 평가 데이터 세트 및 로컬 기능 중요도 배열이 필요합니다."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "데이터 통계",
|
||||
"datapoints": "{0}개 데이터 요소",
|
||||
"features": "{0}개 기능",
|
||||
"filters": "{0}개 필터",
|
||||
"binaryClassifier": "이진 분류자",
|
||||
"regressor": "회귀 변수",
|
||||
"multiclassClassifier": "다중 클래스 분류자",
|
||||
"datasetCohorts": "데이터 세트 코호트",
|
||||
"editCohort": "코호트 편집",
|
||||
"duplicateCohort": "코호트 복제",
|
||||
"addCohort": "코호트 추가",
|
||||
"copy": " 복사"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "예측 값의 분포와 모델 성능 메트릭 값을 탐색하여 모델의 성능을 평가합니다. 데이터 세트의 여러 코호트 또는 하위 그룹에 대한 성능 비교 분석을 확인하여 모델을 추가로 조사할 수 있습니다. 서로 다른 차원에서 잘라낼 y 값과 x 값을 따라 필터를 선택합니다. 그래프 형식을 변경하려면 그래프의 톱니바퀴를 선택합니다.",
|
||||
"modelStatistics": "모델 통계",
|
||||
"cohortPickerLabel": "탐색할 데이터 세트 코호트 선택",
|
||||
"missingParameters": "이 탭에서 모델의 예측 값 배열을 제공해야 합니다.",
|
||||
"missingTrueY": "모델 성능 통계를 사용하려면 예측 결과와 함께 true 결과가 제공되어야 합니다."
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y 값",
|
||||
"numberOfDatapoints": "데이터 요소 수",
|
||||
"xValue": "X 값",
|
||||
"rowIndex": "행 인덱스",
|
||||
"featureImportance": "기능 중요도",
|
||||
"countTooltipPrefix": "개수: {0}",
|
||||
"count": "개수 계산",
|
||||
"featurePrefix": "기능",
|
||||
"importancePrefix": "중요도",
|
||||
"cohort": "코호트",
|
||||
"howToRead": "이 차트를 읽는 방법"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "X, Y 및 색상 축을 따라 다른 필터를 선택하여 데이터를 다른 차원을 따라 분할함으로써 데이터 세트 통계를 살펴봅니다. 예측 결과, 데이터 세트 기능 및 오류 그룹과 같은 필터를 사용하여 데이터 세트 통계를 분석하려면 위의 데이터 세트 코호트를 만드세요. 그래프의 오른쪽 위 모서리에 있는 톱니 모양 아이콘을 사용하여 그래프 유형을 변경합니다.",
|
||||
"colorValue": "색 값",
|
||||
"individualDatapoints": "개별 데이터 요소",
|
||||
"aggregatePlots": "플롯 집계",
|
||||
"chartType": "차트 종류",
|
||||
"missingParameters": "이 탭에서 평가 데이터 세트를 제공해야 합니다.",
|
||||
"noColor": "없음"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "다음의 기능 중요도",
|
||||
"placeholder": "해당 종속 플롯을 표시하려면 위의 가로 막대형 차트에서 기능을 클릭"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "산점도를 클릭하여 데이터 요소를 선택한 후 아래에서 해당 로컬 기능 중요도 값(로컬 설명) 및 개별 ICE(조건부 기대 수준) 플롯을 볼 수 있습니다. 오른쪽에 있는 패널을 사용하여 알려진 데이터 요소의 기능을 약간 변경함으로써 가상 데이터 요소를 만듭니다. 기능 중요도 값은 많은 근사치를 기준으로 하며 예측의 \"원인\"이 아닙니다. 인과 유추의 엄격한 수학적 견고성이 구현되지 않으면 이 도구를 기준으로 하는 실질적인 의사 결정을 사용자에게 권유하지 않습니다.",
|
||||
"panelPlaceholder": "새 데이터 요소에 대해 예측을 수행하려면 모델이 필요합니다.",
|
||||
"cohortPickerLabel": "탐색할 데이터 세트 코호트 선택",
|
||||
"scatterLegendText": "범례 항목을 클릭하여 플롯에서 데이터 요소를 설정/해제합니다.",
|
||||
"realPoint": "실제 데이터 요소",
|
||||
"noneSelectedYet": "아직 선택한 항목 없음",
|
||||
"whatIfDatapoints": "가상 데이터 요소",
|
||||
"noneCreatedYet": "아직 만들어진 항목 없음",
|
||||
"showLabel": "표시:",
|
||||
"featureImportancePlot": "기능 중요도 플롯",
|
||||
"icePlot": "ICE(개별 조건부 예상) 플롯",
|
||||
"featureImportanceLackingParameters": "각 기능이 개별 예측에 영향을 주는 방식을 확인하려면 로컬 기능 중요도를 제공하세요.",
|
||||
"featureImportanceGetStartedText": "기능 중요도를 확인할 요소 선택",
|
||||
"iceLackingParameters": "가상 데이터 요소에 대해 예측하려면 ICE 플롯에 조작 가능한 모델이 필요합니다.",
|
||||
"IceGetStartedText": "ICE 플롯을 보려면 요소를 선택하거나 가상 요소를 만드세요.",
|
||||
"whatIfDatapoint": "가상 데이터 요소",
|
||||
"whatIfHelpText": "플롯에서 요소를 선택하거나 작은 변경에 대한 알려진 데이터 요소 인덱스를 수동으로 입력하고 새 가상 요소로 저장합니다.",
|
||||
"indexLabel": "작은 변경에 대한 데이터 인덱스",
|
||||
"rowLabel": "행 {0}",
|
||||
"whatIfNameLabel": "가상 데이터 요소 이름",
|
||||
"featureValues": "기능 값",
|
||||
"predictedClass": "예측 클래스: ",
|
||||
"predictedValue": "예측 값: ",
|
||||
"probability": "확률: ",
|
||||
"trueClass": "True 클래스: ",
|
||||
"trueValue": "True 값: ",
|
||||
"trueValue.comment": "회귀에 대한 실제 레이블의 접두사",
|
||||
"newPredictedClass": "새 예측 클래스: ",
|
||||
"newPredictedValue": "새 예측 값: ",
|
||||
"newProbability": "새 확률: ",
|
||||
"saveAsNewPoint": "새 요소로 저장",
|
||||
"saveChanges": "변경 내용 저장",
|
||||
"loading": "로드 중...",
|
||||
"classLabel": "클래스: {0}",
|
||||
"minLabel": "최소",
|
||||
"maxLabel": "최대",
|
||||
"stepsLabel": "단계",
|
||||
"disclaimer": "고지 사항: 이는 많은 근사치를 기반으로 한 설명이며 예측의 \"원인\"이 아닙니다. 인과 추론의 엄격한 수학적 견고성 없이는 이 도구를 기반으로 하여 실제 의사 결정을 하지 않는 것이 좋습니다.",
|
||||
"missingParameters": "이 탭에서 평가 데이터 세트를 제공해야 합니다.",
|
||||
"selectionLimit": "최대 3개 지점 선택",
|
||||
"classPickerLabel": "클래스",
|
||||
"tooltipTitleMany": "상위 {0}개의 예측 클래스",
|
||||
"whatIfTooltipTitle": "가상 예측 클래스",
|
||||
"tooltipTitleFew": "예측 클래스",
|
||||
"probabilityLabel": "확률",
|
||||
"deltaLabel": "델타",
|
||||
"nonNumericValue": "값은 숫자여야 합니다.",
|
||||
"icePlotHelperText": "ICE 플롯은 선택한 데이터 요소 예측 값이 최솟값과 최댓값 사이의 기능 값 범위를 따라 변경되는 방식을 보여 줍니다."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "필터 선택",
|
||||
"TreatAsCategorical": "범주로 처리",
|
||||
"addFilter": "필터 추가",
|
||||
"addedFilters": "추가된 필터",
|
||||
"noAddedFilters": "아직 추가된 필터 없음",
|
||||
"defaultFilterState": "데이터 세트 코호트에 매개 변수를 추가하려면 필터를 선택하세요.",
|
||||
"cohortNameLabel": "데이터 세트 코호트 이름",
|
||||
"cohortNamePlaceholder": "코호트 이름 지정",
|
||||
"save": "저장",
|
||||
"delete": "삭제",
|
||||
"cancel": "취소",
|
||||
"cohortNameError": "코호트 이름 누락",
|
||||
"placeholderName": "코호트 {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "선택",
|
||||
"ditherLabel": "디더링해야 함",
|
||||
"selectFilter": "축 값 선택",
|
||||
"selectFeature": "기능 선택",
|
||||
"binLabel": "데이터에 범주화 적용",
|
||||
"TreatAsCategorical": "범주로 처리",
|
||||
"numOfBins": "bin 수",
|
||||
"groupByCohort": "코호트로 그룹화",
|
||||
"selectClass": "클래스 선택",
|
||||
"countHelperText": "점 수의 히스토그램"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "예측 가능성",
|
||||
"predictedY": "예상 Y",
|
||||
"evalData": "평가 데이터 세트",
|
||||
"localFeatureImportance": "로컬 기능 중요도",
|
||||
"inconsistentDimensions": "차원이 일관되지 않습니다. {0}은(는) {1}차원인데, {2}차원이어야 합니다.",
|
||||
"notNonEmpty": "{0} 입력은 비어 있지 않은 배열이 아닙니다.",
|
||||
"varyingLength": "차원이 일관되지 않습니다. {0}에는 다양한 길이의 요소가 포함되어 있습니다.",
|
||||
"notArray": "{0}은(는) 배열이 아닙니다. {1}차원의 배열이 필요합니다.",
|
||||
"errorHeader": "일부 입력 매개 변수가 일관되지 않아 사용되지 않음: ",
|
||||
"datasizeWarning": "평가 데이터 세트가 너무 커서 일부 차트에 효과적으로 표시할 수 없습니다. 필터를 추가하여 코호트의 크기를 줄이세요. ",
|
||||
"datasizeError": "선택한 코호트가 너무 큽니다. 필터를 추가하여 코호트의 크기를 줄이세요.",
|
||||
"addFilters": "필터 추가"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " {0} 포함 ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} 외 {1}명"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R-제곱: {0}",
|
||||
"meanPrediction": "평균 예측 {0}",
|
||||
"accuracy": "정확도: {0}",
|
||||
"precision": "정밀도: {0}",
|
||||
"recall": "회수: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "전체 모델 예측에 영향을 주는 상위 k개의 중요한 기능을 살펴봅니다. 내림차순 기능 중요도를 표시하려면 슬라이더를 사용합니다."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "이러한 설명은 무엇을 의미하나요?",
|
||||
"clickHere": "자세한 정보",
|
||||
"shapTitle": "Shapley 값",
|
||||
"shapDescription": "이 설명자는 모델을 설명하는 게임 이론 방식인 SHAP를 사용합니다. 이 방법은 모델에서 무시를 통해 이러한 기능을 \"숨겨\" 기능 세트의 중요도를 측정합니다. 자세한 내용을 보려면 아래 링크를 클릭하세요.",
|
||||
"limeTitle": "LIME(로컬 해석 가능 모델 중립적 설명)",
|
||||
"limeDescription": "이 설명자는 모델의 선형 근사값을 제공하는 LIME을 사용합니다. 설명을 얻으려면 인스턴스를 약간 변경하고, 모델 예측을 가져온 다음, 이러한 예측을 레이블로 사용하여 로컬로 신뢰할 수 있는 스파스 선형 모델을 학습합니다. 이 선형 모델의 가중치는 '기능 중요도'로 사용됩니다. 자세한 내용을 보려면 아래 링크를 클릭하세요.",
|
||||
"mimicTitle": "모방(전역 서로게이트 설명)",
|
||||
"mimicDescription": "이 설명자는 전역 서로게이트 모델을 학습하여 블랙 박스 모델을 모방하는 개념을 기준으로 합니다. 전역 서로게이트 모델은 모든 블랙 박스 모델 예측 근사치를 가능한 한, 정확하게 구하도록 학습되는 본질적으로 해석 가능한 모델입니다. 기능 중요도 값은 기본 서로게이트 모델(LightGBM, 선형 회귀, 확률적 그라데이션 하강 또는 의사 결정 트리)의 모델 기반 기능 중요도 값입니다.",
|
||||
"pfiTitle": "PFI(순열 기능 중요도)",
|
||||
"pfiDescription": "이 설명자는 전체 데이터 세트에 대해 한 번에 임의로 데이터의 순서를 섞고 관심 분야 변경의 성능 메트릭(기본 성능 메트릭: 이진 분류의 경우 F1 점수, 다중 클래스 분류의 경우 마이크로 평균이 있는 F1 점수, 회귀의 경우 평균 절대 오차) 크기를 계산합니다. 변경 내용이 클수록 기능은 더 중요합니다. 이 설명자는 기본 모델의 전체 동작도 설명하지만 개별 예측은 설명하지 않습니다. 기능의 기능 중요도 값은 해당 특정 기능을 약간 변경하여 모델 성능의 델타를 나타냅니다."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,44 +0,0 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { getLocalization } from "@responsible-ai/mlchartlib";
|
||||
|
||||
import cs from "./cs-CZ.json";
|
||||
import de from "./de-DE.json";
|
||||
import en from "./en.json";
|
||||
import es from "./es-ES.json";
|
||||
import fr from "./fr-FR.json";
|
||||
import hu from "./hu-HU.json";
|
||||
import it from "./it-IT.json";
|
||||
import ja from "./ja-JP.json";
|
||||
import ko from "./ko-KR.json";
|
||||
import nl from "./nl-NL.json";
|
||||
import pl from "./pl-PL.json";
|
||||
import ptbr from "./pt-BR.json";
|
||||
import pt from "./pt-PT.json";
|
||||
import ru from "./ru-RU.json";
|
||||
import sv from "./sv-SE.json";
|
||||
import tr from "./tr-TR.json";
|
||||
import zhcn from "./zh-CN.json";
|
||||
import zhtw from "./zh-TW.json";
|
||||
|
||||
export const localization = getLocalization({
|
||||
cs,
|
||||
de,
|
||||
en,
|
||||
es,
|
||||
fr,
|
||||
hu,
|
||||
it,
|
||||
ja,
|
||||
ko,
|
||||
nl,
|
||||
pl,
|
||||
"pt-BR": ptbr,
|
||||
"pt-PT": pt,
|
||||
ru,
|
||||
sv,
|
||||
tr,
|
||||
"zh-CN": zhcn,
|
||||
"zh-TW": zhtw
|
||||
});
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Selecteer een punt om de lokale uitleg ervan weer te geven",
|
||||
"defaultClassNames": "Klasse {0}",
|
||||
"defaultFeatureNames": "Functie {0}",
|
||||
"absoluteAverage": "Gemiddelde van absolute waarde",
|
||||
"predictedClass": "Voorspelde klasse",
|
||||
"datasetExplorer": "Gegevenssetverkenner",
|
||||
"dataExploration": "Verkenning van gegevensset",
|
||||
"aggregateFeatureImportance": "Belang van de functie Samenvoegen",
|
||||
"globalImportance": "Algemene urgentie",
|
||||
"explanationExploration": "Uitlegverkenning",
|
||||
"individualAndWhatIf": "Belang van afzonderlijke functies en What-If",
|
||||
"summaryImportance": "Urgentiesamenvatting",
|
||||
"featureImportance": "Functie-urgentie",
|
||||
"featureImportanceOf": "Belang van de functie van {0}",
|
||||
"perturbationExploration": "Storingsverkenning",
|
||||
"localFeatureImportance": "Lokale functie-urgentie",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Selectie wissen",
|
||||
"feature": "Functie:",
|
||||
"intercept": "Snijpunt",
|
||||
"modelPerformance": "Modelprestaties",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Gegevens: {0}",
|
||||
"importanceLabel": "Urgentie: {0}",
|
||||
"predictedY": "Voorspelde Y",
|
||||
"index": "Index",
|
||||
"dataGroupLabel": "Gegevens",
|
||||
"output": "Uitvoer",
|
||||
"probabilityLabel": "Waarschijnlijkheid: {0}",
|
||||
"trueY": "Ware Y",
|
||||
"class": "klasse: ",
|
||||
"xValue": "X-waarde:",
|
||||
"yValue": "Y-waarde:",
|
||||
"colorValue": "Kleur:",
|
||||
"count": "Aantal"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Weging voor alle klassen:",
|
||||
"info": "Informatie over de berekening van meerdere klassen",
|
||||
"overviewInfo": "Modellen met meerdere klassen genereren een onafhankelijke functie-urgentievector voor elke klasse. De functie-urgentievector van elke klasse geeft aan welke functies een klasse waarschijnlijk of minder waarschijnlijk hebben gemaakt. U kunt selecteren hoe de wegingen van de functie-urgentievector per klasse wordt samengevat in één waarde:",
|
||||
"absoluteValInfo": "Gemiddelde van absolute waarde: hiermee wordt de som van de urgentie van de functie weergegeven voor alle mogelijke klassen, gedeeld door het aantal klassen",
|
||||
"predictedClassInfo": "Voorspelde klasse: hiermee wordt de functie-urgentiewaarde weergegeven voor de voorspelde klasse van een bepaald punt",
|
||||
"enumeratedClassInfo": "Geïnventariseerde klassenamen: hiermee worden alleen de functie-urgentiewaarden van de opgegeven klasse weergegeven voor alle gegevenspunten.",
|
||||
"close": "Sluiten",
|
||||
"crossClassWeights": "Gewichten in meerdere klassen"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Geschaalde functiewaarde",
|
||||
"low": "Laag",
|
||||
"high": "Hoog",
|
||||
"featureLabel": "Functie: {0}",
|
||||
"valueLabel": "Functiewaarde: {0}",
|
||||
"importanceLabel": "Urgentie: {0}",
|
||||
"predictedClassTooltip": "Voorspelde klasse: {0}",
|
||||
"trueClassTooltip": "Ware klasse: {0}",
|
||||
"predictedOutputTooltip": "Voorspelde uitvoer: {0}",
|
||||
"trueOutputTooltip": "Ware uitvoer: {0}",
|
||||
"topKFeatures": "Belangrijkste K-functies:",
|
||||
"topKInfo": "Zo wordt de top k berekend",
|
||||
"predictedValue": "Voorspelde waarde",
|
||||
"predictedClass": "Voorspelde klasse",
|
||||
"trueValue": "Ware waarde",
|
||||
"trueClass": "Ware klasse",
|
||||
"noColor": "Geen",
|
||||
"tooManyRows": "De opgegeven gegevensset is groter dan door deze grafiek wordt ondersteund"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Klasse: {0}",
|
||||
"sortBy": "Sorteren op",
|
||||
"noData": "Geen gegevens",
|
||||
"absoluteGlobal": "Absoluut algemeen",
|
||||
"absoluteLocal": "Absoluut lokaal",
|
||||
"calculatingExplanation": "Berekeningsuitleg"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Moet numeriek zijn",
|
||||
"integerError": "Moet een geheel getal zijn",
|
||||
"prediction": "Voorspelling",
|
||||
"predictedProbability": "Voorspelde waarschijnlijkheid",
|
||||
"predictionLabel": "Voorspelling: {0}",
|
||||
"probabilityLabel": "Waarschijnlijkheid: {0}",
|
||||
"noModelError": "Geef een operationeel model op om voorspellingen te verkennen in ICE-tekeningen.",
|
||||
"featurePickerLabel": "Functie:",
|
||||
"minimumInputLabel": "Minimum:",
|
||||
"maximumInputLabel": "Maximum:",
|
||||
"stepInputLabel": "Stappen:",
|
||||
"loadingMessage": "Gegevens laden...",
|
||||
"submitPrompt": "Een bereik verzenden om een ICE-tekening weer te geven",
|
||||
"topLevelErrorMessage": "Fout in parameter",
|
||||
"errorPrefix": "Fout aangetroffen: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Laden...",
|
||||
"perturbationLabel": "Storing:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Voorspelde waarde: {0}",
|
||||
"predictedClassLabel": "Voorspelde klasse: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Geen groepering",
|
||||
"groupPredicted": "Voorspelde Y",
|
||||
"groupTrue": "Ware Y",
|
||||
"groupBy": "Groeperen op"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Grafiektype:",
|
||||
"violinText": "Violin",
|
||||
"barText": "Staaf",
|
||||
"boxText": "Vak",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "De algemene functie-urgentie wordt berekend door het gemiddelde te berekenen van de absolute waarde van de functie-urgentie van alle punten (L1-normalisatie). ",
|
||||
"multiclassImportanceAddendum": "Alle punten worden opgenomen in de berekening van de urgentie van een functie voor alle klassen. Er wordt geen differentiële weging gebruikt. Een functie met een grote negatieve urgentie voor veel punten die volgens de voorspelling niet van klasse A zijn, verhoogt het belang van de klasse A-urgentie van die functie aanzienlijk."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Is gelijk aan",
|
||||
"greaterThanComparison": "Groter dan",
|
||||
"greaterThanEqualToComparison": "Groter dan of gelijk aan",
|
||||
"lessThanComparison": "Kleiner dan",
|
||||
"lessThanEqualToComparison": "Kleiner dan of gelijk aan",
|
||||
"inTheRangeOf": "Binnen het bereik van",
|
||||
"categoricalIncludeValues": "Inbegrepen waarden:",
|
||||
"numericValue": "Waarde",
|
||||
"numericalComparison": "Bewerking",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Maximum",
|
||||
"min": "Min: {0}",
|
||||
"max": "Max: {0}",
|
||||
"uniqueValues": "aantal unieke waarden: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Regressiefout",
|
||||
"error": "Fout",
|
||||
"classificationOutcome": "Classificatieresultaten",
|
||||
"truePositive": "Terecht-positief",
|
||||
"trueNegative": "Terecht-negatief",
|
||||
"falsePositive": "Fout-positief",
|
||||
"falseNegative": "Fout-negatief",
|
||||
"dataset": "Gegevensset",
|
||||
"predictedProbabilities": "Voorspellingskansen",
|
||||
"none": "Aantal"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Sluiten",
|
||||
"defaultCustomRootName": "Kopie van rij {0}",
|
||||
"filterFeaturePlaceholder": "Zoeken naar functies"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Cohort",
|
||||
"defaultLabel": "Alle gegevens"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Verken de k belangrijkste functies die invloed hebben op uw algemene modelvoorspellingen (ofwel algemene verklaring). Gebruik de schuifregelaar om waarden voor functie-urgentie aflopend weer te geven. Selecteer maximaal drie cohorten om de bijbehorende waarden voor functie-urgentie naast elkaar weer te geven. Klik op een functiebalk in de grafiek om te bekijken hoe waarden van de geselecteerde functie invloed hebben op de modelvoorspelling.",
|
||||
"topAtoB": "Belangrijkste {0}-{1} functies",
|
||||
"datasetCohorts": "Gegevenssetcohorten",
|
||||
"legendHelpText": "Schakel cohorten in en uit in de plot door op de legenda-items te klikken.",
|
||||
"sortBy": "Sorteren op",
|
||||
"viewDependencePlotFor": "Plot met afhankelijkheden weergeven voor:",
|
||||
"datasetCohortSelector": "Een gegevenssetcohort selecteren",
|
||||
"aggregateFeatureImportance": "Belang van de functie Samenvoegen",
|
||||
"missingParameters": "Op dit tabblad moet de parameter voor het lokale functiebelang worden opgegeven.",
|
||||
"weightOptions": "Gewichten voor urgentie van klassen",
|
||||
"dependencePlotTitle": "Plots met afhankelijkheden",
|
||||
"dependencePlotHelperText": "In deze plot met afhankelijkheden wordt de relatie tussen de waarde van een functie en het bijbehorende belang van de functie in een cohort weergegeven.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Functie selecteren",
|
||||
"datasetRequired": "Plots met afhankelijkheden vereisen de evaluatiegegevensset en de prioriteitsmatrix van lokale functies."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Gegevensstatistieken",
|
||||
"datapoints": "{0} gegevenspunten",
|
||||
"features": "Functies van {0}",
|
||||
"filters": "{0} filters",
|
||||
"binaryClassifier": "Binaire classificatie",
|
||||
"regressor": "Regressor",
|
||||
"multiclassClassifier": "Classificatie voor meerdere klassen",
|
||||
"datasetCohorts": "Gegevenssetcohorten",
|
||||
"editCohort": "Cohort bewerken",
|
||||
"duplicateCohort": "Dubbele cohort",
|
||||
"addCohort": "Cohort toevoegen",
|
||||
"copy": " kopiëren"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Evalueer de prestaties van uw model door de distributie van uw voorspellingswaarden en de waarden van uw metrische prestatiegegevens van het model te verkennen. U kunt uw model nog verder onderzoeken door naar de vergelijkende analyse van de prestaties van verschillende cohorten of subgroepen van uw gegevensset te kijken. Selecteer filters voor de y- en de x-waarde om een beeld van verschillende dimensies te krijgen. Selecteer het tandwieltje in de grafiek om het grafiektype te wijzigen.",
|
||||
"modelStatistics": "Modelstatistieken",
|
||||
"cohortPickerLabel": "Een gegevenssetcohort selecteren om te verkennen",
|
||||
"missingParameters": "Op dit tabblad moet de matrix van voorspelde waarden uit het model worden opgegeven.",
|
||||
"missingTrueY": "Prestatiestatistieken voor modellen moeten naast de voorspelde resultaten ook de echte resultaten weergeven"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y-waarde",
|
||||
"numberOfDatapoints": "Aantal gegevenspunten",
|
||||
"xValue": "X-waarde",
|
||||
"rowIndex": "Rij-index",
|
||||
"featureImportance": "Belang van de functie",
|
||||
"countTooltipPrefix": "Aantal: {0}",
|
||||
"count": "Aantal",
|
||||
"featurePrefix": "Functie",
|
||||
"importancePrefix": "Urgentie",
|
||||
"cohort": "Cohort",
|
||||
"howToRead": "Hoe u deze grafiek moet lezen"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Verken de statistieken van uw gegevensset door verschillende filters langs de X-, Y- en kleuras te selecteren om uw gegevens te segmenteren in verschillende dimensies. Maak hierboven gegevenssetcohorten om statistieken van de gegevensset te analyseren met filters zoals voorspelde resultaten, gegevenssetfuncties en foutgroepen. Gebruik het tandwielpictogram in de rechterbovenhoek van de grafiek om grafiektypen te wijzigen.",
|
||||
"colorValue": "Kleurwaarde",
|
||||
"individualDatapoints": "Afzonderlijke gegevenspunten",
|
||||
"aggregatePlots": "Plots samenvoegen",
|
||||
"chartType": "Grafiektype",
|
||||
"missingParameters": "Op dit tabblad moet een evaluatiegegevensset worden opgegeven.",
|
||||
"noColor": "Geen"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Belang van de functie van",
|
||||
"placeholder": "Klik op een functie in het bovenstaande staafdiagram om de bijbehorende plot met afhankelijkheden weer te geven"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "U kunt een gegevenspunt selecteren door op het spreidingsdiagram te klikken om de bijbehorende waarden voor lokale functie-urgentie (lokale uitleg) en ICE (afzonderlijke voorwaardelijke verwachting) hieronder weer te geven. Maak een hypothetisch what-if-gegevenspunt door het deelvenster aan de rechterkant te gebruiken om functies van een bekend gegevenspunt te verstoren. De waarden voor functie-urgentie zijn op vele benaderingen gebaseerd en niet de 'oorzaak' van voorspellingen. Zonder een strikte wiskundige onderbouwing van causale interferentie is het niet raadzaam dat gebruikers echte beslissingen nemen op basis van dit hulpprogramma.",
|
||||
"panelPlaceholder": "Een model is vereist om voorspellingen te maken voor nieuwe gegevenspunten.",
|
||||
"cohortPickerLabel": "Een gegevenssetcohort selecteren om te verkennen",
|
||||
"scatterLegendText": "Schakel gegevenspunten in en uit in de plot door op de legenda-items te klikken.",
|
||||
"realPoint": "Werkelijke gegevenspunten",
|
||||
"noneSelectedYet": "Nog geen geselecteerd",
|
||||
"whatIfDatapoints": "What-If-gegevenspunten",
|
||||
"noneCreatedYet": "Nog geen gemaakt",
|
||||
"showLabel": "Weergeven:",
|
||||
"featureImportancePlot": "Plot met belang van functies",
|
||||
"icePlot": "ICE-tekenbewerking (Individual Conditional Expectation)",
|
||||
"featureImportanceLackingParameters": "Geef het belang van de lokale functie op om te zien hoe elke functie invloed heeft op afzonderlijke voorspellingen.",
|
||||
"featureImportanceGetStartedText": "Selecteer een punt om het belang van de functie weer te geven",
|
||||
"iceLackingParameters": "Voor ICE-grafieken is een operationeel model vereist om voorspellingen te maken voor hypothetische gegevenspunten.",
|
||||
"IceGetStartedText": "Een punt selecteren of een What-If-punt maken om ICE-plots weer te geven",
|
||||
"whatIfDatapoint": "What-If-gegevenspunt",
|
||||
"whatIfHelpText": "Selecteer een punt in de plot of voer handmatig een bekende gegevenspuntindex voor verstoren en om als een nieuw What-If-punt op te slaan.",
|
||||
"indexLabel": "Gegevensindex die moet worden verstoord",
|
||||
"rowLabel": "Rij {0}",
|
||||
"whatIfNameLabel": "Naam van What-If-gegevenspunt",
|
||||
"featureValues": "Functiewaarden",
|
||||
"predictedClass": "Voorspelde klasse: ",
|
||||
"predictedValue": "Voorspelde waarde: ",
|
||||
"probability": "Waarschijnlijkheid: ",
|
||||
"trueClass": "Werkelijke klasse: ",
|
||||
"trueValue": "Werkelijke waarde: ",
|
||||
"trueValue.comment": "voorvoegsel van werkelijk label voor regressie",
|
||||
"newPredictedClass": "Nieuwe voorspelde klasse: ",
|
||||
"newPredictedValue": "Nieuwe voorspelde waarde: ",
|
||||
"newProbability": "Nieuwe waarschijnlijkheid: ",
|
||||
"saveAsNewPoint": "Opslaan als nieuw punt",
|
||||
"saveChanges": "Wijzigingen opslaan",
|
||||
"loading": "Laden...",
|
||||
"classLabel": "Klasse: {0}",
|
||||
"minLabel": "Minimum",
|
||||
"maxLabel": "Maximum",
|
||||
"stepsLabel": "Stappen",
|
||||
"disclaimer": "Disclaimer: dit zijn verklaringen die op vele schattingen zijn gebaseerd en niet de 'oorzaak' van voorspellingen. Zonder een strikte wiskundige onderbouwing van causale interferentie is het niet raadzaam dat gebruikers echte beslissingen nemen op basis van dit hulpprogramma.",
|
||||
"missingParameters": "Op dit tabblad moet een evaluatiegegevensset worden opgegeven.",
|
||||
"selectionLimit": "Maximaal drie geselecteerde punten",
|
||||
"classPickerLabel": "Klasse",
|
||||
"tooltipTitleMany": "Belangrijkste {0} voorspelde klassen",
|
||||
"whatIfTooltipTitle": "Voorspelde what if-klassen",
|
||||
"tooltipTitleFew": "Voorspelde klassen",
|
||||
"probabilityLabel": "Kans",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "Waarde moet numeriek zijn",
|
||||
"icePlotHelperText": "Met ICE-spreidingsdiagrammen wordt aangegeven hoe de voorspellingswaarden van het geselecteerde gegevenspunt worden gewijzigd langs een aantal functiewaarden tussen een minimum- en maximumwaarde."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Filter selecteren",
|
||||
"TreatAsCategorical": "Beschouwen als categorisch",
|
||||
"addFilter": "Filter toevoegen",
|
||||
"addedFilters": "Toegevoegde filters",
|
||||
"noAddedFilters": "Nog geen filters toegevoegd",
|
||||
"defaultFilterState": "Selecteer een filter om parameters aan uw gegevenssetcohort toe te voegen.",
|
||||
"cohortNameLabel": "Naam van de gegevenssetcohort",
|
||||
"cohortNamePlaceholder": "Een naam voor uw cohort invoeren",
|
||||
"save": "Opslaan",
|
||||
"delete": "Verwijderen",
|
||||
"cancel": "Annuleren",
|
||||
"cohortNameError": "Cohortnaam ontbreekt",
|
||||
"placeholderName": "Cohort {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Selecteren",
|
||||
"ditherLabel": "Moet op raster worden weergegeven",
|
||||
"selectFilter": "Uw aswaarde selecteren",
|
||||
"selectFeature": "Functie selecteren",
|
||||
"binLabel": "Binning toepassen op gegevens",
|
||||
"TreatAsCategorical": "Beschouwen als categorisch",
|
||||
"numOfBins": "Aantal opslaglocaties",
|
||||
"groupByCohort": "Groeperen op cohort",
|
||||
"selectClass": "Klasse selecteren",
|
||||
"countHelperText": "Een histogram van het aantal punten"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Voorspelde waarschijnlijkheid",
|
||||
"predictedY": "Voorspelde Y",
|
||||
"evalData": "Evaluatiegegevensset",
|
||||
"localFeatureImportance": "Belang van lokale functie",
|
||||
"inconsistentDimensions": "Inconsistente dimensies. {0} heeft dimensies {1}, verwacht {2}",
|
||||
"notNonEmpty": "{0} invoer is geen niet-lege matrix",
|
||||
"varyingLength": "Inconsistente dimensies. {0} heeft elementen van verschillende lengte",
|
||||
"notArray": "{0} geen matrix. Er wordt een matrix van de dimensie {1} verwacht",
|
||||
"errorHeader": "Sommige invoerparameters zijn inconsistent en worden niet gebruikt: ",
|
||||
"datasizeWarning": "De gegevensset voor evaluatie is te groot om in sommige grafieken effectief te kunnen worden weergegeven. Voeg filters toe om de grootte van de cohort te verkleinen. ",
|
||||
"datasizeError": "De geselecteerde cohort is te groot. Voeg filters toe om de grootte van de cohort te verkleinen.",
|
||||
"addFilters": "Filters toevoegen"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " bevat {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} en {1} anderen"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R-kwadraat: {0}",
|
||||
"meanPrediction": "Gemiddelde voorspelling {0}",
|
||||
"accuracy": "Nauwkeurigheid: {0}",
|
||||
"precision": "Precisie: {0}",
|
||||
"recall": "Terughalen: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Verken de k belangrijkste functies die invloed hebben op uw algemene modelvoorspellingen. Gebruik de schuifbalk om functies van aflopend belang weer te geven."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Wat betekenen deze toelichtingen?",
|
||||
"clickHere": "Meer informatie",
|
||||
"shapTitle": "Shap-waarden",
|
||||
"shapDescription": "Deze verklarende factor maakt gebruik van SHAP, een speltheoretische benadering voor het uitleggen van modellen, waarbij de betekenis van de functiesets wordt gemeten door deze functies te 'verbergen' voor het model via marginalisatie. Klik op de onderstaande koppeling voor meer informatie.",
|
||||
"limeTitle": "LIME (Local Interpretable Model-Agnostic Explanations)",
|
||||
"limeDescription": "Deze verklarende factor maakt gebruik van LIME, waarmee een lineaire benadering van het model wordt geleverd. Een uitleg wordt verkregen door de instantie te verstoren, modelvoorspellingen op te halen en deze voorspellingen te gebruiken als labels om een eenvoudig lineair model te verkrijgen dat lokaal getrouw is. De gewichten van dit lineaire model wordt gebruikt als functie-urgenties. Klik op de onderstaande koppeling voor meer informatie.",
|
||||
"mimicTitle": "Nabootsen (verklaringen met globale surrogaatmodellen)",
|
||||
"mimicDescription": "Deze verklarende factor is gebaseerd op het idee van het trainen van globale surrogaatmodellen om black box-modellen na te bootsen. Een globaal surrogaatmodel is een intrinsiek interpreteerbaar model dat is getraind om de voorspellingen van elk black box-model zo nauwkeurig mogelijk te benaderen. De waarden voor functie-urgentie zijn modelwaarden voor functie-urgentie van het onderliggende surrogaatmodel (LightGBM, lineaire regressie, Stochastic Gradient Descent of beslissingsstructuur)",
|
||||
"pfiTitle": "PFI (Permutation Feature Importance)",
|
||||
"pfiDescription": "Met deze verklarende factor worden per functie de gegevens van een volledige gegevensset willekeurig gebruikt en wordt berekend in welke mate de metrische prestatiegegevens voor het belang veranderen (standaardwaarden voor metrische prestatiegegevens: F1 voor binaire classificatie, F1-score met microgemiddelde voor classificatie met meerdere klassen en gemiddelde absolute fout voor regressie). Hoe groter de verandering, hoe belangrijker de functie. Deze verklarende factor kan alleen het algemene gedrag van het onderliggende model uitleggen. Er worden geen afzonderlijke voorspellingen uitgelegd. De waarde voor functie-urgentie van een functie geeft het verschil in de prestatie van het model aan wanneer die specifieke functie wordt verstoord."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Wybierz punkt, aby zobaczyć jego lokalne wyjaśnienie",
|
||||
"defaultClassNames": "Klasa {0}",
|
||||
"defaultFeatureNames": "Cecha {0}",
|
||||
"absoluteAverage": "Średnia wartości bezwzględnej",
|
||||
"predictedClass": "Przewidywana klasa",
|
||||
"datasetExplorer": "Eksplorator zestawu danych",
|
||||
"dataExploration": "Eksploracja zestawu danych",
|
||||
"aggregateFeatureImportance": "Ważność funkcji agregującej",
|
||||
"globalImportance": "Ważność globalna",
|
||||
"explanationExploration": "Eksploracja wyjaśnienia",
|
||||
"individualAndWhatIf": "Ważność poszczególnych funkcji i analiza warunkowa",
|
||||
"summaryImportance": "Ważność podsumowania",
|
||||
"featureImportance": "Istotność funkcji",
|
||||
"featureImportanceOf": "Ważność funkcji {0}",
|
||||
"perturbationExploration": "Eksploracja perturbacji",
|
||||
"localFeatureImportance": "Lokalna istotność funkcji",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Wyczyść zaznaczenie",
|
||||
"feature": "Funkcja:",
|
||||
"intercept": "Przecięcie",
|
||||
"modelPerformance": "Wydajność modelu",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Dane: {0}",
|
||||
"importanceLabel": "Ważność: {0}",
|
||||
"predictedY": "Przewidywana wartość Y",
|
||||
"index": "Indeks",
|
||||
"dataGroupLabel": "Dane",
|
||||
"output": "Wyjściowe",
|
||||
"probabilityLabel": "Prawdopodobieństwo: {0}",
|
||||
"trueY": "Prawda Y",
|
||||
"class": "klasa:",
|
||||
"xValue": "Wartość X:",
|
||||
"yValue": "Wartość Y:",
|
||||
"colorValue": "Kolor:",
|
||||
"count": "Liczba"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Waga wieloklasowa:",
|
||||
"info": "Informacje o obliczeniach obejmujących wiele klas",
|
||||
"overviewInfo": "Modele wieloklasowe generują dla każdej klasy niezależny wektor istotności funkcji. Wektor istotności funkcji każdej klasy wskazuje, które funkcje powodują, że klasa jest mniej lub bardziej prawdopodobna. Możesz wybrać, w jaki sposób wagi wektorów istotności funkcji każdej klasy są podsumowywane do pojedynczej wartości:",
|
||||
"absoluteValInfo": "Średnia wartości bezwzględnej: pokazuje sumę ważności cechy dla wszystkich możliwych klas podzieloną przez liczbę klas",
|
||||
"predictedClassInfo": "Przewidywana klasa: pokazuje wartość istotności funkcji dla przewidywanej klasy danego punktu",
|
||||
"enumeratedClassInfo": "Wyliczane nazwy klas: pokazuje tylko określone wartości istotności funkcji klasy dla wszystkich punktów danych.",
|
||||
"close": "Zamknij",
|
||||
"crossClassWeights": "Wagi między klasami"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Skalowana wartości cechy",
|
||||
"low": "Niskie",
|
||||
"high": "Wysoki",
|
||||
"featureLabel": "Cecha: {0}",
|
||||
"valueLabel": "Wartość cechy: {0}",
|
||||
"importanceLabel": "Ważność: {0}",
|
||||
"predictedClassTooltip": "Przewidywana klasa: {0}",
|
||||
"trueClassTooltip": "Prawdziwa klasa: {0}",
|
||||
"predictedOutputTooltip": "Przewidywane dane wyjściowe: {0}",
|
||||
"trueOutputTooltip": "Prawdziwe dane wyjściowe: {0}",
|
||||
"topKFeatures": "Cechy Top K:",
|
||||
"topKInfo": "Jak jest obliczana najwyższa wartość k",
|
||||
"predictedValue": "Wartość prognozowana",
|
||||
"predictedClass": "Przewidywana klasa",
|
||||
"trueValue": "Wartość Prawda",
|
||||
"trueClass": "Prawdziwa klasa",
|
||||
"noColor": "Brak",
|
||||
"tooManyRows": "Podany zestaw danych jest większy niż ten wykres może obsłużyć"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Klasa: {0}",
|
||||
"sortBy": "Sortuj według",
|
||||
"noData": "Brak danych",
|
||||
"absoluteGlobal": "Bezwzględne globalnie",
|
||||
"absoluteLocal": "Bezwzględne lokalnie",
|
||||
"calculatingExplanation": "Obliczanie wyjaśnienia"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Musi być liczbą",
|
||||
"integerError": "Musi być liczbą całkowitą",
|
||||
"prediction": "Przewidywanie",
|
||||
"predictedProbability": "Przewidywane prawdopodobieństwo",
|
||||
"predictionLabel": "Przewidywanie: {0}",
|
||||
"probabilityLabel": "Prawdopodobieństwo: {0}",
|
||||
"noModelError": "Podaj funkcjonujący model, aby eksplorować przewidywania na wykresach ICE.",
|
||||
"featurePickerLabel": "Funkcja:",
|
||||
"minimumInputLabel": "Minimum:",
|
||||
"maximumInputLabel": "Maksimum:",
|
||||
"stepInputLabel": "Kroki:",
|
||||
"loadingMessage": "Trwa ładowanie danych...",
|
||||
"submitPrompt": "Prześlij zakres, aby wyświetlić wykres ICE",
|
||||
"topLevelErrorMessage": "Błąd w parametrze",
|
||||
"errorPrefix": "Wystąpił błąd: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Trwa ładowanie...",
|
||||
"perturbationLabel": "Perturbacja:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Przewidywana wartość: {0}",
|
||||
"predictedClassLabel": "Przewidywana klasa: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Brak grupowania",
|
||||
"groupPredicted": "Przewidywane Y",
|
||||
"groupTrue": "Prawda Y",
|
||||
"groupBy": "Grupuj według"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Typ wykresu:",
|
||||
"violinText": "Violin",
|
||||
"barText": "Słupkowy",
|
||||
"boxText": "Pole",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "Globalna istotność funkcji jest obliczana przez wyliczenie średniej wartości bezwzględnej istotności funkcji we wszystkich punktach (normalizacja L1). ",
|
||||
"multiclassImportanceAddendum": "Wszystkie punkty są uwzględniane przy obliczaniu ważności cechy dla wszystkich klas, nie są używane żadne wagi różnicujące. W ten sposób cecha, która ma duże negatywne znaczenie dla wielu punktów przewidywanych jako nienależące do klasy „Klasa A”, znacznie zwiększy znaczenie tej cechy w obrębie klasy „Klasa A”."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Równe",
|
||||
"greaterThanComparison": "Większe niż",
|
||||
"greaterThanEqualToComparison": "Większe niż lub równe",
|
||||
"lessThanComparison": "Mniejsze niż",
|
||||
"lessThanEqualToComparison": "Mniejsze niż lub równe",
|
||||
"inTheRangeOf": "W zakresie",
|
||||
"categoricalIncludeValues": "Uwzględnione wartości:",
|
||||
"numericValue": "Wartość",
|
||||
"numericalComparison": "Operacja",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Maksimum",
|
||||
"min": "Min. {0}",
|
||||
"max": "Maks. {0}",
|
||||
"uniqueValues": "Liczba unikatowych wartości: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Błąd regresji",
|
||||
"error": "Błąd",
|
||||
"classificationOutcome": "Wynik klasyfikacji",
|
||||
"truePositive": "Wynik prawdziwie dodatni",
|
||||
"trueNegative": "Wynik prawdziwie ujemny",
|
||||
"falsePositive": "Wynik fałszywie dodatni",
|
||||
"falseNegative": "Wynik fałszywie ujemny",
|
||||
"dataset": "Zestaw danych",
|
||||
"predictedProbabilities": "Prawdopodobieństwa przewidywania",
|
||||
"none": "Liczba"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Zamknij",
|
||||
"defaultCustomRootName": "Kopia wiersza {0}",
|
||||
"filterFeaturePlaceholder": "Wyszukaj funkcje"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Kohorta",
|
||||
"defaultLabel": "Wszystkie dane"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Poznaj pierwsze k najważniejszych cech, które mają wpływ na ogólne przewidywania modelu (inaczej nazywane wyjaśnieniem globalnym). Za pomocą suwaka możesz wyświetlać wartości ważności cech w kolejności malejącej. Wybierz maksymalnie trzy kohorty, aby zobaczyć ich wartości ważności cech obok siebie. Kliknij dowolny pasek cechy na wykresie, aby zobaczyć, jak wartości wybranej cechy wpływają na przewidywania modelu.",
|
||||
"topAtoB": "Najważniejsze funkcje: {0}–{1}",
|
||||
"datasetCohorts": "Kohorty zestawów danych",
|
||||
"legendHelpText": "Włącz lub wyłącz kohorty na wykresie, klikając elementy legendy.",
|
||||
"sortBy": "Sortuj według",
|
||||
"viewDependencePlotFor": "Wyświetl wykres zależności dla:",
|
||||
"datasetCohortSelector": "Wybierz kohortę zestawów danych",
|
||||
"aggregateFeatureImportance": "Ważność funkcji agregującej",
|
||||
"missingParameters": "Na tej karcie musi być podany parametr ważności cechy lokalnej.",
|
||||
"weightOptions": "Wagi ważności klasy",
|
||||
"dependencePlotTitle": "Wykresy zależności",
|
||||
"dependencePlotHelperText": "Ten wykres zależności pokazuje relację między wartością cechy a odpowiadającą jej ważnością cechy w kohorcie.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Wybierz cechę",
|
||||
"datasetRequired": "Wykresy zależności wymagają zestawu danych oceny i tablicy ważności cech lokalnych."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Statystyki danych",
|
||||
"datapoints": "Punkty danych: {0}",
|
||||
"features": "Funkcje: {0}",
|
||||
"filters": "Filtry: {0}",
|
||||
"binaryClassifier": "Klasyfikator binarny",
|
||||
"regressor": "Regresor",
|
||||
"multiclassClassifier": "Klasyfikator wieloklasowy",
|
||||
"datasetCohorts": "Kohorty zestawów danych ",
|
||||
"editCohort": "Edytuj kohortę",
|
||||
"duplicateCohort": "Zduplikuj kohortę",
|
||||
"addCohort": "Dodaj kohortę",
|
||||
"copy": " kopia"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Oceń wydajność modelu, badając rozkład wartości prognozowania i wartości metryk wydajności modelu. Możesz dokładniej zbadać model, patrząc na analizę porównawczą jego wydajności w różnych kohortach lub podgrupach zestaw danych. Wybierz filtry wzdłuż wartości y i wartości x, aby przeciąć różne wymiary. Wybierz ikonę koła zębatego na wykresie, aby zmienić typ wykresu.",
|
||||
"modelStatistics": "Statystyki modelu",
|
||||
"cohortPickerLabel": "Wybierz kohortę zestawów danych do zbadania",
|
||||
"missingParameters": "Ta karta wymaga, aby była podana tablica przewidywanych wartości z modelu.",
|
||||
"missingTrueY": "Statystyki wydajności modelu wymagają podania prawdziwych wyników oprócz przewidywanych wyników"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Wartość Y",
|
||||
"numberOfDatapoints": "Liczba punktów danych",
|
||||
"xValue": "Wartość X",
|
||||
"rowIndex": "Indeks wiersza",
|
||||
"featureImportance": "Ważność funkcji",
|
||||
"countTooltipPrefix": "Liczba: {0}",
|
||||
"count": "Liczba",
|
||||
"featurePrefix": "Funkcja",
|
||||
"importancePrefix": "Ważność",
|
||||
"cohort": "Kohorta",
|
||||
"howToRead": "Jak czytać ten wykres"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Eksploruj statystyki zestawów danych, wybierając różne filtry dla osi X, Y i kolorów, aby dzielić dane według różnych wymiarów. Twórz powyżej kohorty zestawów danych, aby analizować statystyki zestawów danych przy użyciu filtrów, takich jak przewidywany wynik, cechy zestawów danych i grupy błędów. Ikona koła zębatego w prawym górnym rogu wykresu umożliwia zmianę typu wykresu.",
|
||||
"colorValue": "Wartość koloru",
|
||||
"individualDatapoints": "Poszczególne punkty danych",
|
||||
"aggregatePlots": "Zagreguj wykresy",
|
||||
"chartType": "Typ wykresu",
|
||||
"missingParameters": "Ta karta wymaga dostarczenia zestawu danych oceny.",
|
||||
"noColor": "Brak"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Ważność funkcji",
|
||||
"placeholder": "Kliknij funkcję na wykresie słupkowym powyżej, aby wyświetlić jej wykres zależności"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Klikając na wykresie punktowym, możesz wybrać punkt danych, aby wyświetlić poniżej jego lokalne wartości ważności cech (lokalne wyjaśnienie) oraz wykres indywidualnego oczekiwania warunkowego (ICE, individual conditional expectation). Utwórz hipotetyczny, warunkowy punkt danych, korzystając z panelu po prawej stronie w celu zaburzenia cech znanego punktu danych. Wartości ważności cech są oparte na wielu przybliżeniach i nie są „przyczyną” przewidywań. Nie zalecamy użytkownikom podejmowania rzeczywistych decyzji na podstawie informacji z tego narzędzia bez ścisłej matematycznej niezawodności wnioskowania przyczyn.",
|
||||
"panelPlaceholder": "Do utworzenia przewidywań dla nowych punktów danych potrzebny jest model.",
|
||||
"cohortPickerLabel": "Wybierz kohortę zestawów danych do zbadania",
|
||||
"scatterLegendText": "Włącz lub wyłącz punkty danych na wykresie, klikając elementy legendy.",
|
||||
"realPoint": "Realne punkty danych",
|
||||
"noneSelectedYet": "Jeszcze nie wybrano",
|
||||
"whatIfDatapoints": "Waurnkowe punkty danych",
|
||||
"noneCreatedYet": "Jeszcze nie utworzono",
|
||||
"showLabel": "Pokaż:",
|
||||
"featureImportancePlot": "Wykres ważności funkcji",
|
||||
"icePlot": "Wykres ICE (individual conditional expectation)",
|
||||
"featureImportanceLackingParameters": "Podaj ważności cech lokalnych, aby zobaczyć, jak każda cecha wpływa na poszczególne przewidywania.",
|
||||
"featureImportanceGetStartedText": "Wybierz punkt, aby wyświetlić ważność funkcji",
|
||||
"iceLackingParameters": "Wykresy ICE wymagają działającego modelu do opracowywania przewidywań dla hipotetycznych punktów danych.",
|
||||
"IceGetStartedText": "Wybierz punkt lub utwórz punkt warunkowy, aby wyświetlić wykresy ICE",
|
||||
"whatIfDatapoint": "Warunkowy punkt danych",
|
||||
"whatIfHelpText": "Wybierz punkt na wykresie lub ręcznie wprowadź znany indeks punktu danych do zakłócenia i zapisz jako nowy punkt warunkowy.",
|
||||
"indexLabel": "Indeks danych do zakłócenia",
|
||||
"rowLabel": "Wiersz {0}",
|
||||
"whatIfNameLabel": "Nazwa warunkowego punktu danych",
|
||||
"featureValues": "Wartości funkcji",
|
||||
"predictedClass": "Przewidywana klasa:",
|
||||
"predictedValue": "Przewidywana wartość:",
|
||||
"probability": "Prawdopodobieństwo:",
|
||||
"trueClass": "Prawdziwa klasa:",
|
||||
"trueValue": "Prawdziwa wartość:",
|
||||
"trueValue.comment": "prefiks do rzeczywistej etykiety na potrzeby regresji",
|
||||
"newPredictedClass": "Nowa przewidywana klasa:",
|
||||
"newPredictedValue": "Nowa przewidywana wartość: ",
|
||||
"newProbability": "Nowe prawdopodobieństwo: ",
|
||||
"saveAsNewPoint": "Zapisz jako nowy punkt",
|
||||
"saveChanges": "Zapisz zmiany",
|
||||
"loading": "Trwa ładowanie...",
|
||||
"classLabel": "Klasa: {0}",
|
||||
"minLabel": "Minimum",
|
||||
"maxLabel": "Maksimum",
|
||||
"stepsLabel": "Kroki",
|
||||
"disclaimer": "Zastrzeżenie: Te wyjaśnienia są oparte na wielu przybliżeniach i nie są „przyczyną” przewidywań. Bez dobrej znajomości matematycznej złożoności wnioskowania przyczynowego nie radzimy użytkownikom podejmować rzeczywistych decyzji w oparciu o to narzędzie.",
|
||||
"missingParameters": "Ta karta wymaga dostarczenia zestawu danych oceny.",
|
||||
"selectionLimit": "Maksymalnie 3 wybrane punkty",
|
||||
"classPickerLabel": "Klasa",
|
||||
"tooltipTitleMany": "Najważniejsze przewidywane klasy ({0})",
|
||||
"whatIfTooltipTitle": "Warunkowe przewidywane klasy",
|
||||
"tooltipTitleFew": "Przewidywane klasy",
|
||||
"probabilityLabel": "Prawdopodobieństwo",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "Wartość powinna być liczbowa",
|
||||
"icePlotHelperText": "Wykresy ICE przedstawiają, jak wartości przewidywań wybranych punktów danych zmieniają się w zależności od wartości cechy z zakresu od minimalnej do maksymalnej."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Wybierz filtr",
|
||||
"TreatAsCategorical": "Traktuj jako kategorie",
|
||||
"addFilter": "Dodaj filtr",
|
||||
"addedFilters": "Dodane filtry",
|
||||
"noAddedFilters": "Nie dodano jeszcze filtrów",
|
||||
"defaultFilterState": "Wybierz filtr, aby dodać parametry do kohorty zestawów danych.",
|
||||
"cohortNameLabel": "Nazwa kohorty zestawu danych",
|
||||
"cohortNamePlaceholder": "Nazwij kohortę",
|
||||
"save": "Zapisz",
|
||||
"delete": "Usuń",
|
||||
"cancel": "Anuluj",
|
||||
"cohortNameError": "Brak nazwy kohorty",
|
||||
"placeholderName": "Kohorta {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Wybierz",
|
||||
"ditherLabel": "Powinny być symulowane",
|
||||
"selectFilter": "Wybierz wartość osi",
|
||||
"selectFeature": "Wybierz funkcję",
|
||||
"binLabel": "Zastosuj pakowanie do danych",
|
||||
"TreatAsCategorical": "Traktuj jako kategorie",
|
||||
"numOfBins": "Liczba pojemników",
|
||||
"groupByCohort": "Grupuj według kohorty",
|
||||
"selectClass": "Wybierz klasę",
|
||||
"countHelperText": "Histogram liczby punktów"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Przewidywane prawdopodobieństwo",
|
||||
"predictedY": "Przewidywana wartość Y",
|
||||
"evalData": "Zestaw danych oceny",
|
||||
"localFeatureImportance": "Istotność cechy lokalnej",
|
||||
"inconsistentDimensions": "Niespójne wymiary. Element {0} ma wymiary {1}, a oczekiwano wymiarów {2}",
|
||||
"notNonEmpty": "Dane wejściowe {0} nie są niepustą tablicą",
|
||||
"varyingLength": "Niespójne wymiary. Element {0} ma elementy o różnej długości",
|
||||
"notArray": "Element {0} nie jest tablicą. Oczekiwano tablicy o wymiarze {1}",
|
||||
"errorHeader": "Niektóre parametry wejściowe były niespójne i nie zostaną użyte: ",
|
||||
"datasizeWarning": "Zestaw danych oceny jest za duży, aby można było go efektywnie wyświetlić na niektórych wykresach. Dodaj filtry, aby zmniejszyć rozmiar kohorty. ",
|
||||
"datasizeError": "Wybrana kohorta jest za duża. Dodaj filtry, aby zmniejszyć jej rozmiar.",
|
||||
"addFilters": "Dodaj filtry"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " obejmuje {0}",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} i {1} innych"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "Błąd średniokwadratowy: {0}",
|
||||
"rSquared": "R-kwadrat: {0}",
|
||||
"meanPrediction": "Średnie przewidywanie {0}",
|
||||
"accuracy": "Dokładność: {0}",
|
||||
"precision": "Precyzja: {0}",
|
||||
"recall": "Kompletność: {0}",
|
||||
"fpr": "Współczynnik wyników fałszywie dodatnich: {0}",
|
||||
"fnr": "Współczynnik wyników fałszywie ujemnych: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Poznaj „top k” ważnych cech, które wpływają na ogólne przewidywania modelu. Użyj suwaka, aby wyświetlić malejące ważności cech."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Co oznaczają te wyjaśnienia?",
|
||||
"clickHere": "Dowiedz się więcej",
|
||||
"shapTitle": "Wartości Shapleya",
|
||||
"shapDescription": "Ten element wyjaśniający używa podejścia SHAP, czyli podejścia z zakresu teorii gier do wyjaśniania modeli, w którym ważność cech jest mierzona przez „ukrywanie” tych cech w modelu za pomocą marginalizacji. Kliknij poniższy link, aby dowiedzieć się więcej.",
|
||||
"limeTitle": "Lokalne, interpretowalne wyjaśnienia niezależne od modelu (LIME)",
|
||||
"limeDescription": "Ten element wyjaśniający używa lokalnych, interpretowalnych wyjaśnień niezależnych od modelu (LIME, Local Interpretable Model-Agnostic Explanations), które zapewniają liniową aproksymację modelu. Aby uzyskać wyjaśnienie, wykonujemy następujące czynności: zaburzamy wystąpienie i pobieramy przewidywania modelu, a następnie używamy tych przewidywań jako etykiet, aby nauczyć rozrzedzony model liniowy, który jest wierny lokalnie. Wagi tego modelu liniowego są używane jako ważności cech. Kliknij poniższy link, aby dowiedzieć się więcej.",
|
||||
"mimicTitle": "Naśladowanie (globalne wyjaśnienia zastępcze)",
|
||||
"mimicDescription": "Ten element wyjaśniający jest oparty na trenowaniu globalnych modeli zastępczych naśladujących modele czarnej skrzynki. Globalny model zastępczy to wewnętrznie interpretowalny model wytrenowany do jak najdokładniejszej aproksymacji przewidywań dowolnego modelu czarnej skrzynki. Wartości ważności cech to oparte na modelu wartości ważności cech podstawowego modelu zastępczego (LightGBM, regresja liniowa, stochastyczny spadek wzdłuż gradientu lub drzewo decyzyjne)",
|
||||
"pfiTitle": "Ważność cechy za pomocą permutacji",
|
||||
"pfiDescription": "Ten element wyjaśniający losowo zmienia dane, modyfikując po jednej cesze całego zestawu danych i obliczając, jak bardzo zmienia się badana metryka wydajności (domyślne metryki wydajności: F1 w przypadku klasyfikacji binarnej, miara F1 z mikrośrednią w przypadku klasyfikacji wielu klas oraz średni błąd bezwzględny w przypadku regresji). Im większa zmiana, tym ważniejsza jest cecha. Ten element wyjaśniający może wyjaśnić tylko ogólne zachowanie podstawowego modelu, ale nie wyjaśnia poszczególnych przewidywań. Wartość ważności cechy reprezentuje zmianę wydajności modelu spowodowaną zaburzeniem tej konkretnej cechy."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Selecione um ponto para ver sua explicação local",
|
||||
"defaultClassNames": "Classe {0}",
|
||||
"defaultFeatureNames": "Recurso {0}",
|
||||
"absoluteAverage": "Média do valor absoluto",
|
||||
"predictedClass": "Classe prevista",
|
||||
"datasetExplorer": "Explorer do Conjunto de Dados",
|
||||
"dataExploration": "Exploração do Conjunto de Dados",
|
||||
"aggregateFeatureImportance": "Importância do Recurso Agregado",
|
||||
"globalImportance": "Importância Global",
|
||||
"explanationExploration": "Exploração de Explicação",
|
||||
"individualAndWhatIf": "Importância do Recurso Individual e Hipótese",
|
||||
"summaryImportance": "Importância do Resumo",
|
||||
"featureImportance": "Importância do Recurso",
|
||||
"featureImportanceOf": "Importância do recurso de {0}",
|
||||
"perturbationExploration": "Exploração de Perturbação",
|
||||
"localFeatureImportance": "Importância do Recurso Local",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Limpar seleção",
|
||||
"feature": "Recurso:",
|
||||
"intercept": "Interceptação",
|
||||
"modelPerformance": "Desempenho do Modelo",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Dados: {0}",
|
||||
"importanceLabel": "Importância: {0}",
|
||||
"predictedY": "Y Previsto",
|
||||
"index": "Índice",
|
||||
"dataGroupLabel": "Dados",
|
||||
"output": "Saída",
|
||||
"probabilityLabel": "Probabilidade: {0}",
|
||||
"trueY": "Y verdadeiro",
|
||||
"class": "classe: ",
|
||||
"xValue": "Valor X:",
|
||||
"yValue": "Valor Y:",
|
||||
"colorValue": "Cor:",
|
||||
"count": "Contagem"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Ponderação entre classes:",
|
||||
"info": "Informações sobre o cálculo entre classes",
|
||||
"overviewInfo": "Os modelos de multiclasse geram um vetor de importância do recurso independente para cada classe. O vetor de importância do recurso de cada classe demonstra quais recursos fizeram uma classe mais ou menos provável. Você pode selecionar como os pesos dos vetores de importância do recurso por classe são resumidos em um único valor:",
|
||||
"absoluteValInfo": "Média do valor absoluto: mostra a soma da importância do recurso em todas as classes possíveis, dividida pelo número de classes",
|
||||
"predictedClassInfo": "Classe prevista: mostra o valor de importância do recurso para a classe prevista de um determinado ponto",
|
||||
"enumeratedClassInfo": "Nomes de classe enumerados: mostra somente os valores de importância do recurso da classe especificada em todos os pontos de dados.",
|
||||
"close": "Fechar",
|
||||
"crossClassWeights": "Pesos de classe cruzada"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Valor de Recurso Dimensionado",
|
||||
"low": "Baixa",
|
||||
"high": "Alta",
|
||||
"featureLabel": "Recurso: {0}",
|
||||
"valueLabel": "Valor do recurso: {0}",
|
||||
"importanceLabel": "Importância: {0}",
|
||||
"predictedClassTooltip": "Classe Prevista: {0}",
|
||||
"trueClassTooltip": "Classe Verdadeira: {0}",
|
||||
"predictedOutputTooltip": "Saída Prevista: {0}",
|
||||
"trueOutputTooltip": "Saída Verdadeira: {0}",
|
||||
"topKFeatures": "K Principais Recursos:",
|
||||
"topKInfo": "Como o k principal é calculado",
|
||||
"predictedValue": "Valor Previsto",
|
||||
"predictedClass": "Classe Prevista",
|
||||
"trueValue": "Valor Verdadeiro",
|
||||
"trueClass": "Classe Verdadeira",
|
||||
"noColor": "Nenhuma",
|
||||
"tooManyRows": "O conjunto de dados fornecido é maior do que este gráfico pode dar suporte"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Classe: {0}",
|
||||
"sortBy": "Classificar por",
|
||||
"noData": "Nenhum Dado",
|
||||
"absoluteGlobal": "Global absoluto",
|
||||
"absoluteLocal": "Local absoluto",
|
||||
"calculatingExplanation": "Calculando a explicação"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Deve ser numérico",
|
||||
"integerError": "Precisa ser um número inteiro",
|
||||
"prediction": "Previsão",
|
||||
"predictedProbability": "Probabilidade prevista",
|
||||
"predictionLabel": "Previsão: {0}",
|
||||
"probabilityLabel": "Probabilidade: {0}",
|
||||
"noModelError": "Forneça um modelo operacionalizado para explorar previsões em gráficos da ICE.",
|
||||
"featurePickerLabel": "Recurso:",
|
||||
"minimumInputLabel": "Mínimo:",
|
||||
"maximumInputLabel": "Máximo:",
|
||||
"stepInputLabel": "Etapas:",
|
||||
"loadingMessage": "Carregando dados...",
|
||||
"submitPrompt": "Enviar um intervalo para exibir um gráfico da ICE",
|
||||
"topLevelErrorMessage": "Erro no parâmetro",
|
||||
"errorPrefix": "Erro encontrado: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Carregando...",
|
||||
"perturbationLabel": "Perturbação:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Valor previsto: {0}",
|
||||
"predictedClassLabel": "Classe prevista: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Nenhum agrupamento",
|
||||
"groupPredicted": "Y Previsto",
|
||||
"groupTrue": "Y verdadeiro",
|
||||
"groupBy": "Agrupar por"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Tipo de gráfico:",
|
||||
"violinText": "Violino",
|
||||
"barText": "Barra",
|
||||
"boxText": "Quadrado",
|
||||
"beehiveText": "Por nuvem",
|
||||
"globalImportanceExplanation": "A importância do recurso global é calculada pela média do valor absoluto da importância do recurso de todos os pontos (normalização de L1). ",
|
||||
"multiclassImportanceAddendum": "Todos os pontos são incluídos no cálculo da importância de um recurso para todas as classes, nenhum peso diferencial é usado. Portanto, um recurso que tenha grande importância negativa para muitos pontos que estimaram que não seja da 'Classe A' aumentará muito a importância de 'Classe A' do recurso."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Igual a",
|
||||
"greaterThanComparison": "Maior que",
|
||||
"greaterThanEqualToComparison": "Maior ou igual a",
|
||||
"lessThanComparison": "Menor que",
|
||||
"lessThanEqualToComparison": "Menor ou igual a",
|
||||
"inTheRangeOf": "No intervalo de",
|
||||
"categoricalIncludeValues": "Valores incluídos:",
|
||||
"numericValue": "Valor",
|
||||
"numericalComparison": "Operação",
|
||||
"minimum": "Mínimo",
|
||||
"maximum": "Máximo",
|
||||
"min": "Mínimo: {0}",
|
||||
"max": "Máximo: {0}",
|
||||
"uniqueValues": "Nº de valores exclusivos: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Erro de regressão",
|
||||
"error": "Erro",
|
||||
"classificationOutcome": "Resultado da classificação",
|
||||
"truePositive": "Verdadeiro positivo",
|
||||
"trueNegative": "Verdadeiro negativo",
|
||||
"falsePositive": "Falso positivo",
|
||||
"falseNegative": "Falso negativo",
|
||||
"dataset": "Conjunto de dados",
|
||||
"predictedProbabilities": "Probabilidades de previsão",
|
||||
"none": "Contagem"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Fechar",
|
||||
"defaultCustomRootName": "Cópia da linha {0}",
|
||||
"filterFeaturePlaceholder": "Pesquisar os recursos"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Coorte",
|
||||
"defaultLabel": "Todos os dados"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Explore os principais recursos importantes de k que afetam as suas previsões de modelo gerais (ou seja, a explicação global). Use o controle deslizante para mostrar os valores de importância do recurso de modo decrescente. Selecione até três coortes para ver os valores de importância do recurso deles lado a lado. Clique em qualquer uma das barras de recursos no grafo para ver como os valores do recurso selecionado afetam a previsão de modelo.",
|
||||
"topAtoB": "Principais {0} – {1} recursos",
|
||||
"datasetCohorts": "Coortes do conjunto de dados",
|
||||
"legendHelpText": "Ative e desative coortes no gráfico clicando nos itens da legenda.",
|
||||
"sortBy": "Classificar por",
|
||||
"viewDependencePlotFor": "Exibir o gráfico de dependência para:",
|
||||
"datasetCohortSelector": "Selecionar um coorte do conjunto de dados",
|
||||
"aggregateFeatureImportance": "Importância do Recurso Agregado",
|
||||
"missingParameters": "Esta guia requer que o parâmetro local de importância do recurso seja fornecido.",
|
||||
"weightOptions": "Pesos de importância de classe",
|
||||
"dependencePlotTitle": "Gráficos de Dependência",
|
||||
"dependencePlotHelperText": "Este gráfico de dependências mostra a relação entre o valor de um recurso e a importância correspondente do recurso em um coorte.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Selecionar o recurso",
|
||||
"datasetRequired": "Os gráficos de dependência exigem o conjunto de dados de avaliação e a matriz local de importância do recurso."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Estatísticas de Dados",
|
||||
"datapoints": "{0} pontos de dados",
|
||||
"features": "{0} recursos",
|
||||
"filters": "{0} filtros",
|
||||
"binaryClassifier": "Classificador Binário",
|
||||
"regressor": "Regressor",
|
||||
"multiclassClassifier": "Classificador Multiclasse",
|
||||
"datasetCohorts": "Coortes do Conjunto de Dados",
|
||||
"editCohort": "Editar o Coorte",
|
||||
"duplicateCohort": "Duplicar o Coorte",
|
||||
"addCohort": "Adicionar Coorte",
|
||||
"copy": " cópia"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Avalie o desempenho do seu modelo explorando a distribuição dos seus valores de previsão e dos valores das suas métricas de desempenho do modelo. Você pode investigar ainda mais o seu modelo examinando uma análise comparativa do desempenho dele em diferentes coortes ou subgrupos do seu conjunto de dados. Selecione os filtros ao longo do valor y e do valor z para cortar em diferentes dimensões. Selecione a engrenagem no grafo para alterar o tipo de grafo.",
|
||||
"modelStatistics": "Estatísticas do Modelo",
|
||||
"cohortPickerLabel": "Selecionar um coorte do conjunto de dados a ser explorado",
|
||||
"missingParameters": "Esta guia requer que a matriz de valores previstos do modelo seja fornecida.",
|
||||
"missingTrueY": "As estatísticas de desempenho do modelo exigem que os verdadeiros resultados sejam fornecidos em adição aos resultados previstos"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Valor Y",
|
||||
"numberOfDatapoints": "Número de pontos de dados",
|
||||
"xValue": "Valor X",
|
||||
"rowIndex": "Índice de linha",
|
||||
"featureImportance": "Importância do recurso",
|
||||
"countTooltipPrefix": "Contagem: {0}",
|
||||
"count": "Contagem",
|
||||
"featurePrefix": "Recurso",
|
||||
"importancePrefix": "Importância",
|
||||
"cohort": "Coorte",
|
||||
"howToRead": "Como ler este gráfico"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Explore as estatísticas do conjunto de dados selecionando filtros diferentes ao longo do eixo X, Y e de cor para dividir os seus dados em dimensões diferentes. Crie conjunto de dados coortes acima para analisar as estatísticas do conjunto de dados com filtros, como resultado previsto, recursos do conjunto de dados e grupos de erros. Use o ícone de engrenagem no canto superior direito do grafo para alterar os tipos de grafo.",
|
||||
"colorValue": "Valor da cor",
|
||||
"individualDatapoints": "Pontos de dados individuais",
|
||||
"aggregatePlots": "Gráficos de agregação",
|
||||
"chartType": "Tipo de gráfico",
|
||||
"missingParameters": "Esta guia requer que um conjunto de dados de avaliação seja fornecido.",
|
||||
"noColor": "Nenhum"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Importância do recurso de",
|
||||
"placeholder": "Clique em um recurso no gráfico de barras acima para mostrar o gráfico de dependência"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Você pode selecionar um ponto de dados clicando no gráfico de dispersão para exibir os valores de importância do recurso local (explicação local) e o gráfico de ICE (expectativa condicional individual) abaixo. Crie um ponto de dados hipotético usando o painel à direita para desorganizar recursos de um ponto de dados conhecido. Os valores de importância do recurso são baseados em muitas aproximações e não são a \"causa\" das previsões. Sem forte robustez matemática de inferência causal, não aconselhamos os usuários a tomarem decisões de vida real com base nessa ferramenta.",
|
||||
"panelPlaceholder": "Um modelo é necessário para fazer previsões para novos pontos de dados.",
|
||||
"cohortPickerLabel": "Selecionar um coorte do conjunto de dados a ser explorado",
|
||||
"scatterLegendText": "Ative e desative os pontos de dados no gráfico clicando nos itens da legenda.",
|
||||
"realPoint": "Pontos de dados reais",
|
||||
"noneSelectedYet": "Nenhum foi selecionado ainda",
|
||||
"whatIfDatapoints": "Pontos de dados de hipótese",
|
||||
"noneCreatedYet": "Nenhum foi criado ainda",
|
||||
"showLabel": "Mostrar:",
|
||||
"featureImportancePlot": "Gráfico de importância do recurso",
|
||||
"icePlot": "Gráfico de ICE (expectativa condicional individual)",
|
||||
"featureImportanceLackingParameters": "Forneça as importâncias do recurso local para ver como cada recurso impacta previsões individuais.",
|
||||
"featureImportanceGetStartedText": "Selecione um ponto para exibir a importância do recurso",
|
||||
"iceLackingParameters": "Os gráficos ICE exigem um modelo operacional para fazer previsões para os pontos de extremidade hipotéticos.",
|
||||
"IceGetStartedText": "Selecione um ponto ou crie um ponto de Hipótese para exibir gráficos da ICE",
|
||||
"whatIfDatapoint": "Ponto de dados de hipótese",
|
||||
"whatIfHelpText": "Selecione um ponto no gráfico ou insira manualmente um índice de ponto de dados conhecido para causar perturbação e salve como um novo ponto de Hipótese.",
|
||||
"indexLabel": "Índice de dados no qual causar perturbação",
|
||||
"rowLabel": "Linha {0}",
|
||||
"whatIfNameLabel": "Nome do ponto de dados de hipótese",
|
||||
"featureValues": "Valores do recurso",
|
||||
"predictedClass": "Classe prevista: ",
|
||||
"predictedValue": "Valor previsto: ",
|
||||
"probability": "Probabilidade: ",
|
||||
"trueClass": "Classe verdadeira: ",
|
||||
"trueValue": "Valor verdadeiro: ",
|
||||
"trueValue.comment": "prefixo do rótulo real para regressão",
|
||||
"newPredictedClass": "Nova classe prevista: ",
|
||||
"newPredictedValue": "Novo valor previsto: ",
|
||||
"newProbability": "Nova probabilidade: ",
|
||||
"saveAsNewPoint": "Salvar como novo ponto",
|
||||
"saveChanges": "Salvar as alterações",
|
||||
"loading": "Carregando...",
|
||||
"classLabel": "Classe: {0}",
|
||||
"minLabel": "Mín.",
|
||||
"maxLabel": "Máx.",
|
||||
"stepsLabel": "Etapas",
|
||||
"disclaimer": "Aviso de isenção de responsabilidade: essas são explicações com base em muitas aproximações e não são a \"causa\" das previsões. Sem a forte robustez matemática da inferência causal, não aconselhamos os usuários a tomar decisões de vida real com base nesta ferramenta.",
|
||||
"missingParameters": "Esta guia requer que um conjunto de dados de avaliação seja fornecido.",
|
||||
"selectionLimit": "No máximo três pontos selecionados",
|
||||
"classPickerLabel": "Classe",
|
||||
"tooltipTitleMany": "{0} principais classes previstas",
|
||||
"whatIfTooltipTitle": "Classes previstas de hipótese",
|
||||
"tooltipTitleFew": "Classes previstas",
|
||||
"probabilityLabel": "Probabilidade",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "O valor deve ser numérico",
|
||||
"icePlotHelperText": "Plotagens de ICE demonstram como os valores de previsão do ponto de dados selecionado são alterados em um intervalo de valores de recurso entre um valor mínimo e um máximo."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Selecionar o Filtro",
|
||||
"TreatAsCategorical": "Tratar como categórico",
|
||||
"addFilter": "Adicionar Filtro",
|
||||
"addedFilters": "Filtros Adicionados",
|
||||
"noAddedFilters": "Nenhum filtro foi adicionado ainda",
|
||||
"defaultFilterState": "Selecione um filtro para adicionar parâmetros ao coorte do seu conjunto de dados.",
|
||||
"cohortNameLabel": "Nome do coorte do conjunto de dados",
|
||||
"cohortNamePlaceholder": "Nomeie o seu coorte",
|
||||
"save": "Salvar",
|
||||
"delete": "Excluir",
|
||||
"cancel": "Cancelar",
|
||||
"cohortNameError": "Nome do coorte ausente",
|
||||
"placeholderName": "Coorte {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Selecionar",
|
||||
"ditherLabel": "Deve pontilhar",
|
||||
"selectFilter": "Selecionar o valor do seu eixo",
|
||||
"selectFeature": "Selecionar o Recurso",
|
||||
"binLabel": "Aplicar compartimentalização aos dados",
|
||||
"TreatAsCategorical": "Tratar como categórico",
|
||||
"numOfBins": "Número de compartimentos",
|
||||
"groupByCohort": "Agrupar por coorte",
|
||||
"selectClass": "Selecionar a classe",
|
||||
"countHelperText": "Um histograma do número de pontos"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Probabilidade prevista",
|
||||
"predictedY": "Y previsto",
|
||||
"evalData": "Conjunto de dados de avaliação",
|
||||
"localFeatureImportance": "Importância do recurso local",
|
||||
"inconsistentDimensions": "Dimensões inconsistentes. {0} tem dimensões {1}, esperava-se {2}",
|
||||
"notNonEmpty": "A entrada {0} não é uma matriz não vazia",
|
||||
"varyingLength": "Dimensões inconsistentes. {0} tem elementos de comprimento variável",
|
||||
"notArray": "{0} não é uma matriz. Matriz de dimensão {1} esperada",
|
||||
"errorHeader": "Alguns parâmetros de entrada eram inconsistentes e não serão usados: ",
|
||||
"datasizeWarning": "O conjunto de dados de avaliação é muito grande para ser exibido com eficácia em alguns gráficos. Adicione filtros para diminuir o tamanho do coorte. ",
|
||||
"datasizeError": "O coorte selecionado é muito grande. Adicione filtros para diminuir o tamanho dele.",
|
||||
"addFilters": "Adicionar filtros"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " inclui {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} e {1} outros"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R ao quadrado: {0}",
|
||||
"meanPrediction": "Previsão média {0}",
|
||||
"accuracy": "Precisão: {0}",
|
||||
"precision": "Precisão: {0}",
|
||||
"recall": "Recall: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Explore os k principais recursos importantes que afetam suas previsões gerais do modelo. Use o controle deslizante para mostrar as importâncias de recurso em ordem decrescente."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "O que significam estas explicações?",
|
||||
"clickHere": "Saiba mais",
|
||||
"shapTitle": "Valores de Shapley",
|
||||
"shapDescription": "Este explicador usa o SHAP, que é uma abordagem de teoria dos jogos para explicar os modelos, na qual a importância dos conjuntos de recursos é medida \"ocultando\" esses recursos do modelo por meio de uso marginal. Clique no link abaixo para saber mais.",
|
||||
"limeTitle": "LIME (Explicações Locais Interpretáveis Independentes de Modelo)",
|
||||
"limeDescription": "Este explicativo usa LIME, que fornece uma aproximação linear do modelo. Para obter uma explicação, fazemos o seguinte: desorganizamos a instância, obtemos previsões de modelo e usamos essas previsões como rótulos para aprender um modelo linear esparso que é localmente fiel. Os pesos desse modelo linear são usados como 'importâncias do recurso'. Clique no link abaixo para saber mais.",
|
||||
"mimicTitle": "Imitação (Explicações Alternativas Globais)",
|
||||
"mimicDescription": "Este explicador se baseia na ideia de treinamento de modelos alternativos globais para imitar modelos de caixa preta. Um modelo alternativo global é um modelo intrinsecamente interpretável que é treinado para aproximar as previsões de qualquer modelo de caixa preta da forma mais exata possível. Os valores de importância do recurso são baseados no seu modelo alternativo subjacente (LightGBM, Regressão Linear, Gradiente Descendente Estocástico ou Árvore de Decisão)",
|
||||
"pfiTitle": "PFI (Importância do Recurso de Permutação)",
|
||||
"pfiDescription": "Este explicador embaralha aleatoriamente dados de um recurso por vez para todo o conjunto de dados e calcula o quanto a métrica de desempenho do interesse é alterada (métricas de desempenho padrão: medida F para classificação binária, medida F com micro média para classificação de multiclasse e erro médio absoluto para regressão). Quanto maior a alteração, mais importante será o recurso. Este explicador pode esclarecer somente o comportamento geral do modelo subjacente, e não previsões individuais. O valor de importância do recurso representa o delta no desempenho do modelo ao desorganizar esse recurso específico."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Selecione um ponto para ver a explicação local",
|
||||
"defaultClassNames": "Classe {0}",
|
||||
"defaultFeatureNames": "Funcionalidade {0}",
|
||||
"absoluteAverage": "Média do valor absoluto",
|
||||
"predictedClass": "Classe prevista",
|
||||
"datasetExplorer": "Explorador de Conjunto de Dados",
|
||||
"dataExploration": "Exploração de Conjuntos de Dados",
|
||||
"aggregateFeatureImportance": "Importância da Funcionalidade Agregada",
|
||||
"globalImportance": "Importância Global",
|
||||
"explanationExploration": "Exploração de Explicação",
|
||||
"individualAndWhatIf": "Importância da Funcionalidade Individual e Hipótese",
|
||||
"summaryImportance": "Importância do Resumo",
|
||||
"featureImportance": "Importância da Funcionalidade",
|
||||
"featureImportanceOf": "Importância da funcionalidade de {0}",
|
||||
"perturbationExploration": "Exploração de Perturbação",
|
||||
"localFeatureImportance": "Importância da Funcionalidade Local",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Limpar seleção",
|
||||
"feature": "Funcionalidade:",
|
||||
"intercept": "Intercetar",
|
||||
"modelPerformance": "Desempenho do Modelo",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Dados: {0}",
|
||||
"importanceLabel": "Importância: {0}",
|
||||
"predictedY": "Y Previsto",
|
||||
"index": "Índice",
|
||||
"dataGroupLabel": "Dados",
|
||||
"output": "Saída",
|
||||
"probabilityLabel": "Probabilidade: {0}",
|
||||
"trueY": "Y Verdadeiro",
|
||||
"class": "classe: ",
|
||||
"xValue": "Valor X:",
|
||||
"yValue": "Valor Y:",
|
||||
"colorValue": "Cor:",
|
||||
"count": "Contagem"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Ponderação entre classes:",
|
||||
"info": "Informações sobre o cálculo entre classes",
|
||||
"overviewInfo": "Os modelos de multiclasse geram um vetor de importância de funcionalidade independente para cada classe. O vetor de importância de funcionalidade de cada classe demonstra que funcionalidades tornaram uma classe mais provável ou menos provável. Pode selecionar a forma como os pesos dos vetores de importância da funcionalidade por classe são resumidos num valor único:",
|
||||
"absoluteValInfo": "Média do valor absoluto: mostra a soma da importância do recurso em todas as classes possíveis, dividida pelo número de classes",
|
||||
"predictedClassInfo": "Classe prevista: mostra o valor de importância da funcionalidade para a classe prevista de um determinado ponto",
|
||||
"enumeratedClassInfo": "Nomes de classe enumerados: mostra apenas os valores de importância da funcionalidade da classe especificada em todos os pontos de dados.",
|
||||
"close": "Fechar",
|
||||
"crossClassWeights": "Pesos entre classes"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Valor de Funcionalidade Dimensionado",
|
||||
"low": "Baixa",
|
||||
"high": "Elevado",
|
||||
"featureLabel": "Funcionalidade: {0}",
|
||||
"valueLabel": "Valor da funcionalidade: {0}",
|
||||
"importanceLabel": "Importância: {0}",
|
||||
"predictedClassTooltip": "Classe Prevista: {0}",
|
||||
"trueClassTooltip": "Classe Verdadeira: {0}",
|
||||
"predictedOutputTooltip": "Saída Prevista: {0}",
|
||||
"trueOutputTooltip": "Saída Verdadeira: {0}",
|
||||
"topKFeatures": "Principais Funcionalidades de K:",
|
||||
"topKInfo": "Como é calculado o top k",
|
||||
"predictedValue": "Valor Previsto",
|
||||
"predictedClass": "Classe Prevista",
|
||||
"trueValue": "Valor Verdadeiro",
|
||||
"trueClass": "Classe Verdadeira",
|
||||
"noColor": "Nenhum",
|
||||
"tooManyRows": "O conjunto de dados fornecido é maior do que este gráfico pode suportar"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Classe: {0}",
|
||||
"sortBy": "Ordenar por",
|
||||
"noData": "Sem Dados",
|
||||
"absoluteGlobal": "Global absoluto",
|
||||
"absoluteLocal": "Local absoluto",
|
||||
"calculatingExplanation": "A calcular explicação"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Tem de ser numérico",
|
||||
"integerError": "Tem de ser um número inteiro",
|
||||
"prediction": "Previsão",
|
||||
"predictedProbability": "Probabilidade prevista",
|
||||
"predictionLabel": "Predição: {0}",
|
||||
"probabilityLabel": "Probabilidade: {0}",
|
||||
"noModelError": "Forneça um modelo operacionalizado para explorar previsões em gráficos ICE.",
|
||||
"featurePickerLabel": "Funcionalidade:",
|
||||
"minimumInputLabel": "Mínimo:",
|
||||
"maximumInputLabel": "Máximo:",
|
||||
"stepInputLabel": "Passos:",
|
||||
"loadingMessage": "A carregar dados...",
|
||||
"submitPrompt": "Submeter um intervalo para ver um gráfico ICE",
|
||||
"topLevelErrorMessage": "Erro no parâmetro",
|
||||
"errorPrefix": "Erro encontrado: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "A carregar...",
|
||||
"perturbationLabel": "Perturbação:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Valor previsto: {0}",
|
||||
"predictedClassLabel": "Classe prevista: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Sem agrupamento",
|
||||
"groupPredicted": "Y Previsto",
|
||||
"groupTrue": "Y Verdadeiro",
|
||||
"groupBy": "Agrupar por"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Tipo de gráfico:",
|
||||
"violinText": "Violino",
|
||||
"barText": "Barras",
|
||||
"boxText": "Caixa",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "A importância global da funcionalidade é calculada através da média do valor absoluto da importância da funcionalidade de todos os pontos (normalização de L1). ",
|
||||
"multiclassImportanceAddendum": "Todos os pontos estão incluídos no cálculo da importância de uma funcionalidade para todas as classes, não é utilizada nenhuma ponderação diferencial. Portanto, uma funcionalidade com importância negativa grande para muitos pontos prevista como não sendo de \"Classe A\" aumentará imensamente a importância \"Classe A\" da funcionalidade."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Igual a",
|
||||
"greaterThanComparison": "Maior do que",
|
||||
"greaterThanEqualToComparison": "Maior que ou igual a",
|
||||
"lessThanComparison": "Menor do que",
|
||||
"lessThanEqualToComparison": "Menor que ou igual a",
|
||||
"inTheRangeOf": "No intervalo de",
|
||||
"categoricalIncludeValues": "Valores incluídos:",
|
||||
"numericValue": "Valor",
|
||||
"numericalComparison": "Operação",
|
||||
"minimum": "Mínimo",
|
||||
"maximum": "Máximo",
|
||||
"min": "Mín.: {0}",
|
||||
"max": "Máx.: {0}",
|
||||
"uniqueValues": "N.º de valores únicos: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Erro de regressão",
|
||||
"error": "Erro",
|
||||
"classificationOutcome": "Resultado da classificação",
|
||||
"truePositive": "Verdadeiro positivo",
|
||||
"trueNegative": "Verdadeiro negativo",
|
||||
"falsePositive": "Falso positivo",
|
||||
"falseNegative": "Falso negativo",
|
||||
"dataset": "Conjunto de dados",
|
||||
"predictedProbabilities": "Probabilidades de predição",
|
||||
"none": "Contagem"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Fechar",
|
||||
"defaultCustomRootName": "Cópia da linha {0}",
|
||||
"filterFeaturePlaceholder": "Procurar funcionalidades"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Coorte",
|
||||
"defaultLabel": "Todos os dados"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Explore as funcionalidades mais importantes de tipo top k que afetam as previsões gerais do seu modelo (também designadas por explicação global). Utilize o controlo de deslize para mostrar as importâncias das funcionalidades por ordem descendente. Selecione até três coortes para ver as importâncias das respetivas funcionalidades lado a lado. Clique em qualquer uma das barras de funcionalidades no gráfico, para ver como os valores da funcionalidade selecionada afetam a previsão do modelo.",
|
||||
"topAtoB": "Principais funcionalidades de {0} a {1}",
|
||||
"datasetCohorts": "Coortes do conjunto de dados",
|
||||
"legendHelpText": "Ative e desative coortes no desenho ao clicar nos itens da legenda.",
|
||||
"sortBy": "Ordenar por",
|
||||
"viewDependencePlotFor": "Ver desenho de dependência para:",
|
||||
"datasetCohortSelector": "Selecionar uma coorte de conjunto de dados",
|
||||
"aggregateFeatureImportance": "Importância da Funcionalidade Agregada",
|
||||
"missingParameters": "Este separador requer que o parâmetro de importância da funcionalidade local seja fornecido.",
|
||||
"weightOptions": "Pesos de importância de classe",
|
||||
"dependencePlotTitle": "Desenhos de Dependência",
|
||||
"dependencePlotHelperText": "Este desenho de dependência mostra a relação entre o valor de uma funcionalidade e a correspondente importância da funcionalidade numa coorte.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Selecionar funcionalidade",
|
||||
"datasetRequired": "Os desenhos de dependência requerem o conjunto de dados de avaliação e a matriz de importância da funcionalidade local."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Estatísticas de Dados",
|
||||
"datapoints": "{0} pontos de dados",
|
||||
"features": "{0} funcionalidades",
|
||||
"filters": "{0} filtros",
|
||||
"binaryClassifier": "Classificador Binário",
|
||||
"regressor": "Regressor",
|
||||
"multiclassClassifier": "Classificador Multiclasse",
|
||||
"datasetCohorts": "Coortes do Conjunto de Dados",
|
||||
"editCohort": "Editar Coorte",
|
||||
"duplicateCohort": "Coorte Duplicada",
|
||||
"addCohort": "Adicionar Coorte",
|
||||
"copy": " cópia"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Avalie o desempenho do modelo ao explorar a distribuição dos seus valores de predição e os valores das métricas de desempenho do modelo. Pode investigar ainda mais o seu modelo ao observar uma análise comparativa do desempenho em diferentes coortes ou subgrupos do seu conjunto de dados. Selecione filtros ao longo dos valores y e x para cortar em diferentes dimensões. Selecione a engrenagem no gráfico para alterar o tipo de gráfico.",
|
||||
"modelStatistics": "Estatísticas do Modelo",
|
||||
"cohortPickerLabel": "Selecionar uma coorte de conjunto de dados para explorar",
|
||||
"missingParameters": "Este separador requer a matriz de valores previstos do modelo seja fornecida.",
|
||||
"missingTrueY": "As estatísticas de desempenho do modelo requerem que os resultados reais sejam fornecidos para além dos resultados previstos"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Valor Y",
|
||||
"numberOfDatapoints": "Número de pontos de dados",
|
||||
"xValue": "Valor X",
|
||||
"rowIndex": "Índice de linha",
|
||||
"featureImportance": "Importância da funcionalidade",
|
||||
"countTooltipPrefix": "Contagem: {0}",
|
||||
"count": "Contagem",
|
||||
"featurePrefix": "Funcionalidade",
|
||||
"importancePrefix": "Importância",
|
||||
"cohort": "Coorte",
|
||||
"howToRead": "Como ler este gráfico"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Explore as estatísticas do conjunto de dados, selecionando filtros diferentes ao longo do eixo X, Y e cor para cortar os seus dados ao longo de diferentes dimensões. Crie coortes de conjunto de dados acima para analisar estatísticas de conjunto de dados com filtros como, por exemplo, resultados previstos, funcionalidades de conjunto de dados e grupos de erro. Utilize o ícone de engrenagem no canto superior direito do gráfico para alterar os tipos de gráficos.",
|
||||
"colorValue": "Valor da cor",
|
||||
"individualDatapoints": "Pontos de dados individuais",
|
||||
"aggregatePlots": "Desenhos agregados",
|
||||
"chartType": "Tipo de gráfico",
|
||||
"missingParameters": "Este separador requer um conjunto de dados de avaliação seja fornecido.",
|
||||
"noColor": "Nenhum"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Importância da funcionalidade de",
|
||||
"placeholder": "Clicar numa funcionalidade no gráfico de barras acima para mostrar o desenho de dependência"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Pode selecionar um ponto de dados clicando no gráfico de dispersão para ver os seus valores de importância de funcionalidade local (explicação local) e o gráfico de expectativa condicional individual (ICE) abaixo. Crie um ponto de dados hipotético, utilizando o painel à direito para funcionalidades de perturbação de um ponto de dados conhecido. Os valores de importância da funcionalidade baseiam-se em muitas aproximações e não são a \"causa\" das previsões. Sem uma robustez matemática rigorosa da inferência causal, não aconselhamos os utilizadores a tomarem decisões da vida real com base nesta ferramenta.",
|
||||
"panelPlaceholder": "É necessário um modelo para fazer predições para novos pontos de dados.",
|
||||
"cohortPickerLabel": "Selecionar uma coorte de conjunto de dados para explorar",
|
||||
"scatterLegendText": "Ative e desative pontos de dados no desenho ao clicar nos itens da legenda.",
|
||||
"realPoint": "Pontos de dados reais",
|
||||
"noneSelectedYet": "Ainda não foi selecionado um",
|
||||
"whatIfDatapoints": "Pontos de dados de hipótese",
|
||||
"noneCreatedYet": "Ainda não foi criado um",
|
||||
"showLabel": "Mostrar:",
|
||||
"featureImportancePlot": "Desenho da importância da funcionalidade",
|
||||
"icePlot": "Desenho de expectativa condicional individual (ICE)",
|
||||
"featureImportanceLackingParameters": "Forneça importâncias de características locais para ver como cada característica afeta as previsões individuais.",
|
||||
"featureImportanceGetStartedText": "Selecionar um ponto para ver a importância da funcionalidade",
|
||||
"iceLackingParameters": "Os desenhos de ICE requerem um modelo operacionalizado para fazer previsões para pontos de dados hipotéticos.",
|
||||
"IceGetStartedText": "Selecionar um ponto ou criar um ponto de Hipótese para ver desenhos ICE",
|
||||
"whatIfDatapoint": "Ponto de dados de hipótese",
|
||||
"whatIfHelpText": "Selecione um ponto no desenho ou introduza manualmente um índice de pontos de dados conhecido para transformar e guardar como um novo ponto de Hipótese.",
|
||||
"indexLabel": "Índice de dados para transformar",
|
||||
"rowLabel": "Linha {0}",
|
||||
"whatIfNameLabel": "Nome do ponto de dados de hipótese",
|
||||
"featureValues": "Valores da funcionalidade",
|
||||
"predictedClass": "Classe prevista: ",
|
||||
"predictedValue": "Valor previsto: ",
|
||||
"probability": "Probabilidade: ",
|
||||
"trueClass": "Classe verdadeira: ",
|
||||
"trueValue": "Valor verdadeiro: ",
|
||||
"trueValue.comment": "prefixo da etiqueta real para regressão",
|
||||
"newPredictedClass": "Nova classe prevista: ",
|
||||
"newPredictedValue": "Novo valor previsto: ",
|
||||
"newProbability": "Nova probabilidade: ",
|
||||
"saveAsNewPoint": "Guardar como novo ponto",
|
||||
"saveChanges": "Guardar alterações",
|
||||
"loading": "A carregar...",
|
||||
"classLabel": "Classe: {0}",
|
||||
"minLabel": "Mín.",
|
||||
"maxLabel": "Máx.",
|
||||
"stepsLabel": "Passos",
|
||||
"disclaimer": "Exclusão de responsabilidade: Estas são explicações que se baseiam em diversas aproximações e não são a \"causa\" das predições. Sem uma robustez matemática rigorosa da inferência causal, não aconselhamos os utilizadores a tomarem decisões da vida real com base nesta ferramenta.",
|
||||
"missingParameters": "Este separador requer um conjunto de dados de avaliação seja fornecido.",
|
||||
"selectionLimit": "Máximo de 3 pontos selecionados",
|
||||
"classPickerLabel": "Classe",
|
||||
"tooltipTitleMany": "{0} classes previstas principais",
|
||||
"whatIfTooltipTitle": "Classes previstas por hipóteses",
|
||||
"tooltipTitleFew": "Classes previstas",
|
||||
"probabilityLabel": "Probabilidade",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "O valor deve ser numérico",
|
||||
"icePlotHelperText": "Os gráficos ICE demonstram como os valores de previsão do ponto de dados selecionados se alteram ao longo de um intervalo de valores de funcionalidades, entre um valor mínimo e máximo."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Selecionar Filtro",
|
||||
"TreatAsCategorical": "Tratar como categórico",
|
||||
"addFilter": "Adicionar Filtro",
|
||||
"addedFilters": "Filtros adicionados",
|
||||
"noAddedFilters": "Ainda não foi adicionado nenhum filtro",
|
||||
"defaultFilterState": "Selecione um filtro para adicionar parâmetros à coorte do conjunto de dados.",
|
||||
"cohortNameLabel": "Nome da coorte do conjunto de dados",
|
||||
"cohortNamePlaceholder": "Atribuir nome à coorte",
|
||||
"save": "Guardar",
|
||||
"delete": "Eliminar",
|
||||
"cancel": "Cancelar",
|
||||
"cohortNameError": "Nome de coorte em falta",
|
||||
"placeholderName": "Coorte {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Selecionar",
|
||||
"ditherLabel": "Deve compor cores",
|
||||
"selectFilter": "Selecionar valor do eixo",
|
||||
"selectFeature": "Selecionar Funcionalidade",
|
||||
"binLabel": "Aplicar discretização aos dados",
|
||||
"TreatAsCategorical": "Tratar como categórico",
|
||||
"numOfBins": "Número de discretizações",
|
||||
"groupByCohort": "Grupo por coorte",
|
||||
"selectClass": "Selecionar classe",
|
||||
"countHelperText": "Um histograma do número de pontos"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Probabilidade prevista",
|
||||
"predictedY": "Y previsto",
|
||||
"evalData": "Conjunto de dados de avaliação",
|
||||
"localFeatureImportance": "Importância da funcionalidade local",
|
||||
"inconsistentDimensions": "Dimensões inconsistentes. {0} tem dimensões {1}, previstas {2}",
|
||||
"notNonEmpty": "A entrada {0} não uma matriz não vazia",
|
||||
"varyingLength": "Dimensões inconsistentes. {0} tem elementos de comprimento variado",
|
||||
"notArray": "{0} não uma matriz. Matriz esperada de dimensão {1}",
|
||||
"errorHeader": "Alguns parâmetros de entrada são inconsistentes e não serão utilizados: ",
|
||||
"datasizeWarning": "O conjunto de dados de avaliação é demasiado grande para ser exibido eficazmente em alguns gráficos. Adicione filtros para diminuir o tamanho da coorte. ",
|
||||
"datasizeError": "A coorte selecionada é demasiado grande. Adicione filtros para diminuir o tamanho da coorte.",
|
||||
"addFilters": "Adicionar filtros"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " inclui {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} e {1} outros"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R-quadrado: {0}",
|
||||
"meanPrediction": "Previsão média {0}",
|
||||
"accuracy": "Precisão: {0}",
|
||||
"precision": "Precisão: {0}",
|
||||
"recall": "Revocação: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Explore as funcionalidades importantes de top k que afetam as suas predições de modelo globais. Utilize o controlo de deslize para mostrar as importâncias das funcionalidades por ordem descendente."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "O que significam estas explicações?",
|
||||
"clickHere": "Saiba mais",
|
||||
"shapTitle": "Valores de Shapley",
|
||||
"shapDescription": "Este explicador utiliza o SHAP, que é uma abordagem teórica de jogos para explicar os modelos, em que a importância dos conjuntos de funcionalidades é medida por \"ocultar\" essas funcionalidades do modelo através de marginalização. Clique na ligação abaixo para saber mais.",
|
||||
"limeTitle": "LIME (Explicações Agnósticas de Modelo Interpretativo Local)",
|
||||
"limeDescription": "Este explicador utiliza LIME, que proporciona uma aproximação linear do modelo. Para obter uma explicação, fazemos o seguinte: perturbar a instância, obter previsões de modelos e utilizar estas previsões como etiquetas para aprender um modelo linear disperso, que é localmente fiel. Os pesos deste modelo linear são utilizados como \"funcionalidades importantes\". Clique na ligação abaixo para saber mais.",
|
||||
"mimicTitle": "Simular (Explicações Globais de Substituição)",
|
||||
"mimicDescription": "Este explicador baseia-se na ideia de formar modelos de substituição globais para simular modelos blackbox. Um modelo de substituição global é um modelo intrinsecamente interpretável, que é formado para aproximar as previsões de qualquer modelo de caixa preta com a maior precisão possível. Os valores de importância da característica são valores de importância de funcionalidade baseados no modelo do seu modelo de substituição subjacente (LightGBM, ou Regressão Linear, ou Descida de Gradiente Stochastic, ou Árvore de Decisões)",
|
||||
"pfiTitle": "Importância da Funcionalidade de Permutação (PFI)",
|
||||
"pfiDescription": "Este explicador baralha aleatoriamente dados uma funcionalidade de cada vez, para todo o conjunto de dados e calcula quanto é a métrica de desempenho das alterações de interesse (métricas de desempenho predefinidas: F1 para classificação binária, Pontuação F1 com micro média para classificação multiclasse e erro absoluto médio para regressão). Quanto maior for a alteração, mais importante é a funcionalidade. Este explicador apenas pode explicar o comportamento geral do modelo subjacente, mas não explica previsões individuais. O valor de importância de uma funcionalidade representa o delta no desempenho do modelo, perturbando essa funcionalidade em particular."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,109 +0,0 @@
|
|||
{
|
||||
"selectPoint": "[pj0ZL][©¥Select a point to see its local explanation !!! !!! !!! !!! ]",
|
||||
"defaultClassNames": "[UtXv6][éÐClass {0} !!! ]",
|
||||
"defaultFeatureNames": "[G3wWC][ÅÛFeature {0} !!! !]",
|
||||
"absoluteAverage": "[O4D5D][¥©Average of absolute value !!! !!! !!]",
|
||||
"predictedClass": "[KT1qT][ËÆPredicted class !!! !!]",
|
||||
"dataExploration": "[lOOHC][ðÇData Exploration !!! !!!]",
|
||||
"globalImportance": "[jTvBo][ÝÂGlobal Importance !!! !!!]",
|
||||
"explanationExploration": "[3kWwz][ïÒExplanation Exploration !!! !!! !]",
|
||||
"summaryImportance": "[g6WS1][¥µSummary Importance !!! !!!]",
|
||||
"featureImportance": "[3or4K][ãÎFeature Importance !!! !!!]",
|
||||
"perturbationExploration": "[pzza7][êÉPerturbation Exploration !!! !!! !]",
|
||||
"localFeatureImportance": "[TdtNk][êLocal Feature Importance !!! !!! !]",
|
||||
"ice": "[prPCi][Í@ICE !!]",
|
||||
"clearSelection": "[zQASF][ÀÜClear selection !!! !!]",
|
||||
"feature": "[tLO6f][íëFeature: !!! ]",
|
||||
"intercept": "[qlPsL][èãIntercept !!! ]",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "[BdvbK][çÏData : {0} !!! !]",
|
||||
"importanceLabel": "[FvbvK][îóImportance : {0} !!! !!!]",
|
||||
"predictedY": "[ziHhe][óïPredictedY !!! !]",
|
||||
"index": "[xlfT6][üÅIndex !!!]",
|
||||
"dataGroupLabel": "[AqAUL][ÀäData !!]",
|
||||
"output": "[j8KrV][õÇOutput !!!]",
|
||||
"probabilityLabel": "[PiqyA][ßáProbability : {0} !!! !!!]",
|
||||
"trueY": "[KQRAA][íéTrue Y !!!]",
|
||||
"class": "[VOC3W][âªclass: !!!]",
|
||||
"xValue": "[sHVVa][ë£X value: !!! ]",
|
||||
"yValue": "[BnzmV][@µY value: !!! ]",
|
||||
"colorValue": "[XnCSF][ÌüColor: !!!]"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "[50Sxt][ùéCross-class weighting: !!! !!! !]",
|
||||
"info": "[08bzl][ÖåInformation on cross-class calculation !!! !!! !!! !!]",
|
||||
"overviewInfo": "[PzmI5][ÖãMulticlass models generate an independent feature importance vector for each class. Each class's feature importance vector demonstrates which features made a class more likely or less likely. You can select how the weights of the per-class feature importance vectors are summarized into a single value: !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!!]",
|
||||
"absoluteValInfo": "[Yv4lf][§ãAverage of absolute value: Shows the sum of the feature's importance across all possible classes, divided by number of classes !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !]",
|
||||
"predictedClassInfo": "[wdPZC][û¢Predicted class: Shows the feature importance value for a given point's predicted class !!! !!! !!! !!! !!! !!! !!! !!!]",
|
||||
"enumeratedClassInfo": "[j44e9][áêEnumerated class names: Shows only the specified class's feature importance values across all data points. !!! !!! !!! !!! !!! !!! !!! !!! !!! !!]",
|
||||
"close": "[zjhsk][ôµClose !!!]"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "[pORh1][ãËScaled Feature Value !!! !!! ]",
|
||||
"low": "[8j80J][ªüLow !!]",
|
||||
"high": "[iPF8V][ÈÇHigh !!]",
|
||||
"featureLabel": "[VfxF0][ÔüFeature: {0} !!! !]",
|
||||
"valueLabel": "[NJ8d6][ÓïFeature value: {0} !!! !!!]",
|
||||
"importanceLabel": "[xMTuc][ÑÔImportance: {0} !!! !!]",
|
||||
"predictedClassTooltip": "[YFF5y][µíPredicted Class: {0} !!! !!! ]",
|
||||
"trueClassTooltip": "[kEOIt][ªÕTrue Class: {0} !!! !!]",
|
||||
"predictedOutputTooltip": "[xxu9k][ýêPredicted Output: {0} !!! !!! ]",
|
||||
"trueOutputTooltip": "[aMYpA][âòTrue Output: {0} !!! !!!]",
|
||||
"topKFeatures": "[HS6IQ][úçTop K Features: !!! !!]",
|
||||
"topKInfo": "[uXrUe][ÔÖHow is top k calculated !!! !!! !]",
|
||||
"predictedValue": "[493X1][ÈäPredicted Value !!! !!]",
|
||||
"predictedClass": "[k8iDy][ßÑPredicted Class !!! !!]",
|
||||
"trueValue": "[xl0EI][ËÿTrue Value !!! !]",
|
||||
"trueClass": "[lj5M0][ªæTrue Class !!! !]",
|
||||
"noColor": "[kzmVi][ÒÃNone !!]",
|
||||
"tooManyRows": "[2WWK2][ýÜThe provided dataset is larger than this chart can support !!! !!! !!! !!! !!! !]"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "[u2tbv][£ÞClass: {0} !!! !]",
|
||||
"sortBy": "[1ewmW][ËÆSort by !!!]",
|
||||
"noData": "[UNQEN][@ýNo Data !!!]",
|
||||
"absoluteGlobal": "[yIM5x][òßAbsolute global !!! !!]",
|
||||
"absoluteLocal": "[Yc3In][ËáAbsolute local !!! !!]",
|
||||
"calculatingExplanation": "[tCp6t][ÃÇCalculating explanation !!! !!! !]"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "[DTMTC][âäMust be numeric !!! !!]",
|
||||
"integerError": "[3vWkI][ðòMust be an integer !!! !!!]",
|
||||
"prediction": "[5h8Q3][ðàPrediction !!! !]",
|
||||
"predictedProbability": "[p7F6t][âÖPredicted probability !!! !!! ]",
|
||||
"predictionLabel": "[r59aK][ÒþPrediction: {0} !!! !!]",
|
||||
"probabilityLabel": "[2RSX2][ÊðProbability: {0} !!! !!!]",
|
||||
"noModelError": "[BVZkF][âÀPlease provide an operationalized model to explore predictions in ICE plots. !!! !!! !!! !!! !!! !!! !!! ]",
|
||||
"featurePickerLabel": "[msNZx][ëÖFeature: !!! ]",
|
||||
"minimumInputLabel": "[rZbXm][ïïMinimum: !!! ]",
|
||||
"maximumInputLabel": "[i6zUS][àøMaximum: !!! ]",
|
||||
"stepInputLabel": "[RGx9j][ÌãSteps: !!!]",
|
||||
"loadingMessage": "[xgP7D][ÃøLoading data... !!! !!]",
|
||||
"submitPrompt": "[w1fJh][ÜùSubmit a range to view an ICE plot !!! !!! !!! !]",
|
||||
"topLevelErrorMessage": "[9p4hQ][ïµError in parameter !!! !!!]",
|
||||
"errorPrefix": "[M9P6g][óìError encountered: {0} !!! !!! !]"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "[zMRsK][©éLoading... !!! !]",
|
||||
"perturbationLabel": "[UBaah][ÙýPerturbation: !!! !!]"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "[sDBkT][ÎéPredicted value : {0} !!! !!! ]",
|
||||
"predictedClassLabel": "[qGwXu][ÑèPredicted class : {0} !!! !!! ]"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "[1T5yN][ÔöNo grouping !!! !]",
|
||||
"groupPredicted": "[A3shb][@ÂPredicted Y !!! !]",
|
||||
"groupTrue": "[2GB3k][ÆðTrue Y !!!]",
|
||||
"groupBy": "[QYE75][ØæGroup by !!! ]"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "[54FZM][ÂÝChart type: !!! !]",
|
||||
"violinText": "[ozu6A][ÌüViolin !!!]",
|
||||
"barText": "[BlBG3][ÔýBar !!]",
|
||||
"boxText": "[JuqVX][àËBox !!]",
|
||||
"beehiveText": "[RjWs9][Ö¥Swarm !!!]",
|
||||
"globalImportanceExplanation": "[HxsS0][ÚóGlobal feature importance is calculated by averaging the absolute value of the feature importance of all points (L1 normalization). !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!!]",
|
||||
"multiclassImportanceAddendum": "[FpXeR][öïAll points are included in calculating a feature's importance for all classes, no differential weighting is used. So a feature that has large negative importance for many points predicted to not be of 'Class A' will greatly increase that feature's 'Class A' importance. !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!! !!!]"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Выберите точку, чтобы просмотреть объяснение ее локальной важности.",
|
||||
"defaultClassNames": "Класс {0}",
|
||||
"defaultFeatureNames": "Функция {0}",
|
||||
"absoluteAverage": "Среднее абсолютного значения",
|
||||
"predictedClass": "Прогнозируемый класс",
|
||||
"datasetExplorer": "Обозреватель наборов данных",
|
||||
"dataExploration": "Исследование набора данных",
|
||||
"aggregateFeatureImportance": "Совокупная важность признаков",
|
||||
"globalImportance": "Глобальный уровень важности",
|
||||
"explanationExploration": "Изучение объяснения",
|
||||
"individualAndWhatIf": "Важность отдельных признаков и гипотетические результаты",
|
||||
"summaryImportance": "Суммарная важность",
|
||||
"featureImportance": "Важность признака",
|
||||
"featureImportanceOf": "Важность признака {0}",
|
||||
"perturbationExploration": "Исследование изменений",
|
||||
"localFeatureImportance": "Локальная важность признака",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Очистить выделение",
|
||||
"feature": "Компонент:",
|
||||
"intercept": "Отсекаемый отрезок",
|
||||
"modelPerformance": "Производительность модели",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Данные: {0}",
|
||||
"importanceLabel": "Важность: {0}",
|
||||
"predictedY": "Прогнозируемое значение Y",
|
||||
"index": "Индекс",
|
||||
"dataGroupLabel": "Данные",
|
||||
"output": "Вывод",
|
||||
"probabilityLabel": "Вероятность: {0}",
|
||||
"trueY": "Истинное значение Y",
|
||||
"class": "класс: ",
|
||||
"xValue": "Значение X:",
|
||||
"yValue": "Значение Y:",
|
||||
"colorValue": "Цвет:",
|
||||
"count": "Количество"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Взвешивание между классами:",
|
||||
"info": "Сведения о многоклассовых вычислениях",
|
||||
"overviewInfo": "Многоклассовые модели создают независимый вектор важности признаков для каждого класса. Этот вектор показывает, какие из признаков делают класс более или менее вероятным. Вы можете выбрать, каким образом веса векторов важности признаков для каждого класса сводятся к одному значению:",
|
||||
"absoluteValInfo": "Среднее абсолютного значения: отображает сумму значений важности функции во всех возможных классах, разделенную на число классов.",
|
||||
"predictedClassInfo": "Прогнозируемый класс: отображает значение важности признака для прогнозируемого класса заданной точки.",
|
||||
"enumeratedClassInfo": "Имена перечислимых классов: отображает только значения важности признаков указанного класса во всех точках данных.",
|
||||
"close": "Закрыть",
|
||||
"crossClassWeights": "Межклассовые весовые коэффициенты"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Масштабированное значение функции",
|
||||
"low": "Низкий",
|
||||
"high": "Высокий",
|
||||
"featureLabel": "Функция: {0}",
|
||||
"valueLabel": "Значение функции: {0}",
|
||||
"importanceLabel": "Важность: {0}",
|
||||
"predictedClassTooltip": "Прогнозируемый класс: {0}",
|
||||
"trueClassTooltip": "Истинный класс: {0}",
|
||||
"predictedOutputTooltip": "Прогнозируемые выходные данные: {0}",
|
||||
"trueOutputTooltip": "Истинные выходные данные: {0}",
|
||||
"topKFeatures": "K самых важных функций:",
|
||||
"topKInfo": "Как вычисляются первые k позиций",
|
||||
"predictedValue": "Предсказанное значение",
|
||||
"predictedClass": "Прогнозируемый класс",
|
||||
"trueValue": "Истина",
|
||||
"trueClass": "Истинный класс",
|
||||
"noColor": "Нет",
|
||||
"tooManyRows": "Размер указанного набора данных превышает значение, поддерживаемое этой диаграммой."
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Класс: {0}",
|
||||
"sortBy": "Сортировать по",
|
||||
"noData": "Нет данных",
|
||||
"absoluteGlobal": "Абсолютное значение глобальной важности",
|
||||
"absoluteLocal": "Абсолютное значение локальной важности",
|
||||
"calculatingExplanation": "Объяснение расчета"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Значение должно быть числовым.",
|
||||
"integerError": "Требуется целое число.",
|
||||
"prediction": "Прогноз",
|
||||
"predictedProbability": "Прогнозируемая вероятность",
|
||||
"predictionLabel": "Прогноз: {0}",
|
||||
"probabilityLabel": "Вероятность: {0}",
|
||||
"noModelError": "Укажите модель, генерирующую прогнозы на основе введенных данных, для исследования прогнозов на графиках ICE.",
|
||||
"featurePickerLabel": "Компонент:",
|
||||
"minimumInputLabel": "Минимум:",
|
||||
"maximumInputLabel": "Максимум:",
|
||||
"stepInputLabel": "Этапы:",
|
||||
"loadingMessage": "Идет загрузка данных...",
|
||||
"submitPrompt": "Отправьте диапазон для просмотра графика ICE",
|
||||
"topLevelErrorMessage": "Ошибка в параметре",
|
||||
"errorPrefix": "Возникла ошибка: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Загрузка...",
|
||||
"perturbationLabel": "Изменение:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Прогнозируемое значение: {0}",
|
||||
"predictedClassLabel": "Прогнозируемый класс: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Без группирования",
|
||||
"groupPredicted": "Прогнозируемое значение Y",
|
||||
"groupTrue": "Истинное значение Y",
|
||||
"groupBy": "Группировать по"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Тип диаграммы:",
|
||||
"violinText": "Скрипка",
|
||||
"barText": "Линейчатая",
|
||||
"boxText": "Рамка",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "Глобальная важность признака вычисляется путем усреднения абсолютного значения важности признака по всем точкам (нормализация L1). ",
|
||||
"multiclassImportanceAddendum": "В вычисление важности функции для всех классов включены все точки, поэтому разностное взвешивание не используется. Таким образом, функция, имеющая большее отрицательное значение важности для многих точек и предсказанная как не являющаяся функцией класса А, значительно увеличит уровень важности функции класса А."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Равно",
|
||||
"greaterThanComparison": "Больше",
|
||||
"greaterThanEqualToComparison": "Больше или равно",
|
||||
"lessThanComparison": "Меньше",
|
||||
"lessThanEqualToComparison": "Меньше или равно",
|
||||
"inTheRangeOf": "В диапазоне",
|
||||
"categoricalIncludeValues": "Включенные значения:",
|
||||
"numericValue": "Значение",
|
||||
"numericalComparison": "Операция",
|
||||
"minimum": "Минимум",
|
||||
"maximum": "Максимум",
|
||||
"min": "Мин.: {0}",
|
||||
"max": "Макс.: {0}",
|
||||
"uniqueValues": "Число уникальных значений: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Ошибка регрессии",
|
||||
"error": "Ошибка",
|
||||
"classificationOutcome": "Результат классификации",
|
||||
"truePositive": "Истинноположительный результат",
|
||||
"trueNegative": "Истинноотрицательный результат",
|
||||
"falsePositive": "Ложноположительный результат",
|
||||
"falseNegative": "Ложноотрицательный результат",
|
||||
"dataset": "Набор данных",
|
||||
"predictedProbabilities": "Прогнозируемые вероятности",
|
||||
"none": "Количество"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Закрыть",
|
||||
"defaultCustomRootName": "Копия строки {0}",
|
||||
"filterFeaturePlaceholder": "Поиск признаков"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Когорта",
|
||||
"defaultLabel": "Все данные"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Изучите первые k важных признаков, влияющих на общие прогнозы по вашей модели (это также называется \"глобальным объяснением\"). Используйте ползунок для отображения важности признаков в порядке убывания. Выберите когорты (не более трех), чтобы просмотреть важность их признаков одновременно. Щелкните столбец с любым из признаков на графике, чтобы увидеть, как значения выбранного признака влияют на прогноз модели.",
|
||||
"topAtoB": "Первые признаки ({1}–{0})",
|
||||
"datasetCohorts": "Когорты наборов данных",
|
||||
"legendHelpText": "Щелкайте элементы условных обозначений для включения или отключения когорт на графике.",
|
||||
"sortBy": "Метод сортировки",
|
||||
"viewDependencePlotFor": "Показать график зависимостей для:",
|
||||
"datasetCohortSelector": "Выберите когорту набора данных",
|
||||
"aggregateFeatureImportance": "Совокупная важность признаков",
|
||||
"missingParameters": "На этой вкладке требуется указать параметр локальной важности признака.",
|
||||
"weightOptions": "Весовые коэффициенты важности класса",
|
||||
"dependencePlotTitle": "Графики зависимостей",
|
||||
"dependencePlotHelperText": "Эта график зависимости отражает связь между значением признака и соответствующим значением важности признака в когорте.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Выберите признак",
|
||||
"datasetRequired": "Для построения графиков зависимостей требуется набор данных для оценки и массив локальных значений важности признаков."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Статистика данных",
|
||||
"datapoints": "Точек данных: {0}",
|
||||
"features": "Признаков: {0}",
|
||||
"filters": "Фильтров: {0}",
|
||||
"binaryClassifier": "Двоичный классификатор",
|
||||
"regressor": "Регрессор",
|
||||
"multiclassClassifier": "Многоклассовый классификатор",
|
||||
"datasetCohorts": "Когорты набора данных",
|
||||
"editCohort": "Изменить когорту",
|
||||
"duplicateCohort": "Дублировать когорту",
|
||||
"addCohort": "Добавить когорту",
|
||||
"copy": " (копия)"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Оцените производительность вашей модели, изучив распределение значений прогнозирования и значений ее метрик производительности. Для дальнейшего изучения вы можете просмотреть сравнительный анализ производительности модели по различным когортам или подгруппам своего набора данных. Выберите фильтры для значений X и Y, чтобы рассмотреть различные измерения. Значок шестеренки на графе позволяет изменить тип графа.",
|
||||
"modelStatistics": "Статистика модели",
|
||||
"cohortPickerLabel": "Выберите когорту набора данных для изучения",
|
||||
"missingParameters": "На этой вкладке требуется указать массив прогнозируемых значений из модели.",
|
||||
"missingTrueY": "Для получения статистики по производительности модели, помимо прогнозируемых результатов, требуется указать истинные результаты."
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Значение Y",
|
||||
"numberOfDatapoints": "Число точек данных",
|
||||
"xValue": "Значение X",
|
||||
"rowIndex": "Индекс строки",
|
||||
"featureImportance": "Важность функции",
|
||||
"countTooltipPrefix": "Количество: {0}",
|
||||
"count": "Количество",
|
||||
"featurePrefix": "Функция",
|
||||
"importancePrefix": "Важность",
|
||||
"cohort": "Когорта",
|
||||
"howToRead": "Как читать эту диаграмму"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Изучайте статистику для набора данных, выбирая различные фильтры для осей X, Y и оси цвета, чтобы разделить данные по разным измерениям. Создавайте когорты наборов данных, чтобы анализировать статистику для набора данных с помощью фильтров, таких как спрогнозированный результат, признаки набора данных и группы ошибок. Чтобы изменить тип диаграммы, используйте значок шестеренки в верхнем правом углу графика.",
|
||||
"colorValue": "Цвет значения",
|
||||
"individualDatapoints": "Отдельные точки данных",
|
||||
"aggregatePlots": "Агрегировать графики",
|
||||
"chartType": "Тип диаграммы",
|
||||
"missingParameters": "На этой вкладке требуется указать набор данных для оценки.",
|
||||
"noColor": "Нет"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Важность признака",
|
||||
"placeholder": "Щелкните признак на линейчатой диаграмме выше для отображения графика его зависимостей"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Вы можете выбрать точку данных, щелкнув диаграмму рассеяния, и просмотреть локальные значения важности признаков (локальное объяснение) и график индивидуального условного математического ожидания (ICE) ниже. Создайте гипотетическую точку данных \"что если\", используя панель справа, чтобы внести отклонения в признаки известной точки данных. Значения важности признаков основаны на многочисленных приближениях и не являются \"причинами\" прогнозов. Если причинный вывод не обладает строгой математической корректностью, мы не рекомендуем пользователям принимать реальные решения на основе этого инструмента.",
|
||||
"panelPlaceholder": "Для создания прогнозов для новых точек данных необходима модель.",
|
||||
"cohortPickerLabel": "Выберите когорту набора данных для изучения",
|
||||
"scatterLegendText": "Щелкайте элементы условных обозначений для включения или отключения точек данных на графике.",
|
||||
"realPoint": "Реальные точки данных",
|
||||
"noneSelectedYet": "Пока ничего не выбрано",
|
||||
"whatIfDatapoints": "Гипотетические точки данных",
|
||||
"noneCreatedYet": "Пока ничего не создано",
|
||||
"showLabel": "Показать:",
|
||||
"featureImportancePlot": "График важности признаков",
|
||||
"icePlot": "График индивидуального условного ожидания (ICE)",
|
||||
"featureImportanceLackingParameters": "Укажите важность локальных признаков, чтобы узнать, как каждый из них влияет на отдельные прогнозы.",
|
||||
"featureImportanceGetStartedText": "Выберите точку для просмотра важности признака",
|
||||
"iceLackingParameters": "Для создания прогнозов для гипотетических точек данных в графиках ICE требуется активированная модель.",
|
||||
"IceGetStartedText": "Выберите точку или создайте гипотетическую точку для отображения графиков ICE",
|
||||
"whatIfDatapoint": "Гипотетическая точка данных",
|
||||
"whatIfHelpText": "Выберите точку на графике или вручную введите известный индекс точки данных для воздействия и сохраните ее как новую гипотетическую точку.",
|
||||
"indexLabel": "Индекс данных для воздействия",
|
||||
"rowLabel": "Строка {0}",
|
||||
"whatIfNameLabel": "Имя гипотетической точки данных",
|
||||
"featureValues": "Значения признаков",
|
||||
"predictedClass": "Спрогнозированный класс: ",
|
||||
"predictedValue": "Спрогнозированное значение: ",
|
||||
"probability": "Вероятность: ",
|
||||
"trueClass": "Истинный класс: ",
|
||||
"trueValue": "Истинное значение: ",
|
||||
"trueValue.comment": "Префикс к фактической метке регрессии",
|
||||
"newPredictedClass": "Новый спрогнозированный класс: ",
|
||||
"newPredictedValue": "Новое спрогнозированное значение: ",
|
||||
"newProbability": "Новая вероятность: ",
|
||||
"saveAsNewPoint": "Сохранить как новую точку",
|
||||
"saveChanges": "Сохранить изменения",
|
||||
"loading": "Загрузка…",
|
||||
"classLabel": "Класс: {0}",
|
||||
"minLabel": "Мин.",
|
||||
"maxLabel": "Макс.",
|
||||
"stepsLabel": "Число шагов",
|
||||
"disclaimer": "Отказ от ответственности. Приведенные здесь пояснения основаны на многочисленных аппроксимациях и не являются \"причиной\" прогнозов. Этот инструмент не предоставляет математически обоснованный причинно-следственный вывод, поэтому мы не рекомендуем пользователям принимать на его основе решения в реальных ситуациях.",
|
||||
"missingParameters": "На этой вкладке требуется указать набор данных для оценки.",
|
||||
"selectionLimit": "Можно выбрать не более 3 точек.",
|
||||
"classPickerLabel": "Класс",
|
||||
"tooltipTitleMany": "Основные спрогнозированные классы ({0})",
|
||||
"whatIfTooltipTitle": "Спрогнозированные классы \"Что если\"",
|
||||
"tooltipTitleFew": "Спрогнозированные классы",
|
||||
"probabilityLabel": "Вероятность",
|
||||
"deltaLabel": "Различие",
|
||||
"nonNumericValue": "Значение должно быть числовым.",
|
||||
"icePlotHelperText": "Графики ICE показывают, как значения прогнозирования для выбранной точки данных изменяются вдоль диапазона значений признаков между минимальным и максимальным значениями."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Выберите фильтр",
|
||||
"TreatAsCategorical": "Рассматривать как категориальные",
|
||||
"addFilter": "Добавить фильтр",
|
||||
"addedFilters": "Добавленные фильтры",
|
||||
"noAddedFilters": "Фильтры пока не добавлены",
|
||||
"defaultFilterState": "Выберите фильтр, чтобы добавить параметры в когорту набора данных.",
|
||||
"cohortNameLabel": "Имя когорты набора данных",
|
||||
"cohortNamePlaceholder": "Присвойте когорте имя",
|
||||
"save": "Сохранить",
|
||||
"delete": "Удалить",
|
||||
"cancel": "Отмена",
|
||||
"cohortNameError": "Отсутствует имя когорты.",
|
||||
"placeholderName": "Когорта {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Выберите",
|
||||
"ditherLabel": "Размывание значений",
|
||||
"selectFilter": "Выберите значение по оси",
|
||||
"selectFeature": "Выберите признак",
|
||||
"binLabel": "Применить группирование к данным",
|
||||
"TreatAsCategorical": "Рассматривать как категориальные",
|
||||
"numOfBins": "Число групп",
|
||||
"groupByCohort": "Группировать по когорте",
|
||||
"selectClass": "Выбор класса",
|
||||
"countHelperText": "Гистограмма количества точек"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Прогнозируемая вероятность",
|
||||
"predictedY": "Прогнозируемое значение Y",
|
||||
"evalData": "Набор данных для оценки",
|
||||
"localFeatureImportance": "Локальная важность признака",
|
||||
"inconsistentDimensions": "Несогласованные размеры. {0} имеет следующие размеры: {1}, ожидается: {2}",
|
||||
"notNonEmpty": "Входные данные {0} не являются непустым массивом.",
|
||||
"varyingLength": "Несогласованные размеры. {0} содержит элементы различной длины.",
|
||||
"notArray": "{0} не является массивом. Ожидается массив с размерностью {1}.",
|
||||
"errorHeader": "Некоторые входные параметры оказались несогласованными и не будут использоваться: ",
|
||||
"datasizeWarning": "Анализируемый набор данных слишком велик для отображения на некоторых диаграммах. Добавьте фильтры, чтобы уменьшить размер когорты. ",
|
||||
"datasizeError": "Выбранная когорта слишком велика. Добавьте фильтры, чтобы уменьшить размер когорты.",
|
||||
"addFilters": "Добавьте фильтры"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " включает {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} и {1} других"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "Среднеквадратичная ошибка: {0}",
|
||||
"rSquared": "R-квадрат: {0}",
|
||||
"meanPrediction": "Средний прогноз {0}",
|
||||
"accuracy": "Правильность: {0}",
|
||||
"precision": "Точность: {0}",
|
||||
"recall": "Полнота: {0}",
|
||||
"fpr": "Ложноположительные: {0}",
|
||||
"fnr": "Ложноотрицательные: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Изучите первые k важных функций, влияющих на общие прогнозы по вашей модели. Используйте ползунок для отображения важности признаков в порядке убывания."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Что означают эти пояснения?",
|
||||
"clickHere": "Дополнительные сведения",
|
||||
"shapTitle": "Значения Шепли",
|
||||
"shapDescription": "Этот модуль объяснения использует подход SHAP, который представляет собой теоретико-игровой подход для объяснения моделей, в котором важность наборов признаков измеряется путем \"скрытия\" этих признаков из модели с помощью маргинализации. Для получения дополнительных сведений перейдите по ссылке ниже.",
|
||||
"limeTitle": "LIME (локально интерпретируемое объяснение, которое не зависит от модели)",
|
||||
"limeDescription": "Этот модуль объяснения использует LIME, что обеспечивает линейное приближение модели. Чтобы получить объяснение, мы выполняем следующие действия: вносим отклонения в экземпляр, получаем прогнозы модели и используем эти прогнозы в качестве меток для изучения разреженной линейной модели, которая является локально достоверной. Весовые коэффициенты этой линейной модели используются в качестве значений \"важности признаков\". Для получения дополнительных сведений перейдите по ссылке ниже.",
|
||||
"mimicTitle": "Имитация (глобальные суррогатные объяснения)",
|
||||
"mimicDescription": "В основе этого модуля объяснения лежит идея обучения глобальных суррогатных моделей для имитации моделей черного ящика. Глобальная суррогатная модель — это внутренняя интерпретируемая модель, которая обучается для как можно более точного приближения прогнозов всех моделей черного ящика. Значения важности признаков — это значения важности признаков на основе модели для базовой суррогатной модели (LightGBM, линейная регрессия, стохастический градиентный спуск или дерево решений)",
|
||||
"pfiTitle": "Важность признаков, определяемая путем перестановок (PFI)",
|
||||
"pfiDescription": "Этот модуль объяснения случайным образом перемешивает данные во всем наборе данных по одному признаку за раз и вычисляет изменение метрики производительности для интереса (метрики производительности по умолчанию: F1 для двоичной классификации, оценка F1 с микроусреднением для многоклассовой классификации и средняя абсолютная погрешность для регрессии). Чем больше это изменение, тем важнее признак. Этот модуль объяснения может объяснить только общее поведение базовой модели, но не отдельные прогнозы. Значение важности признака представляет собой разницу в производительности модели, измеренную путем внесения отклонений в этот конкретный признак."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Välj en punkt om du vill se dess lokala förklaring",
|
||||
"defaultClassNames": "Klass {0}",
|
||||
"defaultFeatureNames": "Funktionen {0}",
|
||||
"absoluteAverage": "Genomsnittligt absolutvärde",
|
||||
"predictedClass": "Förutsagd klass",
|
||||
"datasetExplorer": "Datauppsättningsutforskaren",
|
||||
"dataExploration": "Datauppsättningsutforskning",
|
||||
"aggregateFeatureImportance": "Mängdfunktionsprioritet",
|
||||
"globalImportance": "Global betydelse",
|
||||
"explanationExploration": "Förklaringsutforskning",
|
||||
"individualAndWhatIf": "Enskild funktions prioritet och What If",
|
||||
"summaryImportance": "Sammanfattningsbetydelse",
|
||||
"featureImportance": "Funktionsrelevans",
|
||||
"featureImportanceOf": "Funktionsprioritet för {0}",
|
||||
"perturbationExploration": "Störningsutforskning",
|
||||
"localFeatureImportance": "Lokal funktionsrelevans",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Rensa val",
|
||||
"feature": "Funktion:",
|
||||
"intercept": "Skärningspunkt",
|
||||
"modelPerformance": "Modellprestanda",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Data: {0}",
|
||||
"importanceLabel": "Relevans: {0}",
|
||||
"predictedY": "Förutsade Y",
|
||||
"index": "Index",
|
||||
"dataGroupLabel": "Data",
|
||||
"output": "Utdata",
|
||||
"probabilityLabel": "Sannolikhet: {0}",
|
||||
"trueY": "Sant Y",
|
||||
"class": "klass: ",
|
||||
"xValue": "X-värde:",
|
||||
"yValue": "Y-värde:",
|
||||
"colorValue": "Färg:",
|
||||
"count": "Antal"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Klassomfattande vägning:",
|
||||
"info": "Information om korsklassberäkning",
|
||||
"overviewInfo": "Multiklassmodeller skapar en oberoende funktionsrelevansvektor för varje klass. Varje klass funktionsrelevansvektor visar vilka funktioner som gör klassen mer eller mindre sannolik. Du kan välja hur vikterna i funktionsrelevansvektorerna per klass ska summeras till ett enda värde:",
|
||||
"absoluteValInfo": "Genomsnittligt absolutvärde: Visar summan av funktionens betydelse i alla möjliga klasser, delat med antalet klasser",
|
||||
"predictedClassInfo": "Förutsagd klass: Visar funktionsrelevansvärdet för en angiven punkts förutsagda klass",
|
||||
"enumeratedClassInfo": "Uppräknade klassnamn: Visar endast funktionsrelevansen för alla datapunkter för den angivna klassens funktion.",
|
||||
"close": "Stäng",
|
||||
"crossClassWeights": "Vikter över klasser"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Skalat funktionsvärde",
|
||||
"low": "Låg",
|
||||
"high": "Hög",
|
||||
"featureLabel": "Funktion: {0}",
|
||||
"valueLabel": "Funktionsvärde: {0}",
|
||||
"importanceLabel": "Relevans: {0}",
|
||||
"predictedClassTooltip": "Förutsagd klass: {0}",
|
||||
"trueClassTooltip": "Sann klass: {0}",
|
||||
"predictedOutputTooltip": "Förutsagda utdata: {0}",
|
||||
"trueOutputTooltip": "Sanna utdata: {0}",
|
||||
"topKFeatures": "K-toppfunktioner:",
|
||||
"topKInfo": "Hur beräknas högsta K",
|
||||
"predictedValue": "Förutsagt värde",
|
||||
"predictedClass": "Förutsagd klass",
|
||||
"trueValue": "Sant värde",
|
||||
"trueClass": "Sann klass",
|
||||
"noColor": "Inget",
|
||||
"tooManyRows": "Den tillhandhållna datamängden är större än vad det här diagrammet kan stödja"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Klass: {0}",
|
||||
"sortBy": "Sortera efter",
|
||||
"noData": "Inga data",
|
||||
"absoluteGlobal": "Absolut global",
|
||||
"absoluteLocal": "Absolut lokal",
|
||||
"calculatingExplanation": "Beräkna förklaring"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Måste vara numeriskt",
|
||||
"integerError": "Måste vara ett heltal",
|
||||
"prediction": "Förutsägelse",
|
||||
"predictedProbability": "Förutsagd sannolikhet",
|
||||
"predictionLabel": "Förutsägelse: {0}",
|
||||
"probabilityLabel": "Sannolikhet: {0}",
|
||||
"noModelError": "Ange en driftsmodell för att utforska förutsägelser i ICE-kurvor.",
|
||||
"featurePickerLabel": "Funktion:",
|
||||
"minimumInputLabel": "Minimum:",
|
||||
"maximumInputLabel": "Maximum:",
|
||||
"stepInputLabel": "Steg:",
|
||||
"loadingMessage": "Läser in data …",
|
||||
"submitPrompt": "Skicka ett intervall så att du kan visa en ICE-rityta",
|
||||
"topLevelErrorMessage": "Fel i parametern",
|
||||
"errorPrefix": "Ett fel påträffades: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Läser in...",
|
||||
"perturbationLabel": "Störning:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Förutsagt värde: {0}",
|
||||
"predictedClassLabel": "Förutsagd klass: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Ingen gruppering",
|
||||
"groupPredicted": "Förutsagt Y",
|
||||
"groupTrue": "Sant Y",
|
||||
"groupBy": "Gruppera efter"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Diagramtyp:",
|
||||
"violinText": "Violin",
|
||||
"barText": "Liggande",
|
||||
"boxText": "Ruta",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "Den globala funktionsrelevansen beräknas genom att det absoluta värdet för funktionsrelevansen för alla punkter (L1-normalisering) beräknas. ",
|
||||
"multiclassImportanceAddendum": "Alla punkter tas med vid beräkning av en funktions betydelse för alla klasser. Ingen differentiell vägning används. Så en funktion som har stor negativ betydelse för många punkter, och som förutsägs inte tillhöra klass A, ökar t.ex. relevansen för funktionens klass A."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Lika med",
|
||||
"greaterThanComparison": "Större än",
|
||||
"greaterThanEqualToComparison": "Större än eller lika med",
|
||||
"lessThanComparison": "Mindre än",
|
||||
"lessThanEqualToComparison": "Mindre än eller lika med",
|
||||
"inTheRangeOf": "I intervallet",
|
||||
"categoricalIncludeValues": "Inkluderade värden:",
|
||||
"numericValue": "Värde",
|
||||
"numericalComparison": "Åtgärd",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Max",
|
||||
"min": "Min: {0}",
|
||||
"max": "Max: {0}",
|
||||
"uniqueValues": "antal unika värden: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Regressionsfel",
|
||||
"error": "Fel",
|
||||
"classificationOutcome": "Klassificeringsutfall",
|
||||
"truePositive": "Sant positiv identifiering",
|
||||
"trueNegative": "Sant negativ identifiering",
|
||||
"falsePositive": "Falsk positiv identifiering",
|
||||
"falseNegative": "Falsk negativ identifiering",
|
||||
"dataset": "Datauppsättning",
|
||||
"predictedProbabilities": "Förutsägelsesannolikheter",
|
||||
"none": "Antal"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Stäng",
|
||||
"defaultCustomRootName": "Kopia av rad {0}",
|
||||
"filterFeaturePlaceholder": "Sök efter funktioner"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Kohort",
|
||||
"defaultLabel": "Alla data"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Utforska de viktigaste k funktionerna som påverkar dina övergripande modellförutsägelser (global förklaring). Använd skjutreglaget för att visa fallande funktionsprioriteter. Välj upp till tre kohorter för att visa deras funktionsprioriteter sida vid sida. Klicka på någon av funktionerna i diagrammet om du vill se hur värdena för den valda funktionen påverkar modellförutsägelse.",
|
||||
"topAtoB": "Främsta {0}–{1} funktioner",
|
||||
"datasetCohorts": "Datauppsättningskohorter",
|
||||
"legendHelpText": "Växla kohorter på och av i området genom att klicka på förklaringselementen.",
|
||||
"sortBy": "Sortera efter",
|
||||
"viewDependencePlotFor": "Visa beroendediagram för:",
|
||||
"datasetCohortSelector": "Välj en datauppsättningskohort",
|
||||
"aggregateFeatureImportance": "Mängdfunktionsprioritet",
|
||||
"missingParameters": "På den här fliken måste prioritetsparametern för den lokala funktionen anges.",
|
||||
"weightOptions": "Vikter för klassbetydelse",
|
||||
"dependencePlotTitle": "Beroenderitytor",
|
||||
"dependencePlotHelperText": "Den här beroendekurvan visar förhållandet mellan värdet för en funktion och motsvarande betydelse för funktionen i en kohort.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Välj funktion",
|
||||
"datasetRequired": "Beroenderitytor kräver utvärderingsdatauppsättningen och den lokala funktionsviktsmatrisen."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Datastatistik",
|
||||
"datapoints": "{0} datapunkter",
|
||||
"features": "{0} funktioner",
|
||||
"filters": "{0} filter",
|
||||
"binaryClassifier": "Binär klassificerare",
|
||||
"regressor": "Regressor",
|
||||
"multiclassClassifier": "Multiklass-klassificerare",
|
||||
"datasetCohorts": "Datauppsättningskohorter",
|
||||
"editCohort": "Redigera kohort",
|
||||
"duplicateCohort": "Duplicera kohort",
|
||||
"addCohort": "Lägg till kohort",
|
||||
"copy": " kopiera"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Utvärdera modellens prestanda genom att utforska dina förutsägelsevärden och värdena för modellprestandamåtten. Du kan undersöka din modell ytterligare genom att titta på en jämförande analys av dess prestanda över olika kohorter eller undergrupper av din datauppsättning. Välj filter utmed y-värde och x-värde för att skära över olika dimensioner. Välj växeln i diagrammet om du vill ändra diagramtyp.",
|
||||
"modelStatistics": "Modellstatistik",
|
||||
"cohortPickerLabel": "Välj en datauppsättningskohort att utforska",
|
||||
"missingParameters": "Den här fliken kräver att matrisen med förutsagda värden från modellen anges.",
|
||||
"missingTrueY": "Statistik för modellprestanda kräver att de verkliga utfallen anges utöver de förväntade utfallen"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y-värde",
|
||||
"numberOfDatapoints": "Antalet datapunkter",
|
||||
"xValue": "X-värde",
|
||||
"rowIndex": "Radindex",
|
||||
"featureImportance": "Funktionsprioritet",
|
||||
"countTooltipPrefix": "Antal: {0}",
|
||||
"count": "Antal",
|
||||
"featurePrefix": "Funktion",
|
||||
"importancePrefix": "Prioritet",
|
||||
"cohort": "Kohort",
|
||||
"howToRead": "Så här läser du det här diagrammet"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Utforska din datauppsättningsstatistik genom att välja olika filter längsmed X-, Y- och färgaxeln för att segmentera dina data längs olika dimensioner. Skapa datauppsättningskohorter ovan för att analysera datauppsättningsstatistik med filter som förutsedda resultat, datauppsättningsfunktioner och felgrupper. Använd kugghjulsikonen i det övre högra hörnet i diagrammet för att ändra diagramtyper.",
|
||||
"colorValue": "Färgvärde",
|
||||
"individualDatapoints": "Enskilda datapunkter",
|
||||
"aggregatePlots": "Sammansatta områden",
|
||||
"chartType": "Diagramtyp",
|
||||
"missingParameters": "En utvärderingsdatauppsättning krävs för den här fliken.",
|
||||
"noColor": "Inget"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Funktionsprioritet för",
|
||||
"placeholder": "Klicka på en funktion i stapeldiagrammet ovan för att visa dess beroendekurva"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Du kan välja en datapunkt genom att klicka på punktdiagrammet för att visa dess lokala funktions prioritetsvärden (lokal förklaring) och ett diagram över enskild villkorsstyrd förväntning (ICE) nedan. Skapa en hypotetiskisk What If-datapunkt genom att använda panelen till höger för att rubba funktioner för en känd datapunkt. Funktionens prioritetsvärden baseras på många approximationer och är inte orsaken till förutsägelserna. Utan strikt matematiskt robusthet av orsakssambandet så uppmanas användare att inte fatta verkliga beslut baserat på det här verktyget.",
|
||||
"panelPlaceholder": "En modell krävs för att göra förutsägelser för nya datapunkter.",
|
||||
"cohortPickerLabel": "Välj en datauppsättningskohort att utforska",
|
||||
"scatterLegendText": "Växla datapunkter på och av i diagrammet genom att klicka på förklaringselementen.",
|
||||
"realPoint": "Verkliga datapunkter",
|
||||
"noneSelectedYet": "Inga har markerats ännu",
|
||||
"whatIfDatapoints": "What If-datapunkter",
|
||||
"noneCreatedYet": "Inget har skapats ännu",
|
||||
"showLabel": "Visa:",
|
||||
"featureImportancePlot": "Funktionsprioritetskurva",
|
||||
"icePlot": "Diagram över enskild villkorsstyrd förväntan (ICE)",
|
||||
"featureImportanceLackingParameters": "Ange lokal funktionsrelevans för att se hur varje funktion påverkar enskilda förutsägelser.",
|
||||
"featureImportanceGetStartedText": "Välj en punkt för att visa funktionsprioritet",
|
||||
"iceLackingParameters": "ICE-områden kräver en operationell modell för att göra förutsägelser för hypotetiska datapunkter.",
|
||||
"IceGetStartedText": "Välj en punkt eller skapa en What If-punkt för att visa ICE-kurvor",
|
||||
"whatIfDatapoint": "What If-datapunkt",
|
||||
"whatIfHelpText": "Välj en punkt i diagrammet eller ange ett okänt datapunktsindex manuellt för perturb och spara som en ny konsekvenspunkt.",
|
||||
"indexLabel": "Dataindex för perturb",
|
||||
"rowLabel": "Rad {0}",
|
||||
"whatIfNameLabel": "What If-datapunktsnamn",
|
||||
"featureValues": "Funktionsvärden",
|
||||
"predictedClass": "Förutsagd klass: ",
|
||||
"predictedValue": "Förutsagt värde: ",
|
||||
"probability": "Sannolikhet: ",
|
||||
"trueClass": "Sann klass: ",
|
||||
"trueValue": "Sant värde: ",
|
||||
"trueValue.comment": "prefix till faktisk etikett för regression",
|
||||
"newPredictedClass": "Ny förutsagd klass: ",
|
||||
"newPredictedValue": "Nytt förutsagt värde: ",
|
||||
"newProbability": "Ny sannolikhet: ",
|
||||
"saveAsNewPoint": "Spara som ny punkt",
|
||||
"saveChanges": "Spara ändringarna",
|
||||
"loading": "Läser in...",
|
||||
"classLabel": "Klass: {0}",
|
||||
"minLabel": "Min",
|
||||
"maxLabel": "Max",
|
||||
"stepsLabel": "Steg",
|
||||
"disclaimer": "Ansvarsfriskrivning: dessa är förklaringar som baseras på många approximationer och är inte orsaken till förutsägelser. Utan en strikt matematiskt robusthet av orsakssamband, så rekommenderar vi inte användare att fatta verkliga beslut baserat på det här verktyget.",
|
||||
"missingParameters": "En utvärderingsdatauppsättning krävs för den här fliken.",
|
||||
"selectionLimit": "Högst 3 valda punkter",
|
||||
"classPickerLabel": "Klass",
|
||||
"tooltipTitleMany": "Viktigaste {0} förväntade klasser",
|
||||
"whatIfTooltipTitle": "What If förväntade klasser",
|
||||
"tooltipTitleFew": "Förutsagda klasser",
|
||||
"probabilityLabel": "Sannolikhet",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "Värdet måste vara numeriskt",
|
||||
"icePlotHelperText": "ICE-diagram visar hur den valda datapunktens förutsägelsevärden ändras längs ett intervall med funktionsvärden mellan ett minimi- och maxvärde."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Välj filter",
|
||||
"TreatAsCategorical": "Hantera som kategorisk",
|
||||
"addFilter": "Lägg till filter",
|
||||
"addedFilters": "Tillagda filter",
|
||||
"noAddedFilters": "Inga filter har lagts till ännu",
|
||||
"defaultFilterState": "Välj ett filter för att lägga till parametrar till din datauppsättningskohort.",
|
||||
"cohortNameLabel": "Datauppsättningens kohortnamn",
|
||||
"cohortNamePlaceholder": "Namnge din kohort",
|
||||
"save": "Spara",
|
||||
"delete": "Ta bort",
|
||||
"cancel": "Avbryt",
|
||||
"cohortNameError": "Kohortnamn saknas",
|
||||
"placeholderName": "Kohort {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Välj",
|
||||
"ditherLabel": "Ska använda raster",
|
||||
"selectFilter": "Välj axelvärde",
|
||||
"selectFeature": "Välj funktion",
|
||||
"binLabel": "Använd gruppering för data",
|
||||
"TreatAsCategorical": "Hantera som kategorisk",
|
||||
"numOfBins": "Antal diskretiseringar",
|
||||
"groupByCohort": "Gruppera efter kohort",
|
||||
"selectClass": "Välj klass",
|
||||
"countHelperText": "Ett histogram med antalet punkter"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Förutsagd sannolikhet",
|
||||
"predictedY": "Förutsade Y",
|
||||
"evalData": "Utvärderingsdatauppsättning",
|
||||
"localFeatureImportance": "Lokal funktionsrelevans",
|
||||
"inconsistentDimensions": "Inkonsekventa dimensioner. {0} har dimensionerna {1}, förväntade {2}",
|
||||
"notNonEmpty": "{0} indata är inte en matris som inte är tom",
|
||||
"varyingLength": "Inkonsekventa dimensioner. {0} har element av varierande längd",
|
||||
"notArray": "{0} är inte en matris. Förväntad matris med dimensionen {1}",
|
||||
"errorHeader": "Vissa indataparametrar var inkonsekventa och kommer inte att användas: ",
|
||||
"datasizeWarning": "Utvärderingsdatauppsättningen är för stor för att kunna visas effektivt i vissa diagram. Lägg till filter för att minska storleken på kohorten. ",
|
||||
"datasizeError": "Den valda kohorten är för stor. Lägg till filter för att minska storleken på kohorten.",
|
||||
"addFilters": "Lägg till filter"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": "<= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " inkluderar {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} och {1} andra"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R-kvadrat: {0}",
|
||||
"meanPrediction": "Genomsnittlig förutsägelse {0}",
|
||||
"accuracy": "Noggrannhet: {0}",
|
||||
"precision": "Precision: {0}",
|
||||
"recall": "Träffsäkerhet: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Utforska de viktigaste k funktionerna som påverkar dina övergripande modellförutsägelser. Använd skjutreglaget för att visa fallande funktionsvikter."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Vad betyder dessa förklaringar?",
|
||||
"clickHere": "Läs mer",
|
||||
"shapTitle": "Shapley-värden",
|
||||
"shapDescription": "Den här förklaringen använder SHAP, som är ett spelteoretiskt tillvägagångssätt för att förklara modeller, där betydelsen av funktionsuppsättningar mäts genom att dölja de funktionerna från modellen genom marginalisering. Klicka på länken nedan om du vill läsa mer.",
|
||||
"limeTitle": "LIME (lokala tolkningsbara modellagnostiska förklaringar)",
|
||||
"limeDescription": "Den här förklaringen använder sig av LIME, vilket ger en linjär approximering av modellen. För att få en förklaring så gör vi följande: rubba instansen, Hämta modellförutsägelser och använd dessa förutsägelser som etiketter för att lära en gles linjär modell som är lokalt trogen. Vikterna för den här linjära modellen används som funktionsbetydelser. Klicka på länken nedan om du vill läsa mer.",
|
||||
"mimicTitle": "Efterlikna (globala surrogatförklaringar)",
|
||||
"mimicDescription": "Den här förklaringen baseras på idén att träna globala surrogatmodeller att efterlikna svart låda-modeller. En global surrogatmodell är en i sig tolkbar modell som tränas att närma sig förutsägelserna för en svart låda-modell så exakt som möjligt. Funktionsprioritetsvärden är modellbaserade funktionsvärden för den underliggande surrogatmodellen (LightGBM, linjär regression, Stochastic Gradient Descent eller beslutsträd)",
|
||||
"pfiTitle": "Permuteringsfunktionsprioritet (PFI)",
|
||||
"pfiDescription": "Den här förklaringen blandar slumpmässigt data en funktion i taget för hela datauppsättningen och beräknar hur mycket prestandamåttet för ränta ändras (standard prestandamått: F1 för binär klassificering, F1-poäng med mikromedelvärde för multiklass-klassificering och medelvärde för absolut fel för regression). Ju större ändringen är, desto viktigare är funktionen. Den här förklaringen kan bara förklara den underliggande modellens övergripande beteende, men inte enskilda förutsägelser. Funktionsprioritetsvärdet för en funktion representerar delta i modellens prestanda genom att rubba den funktionen."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "Yerel açıklamasını görmek için bir nokta seçin",
|
||||
"defaultClassNames": "{0} sınıfı",
|
||||
"defaultFeatureNames": "Özellik {0}",
|
||||
"absoluteAverage": "Mutlak değer ortalaması",
|
||||
"predictedClass": "Tahmin edilen sınıf",
|
||||
"datasetExplorer": "Veri Kümesi Gezgini",
|
||||
"dataExploration": "Veri Kümesi Keşfi",
|
||||
"aggregateFeatureImportance": "Kümelenmiş Özellik Önem Derecesi",
|
||||
"globalImportance": "Genel Önem Derecesi",
|
||||
"explanationExploration": "Açıklama Araştırması",
|
||||
"individualAndWhatIf": "Tek Özellik Önem Derecesi ve Durum Değerlendirmesi",
|
||||
"summaryImportance": "Özet Önem Derecesi",
|
||||
"featureImportance": "Özellik Önem Derecesi",
|
||||
"featureImportanceOf": "{0} özellik önem derecesi",
|
||||
"perturbationExploration": "Sapma Araştırması",
|
||||
"localFeatureImportance": "Yerel Özellik Önem Derecesi",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "Seçimi temizle",
|
||||
"feature": "Özellik:",
|
||||
"intercept": "kesim noktası",
|
||||
"modelPerformance": "Model Performansı",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "Veri: {0}",
|
||||
"importanceLabel": "Önem derecesi: {0}",
|
||||
"predictedY": "Tahmin edilen Y",
|
||||
"index": "Dizin",
|
||||
"dataGroupLabel": "Veri",
|
||||
"output": "Çıktı",
|
||||
"probabilityLabel": "Olasılık: {0}",
|
||||
"trueY": "Doğru Y",
|
||||
"class": "sınıf: ",
|
||||
"xValue": "X değeri:",
|
||||
"yValue": "Y değeri:",
|
||||
"colorValue": "Renk:",
|
||||
"count": "Sayı"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "Sınıflar arası ağırlık:",
|
||||
"info": "Sınıflar arası hesaplama hakkında bilgiler",
|
||||
"overviewInfo": "Çok sınıflı modeller, her sınıf için bağımsız bir özellik önem derecesi vektörü oluşturur. Her sınıfın özellik önem derecesi vektörü, sınıfı hangi özelliklerin daha olası veya daha az olası yaptığını gösterir. Sınıf başına özellik önem derecesi vektörlerinin ağırlığının tek bir değerde nasıl özetleneceğini seçebilirsiniz:",
|
||||
"absoluteValInfo": "Mutlak değer ortalaması: Özelliğin tüm olası sınıflardaki önem derecesi toplamının sınıf sayısına bölümünü gösterir",
|
||||
"predictedClassInfo": "Tahmin edilen sınıf: Belirli bir noktanın tahmin edilen sınıfı için özellik önem derecesi değerini gösterir",
|
||||
"enumeratedClassInfo": "Listelenen sınıf adları: Tüm veri noktalarında yalnızca belirtilen sınıfın özellik önem derecesi değerlerini gösterir.",
|
||||
"close": "Kapat",
|
||||
"crossClassWeights": "Sınıflar arası ağırlıklar"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "Ölçeklendirilmiş Özellik Değeri",
|
||||
"low": "Düşük",
|
||||
"high": "Yüksek",
|
||||
"featureLabel": "Özellik: {0}",
|
||||
"valueLabel": "Özellik değeri: {0}",
|
||||
"importanceLabel": "Önem derecesi: {0}",
|
||||
"predictedClassTooltip": "Tahmin Edilen Sınıf: {0}",
|
||||
"trueClassTooltip": "Doğru Sınıf: {0}",
|
||||
"predictedOutputTooltip": "Tahmin Edilen Çıkış: {0}",
|
||||
"trueOutputTooltip": "Doğru Çıkış: {0}",
|
||||
"topKFeatures": "En Yüksek K Özellikleri:",
|
||||
"topKInfo": "İlk k nasıl hesaplanır?",
|
||||
"predictedValue": "Tahmin Edilen Değer",
|
||||
"predictedClass": "Tahmin Edilen Sınıf",
|
||||
"trueValue": "Gerçek Değer",
|
||||
"trueClass": "Doğru Sınıf",
|
||||
"noColor": "Yok",
|
||||
"tooManyRows": "Sağlanan veri kümesi, bu grafiğin destekleyebileceğinden daha büyük"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "Sınıf: {0}",
|
||||
"sortBy": "Sıralama Ölçütü",
|
||||
"noData": "Veri Yok",
|
||||
"absoluteGlobal": "Mutlak genel",
|
||||
"absoluteLocal": "Mutlak yerel",
|
||||
"calculatingExplanation": "Açıklama hesaplanıyor"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "Sayısal bir değer olmalıdır",
|
||||
"integerError": "Bir tamsayı olmalıdır",
|
||||
"prediction": "Tahmin",
|
||||
"predictedProbability": "Tahmin edilen olasılık",
|
||||
"predictionLabel": "Tahmin: {0}",
|
||||
"probabilityLabel": "Olasılık: {0}",
|
||||
"noModelError": "ICE çizimlerindeki tahminleri keşfetmek için lütfen işlevselleştirilmiş bir model belirtin.",
|
||||
"featurePickerLabel": "Özellik:",
|
||||
"minimumInputLabel": "Minimum:",
|
||||
"maximumInputLabel": "Maksimum:",
|
||||
"stepInputLabel": "Adımlar:",
|
||||
"loadingMessage": "Veriler yükleniyor...",
|
||||
"submitPrompt": "ICE çizimini görüntülemek için bir aralık gönderin",
|
||||
"topLevelErrorMessage": "Parametrede hata",
|
||||
"errorPrefix": "Hatayla karşılaşıldı: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "Yükleniyor...",
|
||||
"perturbationLabel": "Sapma:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "Tahmin edilen değer: {0}",
|
||||
"predictedClassLabel": "Tahmin edilen sınıf: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "Gruplandırma yok",
|
||||
"groupPredicted": "Tahmin edilen Y",
|
||||
"groupTrue": "Doğru Y",
|
||||
"groupBy": "Gruplandır"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "Grafik türü:",
|
||||
"violinText": "Keman",
|
||||
"barText": "Çubuk",
|
||||
"boxText": "Kutu",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "Genel özellik önem derecesi, tüm noktaların özellik önem derecelerine ait mutlak değerinin ortalaması alınarak hesaplanır (L1 normalleştirme). ",
|
||||
"multiclassImportanceAddendum": "Tüm sınıflar için özelliğin önem derecesi hesaplanırken tüm noktalar dahil edilir ve fark ağırlığı kullanılmaz Bu nedenle, 'A Sınıfı' olmadığı tahmin edilen çok sayıda nokta için büyük negatif önem derecesine sahip bir özellik, bu özelliğin 'A Sınıfı' önem derecesini büyük ölçüde artırır."
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "Eşittir",
|
||||
"greaterThanComparison": "Büyüktür",
|
||||
"greaterThanEqualToComparison": "Büyük veya eşit",
|
||||
"lessThanComparison": "Küçüktür",
|
||||
"lessThanEqualToComparison": "Küçük veya eşit",
|
||||
"inTheRangeOf": "Şu aralıkta:",
|
||||
"categoricalIncludeValues": "Eklenen değerler:",
|
||||
"numericValue": "Değer",
|
||||
"numericalComparison": "İşlem",
|
||||
"minimum": "Minimum",
|
||||
"maximum": "Maksimum",
|
||||
"min": "En az: {0}",
|
||||
"max": "En fazla: {0}",
|
||||
"uniqueValues": "benzersiz değerlerin sayısı: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "Regresyon hatası",
|
||||
"error": "Hata",
|
||||
"classificationOutcome": "Sınıflandırma sonucu",
|
||||
"truePositive": "Gerçek pozitif",
|
||||
"trueNegative": "Gerçek negatif",
|
||||
"falsePositive": "Hatalı pozitif sonuç",
|
||||
"falseNegative": "Hatalı negatif sonuç",
|
||||
"dataset": "Veri kümesi",
|
||||
"predictedProbabilities": "Tahmin olasılıkları",
|
||||
"none": "Sayı"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "Kapat",
|
||||
"defaultCustomRootName": "{0} satırının kopyası",
|
||||
"filterFeaturePlaceholder": "Özelliklerde arama yapın"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "Kohort",
|
||||
"defaultLabel": "Tüm veriler"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "Genel model tahminlerinizi etkileyen en önemli k özelliği (yani genel açıklama) keşfedin. Azalan özellik önem değerlerini görüntülemek için kaydırıcıyı kullanın. Özellik önem değerlerini yan yana görmek için en fazla üç kohort seçin. Seçilen özelliğin değerlerinin model tahminini nasıl etkilediğini görmek için grafik çubuklarından birine tıklayın.",
|
||||
"topAtoB": "İlk {0}-{1} özellik",
|
||||
"datasetCohorts": "Veri kümesi kohortları",
|
||||
"legendHelpText": "Gösterge öğelerine tıklayarak çizimdeki kohortları açıp kapatın.",
|
||||
"sortBy": "Sıralama ölçütü:",
|
||||
"viewDependencePlotFor": "Şunun için bağımlılık çizimini görüntüle:",
|
||||
"datasetCohortSelector": "Veri kümesi kohortu seçin",
|
||||
"aggregateFeatureImportance": "Kümelenmiş Özellik Önem Derecesi",
|
||||
"missingParameters": "Bu sekme için yerel özellik önem derecesi parametresinin sağlanması gerekir.",
|
||||
"weightOptions": "Sınıf önem ağırlıkları",
|
||||
"dependencePlotTitle": "Bağımlılık Çizimleri",
|
||||
"dependencePlotHelperText": "Bu bağımlılık çizimi, bir özelliğin değeri ile bir kohort arasındaki özelliğin önem derecesi arasındaki ilişkiyi gösterir.",
|
||||
"dependencePlotFeatureSelectPlaceholder": "Özellik seçin",
|
||||
"datasetRequired": "Bağımlılık çizimleri için değerlendirme veri kümesi ve yerel özellik önem dizesi gereklidir."
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "Veri İstatistikleri",
|
||||
"datapoints": "{0} veri noktası",
|
||||
"features": "{0} özellik",
|
||||
"filters": "{0} filtre",
|
||||
"binaryClassifier": "İkili Sınıflandırıcı",
|
||||
"regressor": "Regresör",
|
||||
"multiclassClassifier": "Çoklu Sınıf Sınıflandırıcısı",
|
||||
"datasetCohorts": "Veri Kümesi Kohortları",
|
||||
"editCohort": "Kohortu Düzenle",
|
||||
"duplicateCohort": "Yinelenen Kohort",
|
||||
"addCohort": "Kohort Ekle",
|
||||
"copy": " kopyası"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "Tahmin değerlerinizin dağılımını ve model performans ölçümlerinizin değerlerini keşfederek modelinizin performansını değerlendirin. Veri kümenizin farklı kohortlarında veya alt gruplarındaki performansının karşılaştırmalı analizine bakarak modelinizi daha ileri düzeyde araştırabilirsiniz. Farklı boyutlarda kesmek için y değeri ve x değeri boyunca filtreleri seçin. Grafik türünü değiştirmek için grafikteki dişliyi seçin.",
|
||||
"modelStatistics": "Model İstatistikleri",
|
||||
"cohortPickerLabel": "Keşfedilecek veri kümesi kohortunu seçin",
|
||||
"missingParameters": "Bu sekme için modeldeki tahmini değerler dizisinin sağlanması gerekir.",
|
||||
"missingTrueY": "Model performans istatistikleri için tahmini sonuçların yanı sıra, doğru sonuçların da sağlanması gerekir"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y değeri",
|
||||
"numberOfDatapoints": "Veri noktası sayısı",
|
||||
"xValue": "X değeri",
|
||||
"rowIndex": "Satır dizini",
|
||||
"featureImportance": "Özelliğin önem derecesi",
|
||||
"countTooltipPrefix": "Sayı: {0}",
|
||||
"count": "Sayı",
|
||||
"featurePrefix": "Özellik",
|
||||
"importancePrefix": "Önem derecesi",
|
||||
"cohort": "Kohort",
|
||||
"howToRead": "Bu grafiği okuma"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "Verilerinizi farklı boyutlarda dilimlemek için X, Y ve renk ekseni boyunca farklı filtreler seçerek veri kümesi istatistiklerinizi keşfedin. Veri kümesi istatistiklerini tahmin edilen sonuç, veri kümesi özellikleri ve hata grupları gibi filtrelerle analiz etmek için yukarıda veri kümesi kohortları oluşturun. Grafik türlerini değiştirmek için grafiğin sağ üst köşesindeki dişli simgesini kullanın.",
|
||||
"colorValue": "Renk değeri",
|
||||
"individualDatapoints": "Ayrı ayrı veri noktaları",
|
||||
"aggregatePlots": "Kümelenmiş çizimler",
|
||||
"chartType": "Grafik türü",
|
||||
"missingParameters": "Bu sekme için değerlendirme veri kümesi sağlanması gerekir.",
|
||||
"noColor": "Yok"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "Özellik önem derecesi:",
|
||||
"placeholder": "Bağımlılık çizimini görüntülemek için yukarıdaki çubuk grafiğinde bulunan bir özelliğe tıklayın"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "Aşağıda yerel özellik önem değerlerini (yerel açıklama) ve bireysel koşullu beklenti (ICE) çizimini görüntülemek için dağılıma tıklayarak bir veri noktası seçebilirsiniz. Bilinen bir veri noktasının özelliklerini karıştırmak için sağdaki paneli kullanarak varsayımsal bir durum veri noktası oluşturun. Özellik önem değerleri birçok yaklaşıma dayanır ve tahminlerin \"nedeni\" değildir. Nedensel çıkarıma dayalı katı matematiksel sağlamlık olmadan, kullanıcıların bu araca dayalı olarak gerçek hayata yönelik kararlar vermelerini önermiyoruz.",
|
||||
"panelPlaceholder": "Yeni veri noktalarına yönelik tahminlerde bulunmak için bir model gerekir.",
|
||||
"cohortPickerLabel": "Keşfedilecek veri kümesi kohortunu seçin",
|
||||
"scatterLegendText": "Gösterge öğelerine tıklayarak çizimdeki veri noktalarını açıp kapatın.",
|
||||
"realPoint": "Gerçek veri noktaları",
|
||||
"noneSelectedYet": "Henüz seçilmedi",
|
||||
"whatIfDatapoints": "Durum değerlendirmesi veri noktaları",
|
||||
"noneCreatedYet": "Henüz oluşturulmadı",
|
||||
"showLabel": "Göster:",
|
||||
"featureImportancePlot": "Özellik önem derecesi çizimi",
|
||||
"icePlot": "Tek koşullu beklenti (ICE) çizimi",
|
||||
"featureImportanceLackingParameters": "Her bir özelliğin tahminleri nasıl etkilediğini görmek için yerel özellik önem derecelerini belirtin.",
|
||||
"featureImportanceGetStartedText": "Özellik önem derecesini görüntülemek için bir nokta seçin",
|
||||
"iceLackingParameters": "ICE çizimleri, kuramsal veri noktalarına yönelik tahminler yapmak için kullanıma hazır hale getirilmiş bir model gerektirir.",
|
||||
"IceGetStartedText": "ICE çizimlerini görüntülemek için bir nokta seçin veya Durum değerlendirmesi noktası oluşturun",
|
||||
"whatIfDatapoint": "Durum değerlendirmesi veri noktası",
|
||||
"whatIfHelpText": "Çizimde bir nokta seçin veya saptırmak ve yeni bir Durum değerlendirmesi noktası olarak kaydetmek için bilinen veri noktası dizinini kendiniz girin.",
|
||||
"indexLabel": "Karıştırılacak veri dizini",
|
||||
"rowLabel": "{0}. satır ",
|
||||
"whatIfNameLabel": "Durum değerlendirmesi veri noktası adı",
|
||||
"featureValues": "Özellik değerleri",
|
||||
"predictedClass": "Tahmin edilen sınıf: ",
|
||||
"predictedValue": "Tahmin edilen değer: ",
|
||||
"probability": "Olasılık: ",
|
||||
"trueClass": "Doğru sınıf: ",
|
||||
"trueValue": "Doğru değer: ",
|
||||
"trueValue.comment": "gerileme için gerçek etiket ön eki",
|
||||
"newPredictedClass": "Tahmin edilen yeni sınıf: ",
|
||||
"newPredictedValue": "Tahmin edilen yeni değer: ",
|
||||
"newProbability": "Yeni olasılık: ",
|
||||
"saveAsNewPoint": "Yeni nokta olarak kaydet",
|
||||
"saveChanges": "Değişiklikleri kaydet",
|
||||
"loading": "Yükleniyor...",
|
||||
"classLabel": "Sınıf: {0}",
|
||||
"minLabel": "Minimum",
|
||||
"maxLabel": "Maksimum",
|
||||
"stepsLabel": "Adımlar",
|
||||
"disclaimer": "Yasal uyarı: Bunlar birçok tahmine dayanan açıklamalar olup tahminlerin “nedeni” değildir. Nedensel çıkarımın sağlam matematiksel temeli olmadan kullanıcıların bu araca dayalı olarak gerçek hayata yönelik karar vermeleri önerilmez.",
|
||||
"missingParameters": "Bu sekme için değerlendirme veri kümesi sağlanması gerekir.",
|
||||
"selectionLimit": "En fazla 3 nokta seçilebilir",
|
||||
"classPickerLabel": "Sınıf",
|
||||
"tooltipTitleMany": "En iyi {0} tahmin edilen sınıf",
|
||||
"whatIfTooltipTitle": "Durum tahmini sınıfları",
|
||||
"tooltipTitleFew": "Tahmin edilen sınıflar",
|
||||
"probabilityLabel": "Olasılık",
|
||||
"deltaLabel": "Delta",
|
||||
"nonNumericValue": "Değer sayısal olmalıdır",
|
||||
"icePlotHelperText": "ICE çizimleri, seçilen veri noktasının tahmin değerlerinin en küçük ve en büyük değerler arasındaki özellik değerleri aralığı boyunca nasıl değiştiğini gösterir."
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "Filtre Seçin",
|
||||
"TreatAsCategorical": "Kategorik olarak işle",
|
||||
"addFilter": "Filtre Ekle",
|
||||
"addedFilters": "Eklenen Filtreler",
|
||||
"noAddedFilters": "Henüz filtre eklenmedi",
|
||||
"defaultFilterState": "Veri kümesi kohortunuza parametre eklemek için bir filtre seçin.",
|
||||
"cohortNameLabel": "Veri kümesi kohortu adı",
|
||||
"cohortNamePlaceholder": "Kohortunuzu adlandırın",
|
||||
"save": "Kaydet",
|
||||
"delete": "Sil",
|
||||
"cancel": "İptal",
|
||||
"cohortNameError": "Kohort adı eksik",
|
||||
"placeholderName": "{0} kohortu"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "Seçin",
|
||||
"ditherLabel": "Titreşmeli",
|
||||
"selectFilter": "Eksen değerinizi seçin",
|
||||
"selectFeature": "Özellik Seçin",
|
||||
"binLabel": "Verilere gruplama uygula",
|
||||
"TreatAsCategorical": "Kategorik olarak işle",
|
||||
"numOfBins": "Bölme sayısı",
|
||||
"groupByCohort": "Kohorta göre gruplandır",
|
||||
"selectClass": "Sınıf seçin",
|
||||
"countHelperText": "Nokta sayısının histogramı"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "Tahmini olasılık",
|
||||
"predictedY": "Tahmini Y",
|
||||
"evalData": "Değerlendirme veri kümesi",
|
||||
"localFeatureImportance": "Yerel özellik önem derecesi",
|
||||
"inconsistentDimensions": "Boyutlar tutarsız. {0}, {1} boyutlarına sahip. {2} bekleniyordu",
|
||||
"notNonEmpty": "{0} girişi, boş olmayan bir dizi değil",
|
||||
"varyingLength": "Boyutlar tutarsız. {0}, çeşitli uzunlukta öğeler içeriyor",
|
||||
"notArray": "{0}, bir dizi değil. Beklenen boyut dizisi {1}",
|
||||
"errorHeader": "Bazı giriş parametreleri tutarsız olduğundan kullanılmayacak: ",
|
||||
"datasizeWarning": "Değerlendirme veri kümesi, bazı grafiklerde etkili bir şekilde görüntülenemeyecek kadar büyük. Lütfen kohortun boyutunu azaltmak için filtreler ekleyin. ",
|
||||
"datasizeError": "Seçilen kohort çok büyük. Kohortun boyutunu azaltmak için lütfen filtre ekleyin.",
|
||||
"addFilters": "Filtre ekle"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " {0} içerir ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} ve diğer {1} kişi"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R-kare: {0}",
|
||||
"meanPrediction": "Ortalama tahmin {0}",
|
||||
"accuracy": "Doğruluk: {0}",
|
||||
"precision": "Duyarlık: {0}",
|
||||
"recall": "Geri çağırma: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "Genel model tahminlerinizi etkileyen en önemli k özelliği keşfedin. Azalan özellik önceliklerini göstermek için kaydırıcıyı kullanın."
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "Bu açıklamalar ne anlama geliyor?",
|
||||
"clickHere": "Daha fazla bilgi",
|
||||
"shapTitle": "Shapley değerleri",
|
||||
"shapDescription": "Bu açıklayıcı, modelleri açıklamaya yönelik bir oyun teorisi yaklaşımı olan SHAP'yi kullanır. Bu yöntemde özellik kümelerinin önemi, bu özellikleri marjinalleştirme yoluyla modelden \"gizleyerek\" ölçülür. Daha fazla bilgi için aşağıdaki bağlantıya tıklayın.",
|
||||
"limeTitle": "LIME (Yerel Yorumlanabilir Modelden Bağımsız Açıklamalar)",
|
||||
"limeDescription": "Bu açıklayıcı, modelin doğrusal bir yaklaşımını sağlayan LIME kullanır. Bir açıklama almak için, aşağıdakileri yaparız: örneği karıştırır, model tahminleri alır ve yerel olarak sadık olan seyrek bir doğrusal modeli öğrenmek için bu tahminleri etiket olarak kullanırız. Bu doğrusal modelin ağırlıkları 'özellik önemi' olarak kullanılır. Daha fazla bilgi için aşağıdaki bağlantıya tıklayın.",
|
||||
"mimicTitle": "Taklit (Genel Vekil Açıklamaları)",
|
||||
"mimicDescription": "Bu açıklayıcı, kara kutu modellerini taklit etmek için küresel vekil modellerini eğitme fikrine dayanmaktadır. Küresel bir vekil modeli, herhangi bir kara kutu modelini mümkün olduğunca doğru olarak tahmin etmek için eğitilmiş içsel olarak yorumlanabilir bir modeldir. Özellik önem değerleri, temel vekil modelinizin (LightGBM veya Lineer Regresyon veya Stokastik Gradyan İniş veya Karar Ağacı) model tabanlı özellik önem değerleridir.",
|
||||
"pfiTitle": "Permütasyon Özelliği Önem Derecesi (PFI)",
|
||||
"pfiDescription": "Bu açıklayıcı, tüm veri kümesi için verileri tek tek özellik bazında rastgele karıştırır ve ilgilenilen performans ölçümündeki değişim oranını hesaplar (varsayılan performans ölçümleri: ikili sınıflandırma için F1, çok sınıflı sınıflandırma için mikro ortalamalı F1 Puanı ve regresyon için ortalama mutlak hata). Değişiklik ne kadar büyükse, bu özellik o kadar önemlidir. Bu açıklayıcı yalnızca temel modelin genel davranışını açıklayabilir, ancak tek tek tahminleri açıklamaz. Bir özelliğin önem değeri, bu özelliği değiştirerek modelin performansında elde edilen deltayı temsil eder."
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "选择一个点以查看其本地说明",
|
||||
"defaultClassNames": "类 {0}",
|
||||
"defaultFeatureNames": "特征 {0}",
|
||||
"absoluteAverage": "绝对值的平均值",
|
||||
"predictedClass": "预测类",
|
||||
"datasetExplorer": "数据集资源管理器",
|
||||
"dataExploration": "数据集浏览",
|
||||
"aggregateFeatureImportance": "聚合特征重要性",
|
||||
"globalImportance": "全局重要性",
|
||||
"explanationExploration": "解释探索",
|
||||
"individualAndWhatIf": "单个特征重要性和模拟",
|
||||
"summaryImportance": "摘要重要性",
|
||||
"featureImportance": "特征重要性",
|
||||
"featureImportanceOf": "特征重要性为 {0}",
|
||||
"perturbationExploration": "小改动探索",
|
||||
"localFeatureImportance": "本地特征重要性",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "清除选定内容",
|
||||
"feature": "功能:",
|
||||
"intercept": "截获",
|
||||
"modelPerformance": "模型性能",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "数据: {0}",
|
||||
"importanceLabel": "重要性: {0}",
|
||||
"predictedY": "预测的 Y",
|
||||
"index": "索引",
|
||||
"dataGroupLabel": "数据",
|
||||
"output": "输出",
|
||||
"probabilityLabel": "概率: {0}",
|
||||
"trueY": "真实 Y",
|
||||
"class": "类:",
|
||||
"xValue": "X 值:",
|
||||
"yValue": "Y 值:",
|
||||
"colorValue": "颜色:",
|
||||
"count": "项计数"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "交叉类权重:",
|
||||
"info": "有关跨类计算的信息",
|
||||
"overviewInfo": "多类分类模型为每个类生成一个独立的特征重要性向量。每个类的特征重要性向量说明了哪些特征使类的可能性更高或更低。可以选择如何将每个类特征重要性向量的权重汇总为单个值:",
|
||||
"absoluteValInfo": "绝对值的平均值: 显示所有可能类中特征重要性的总和，除以类的数量",
|
||||
"predictedClassInfo": "预测类: 显示给定点的预测类的特征重要性值",
|
||||
"enumeratedClassInfo": "枚举的类名: 仅显示所有数据点中指定的类的特征重要性值。",
|
||||
"close": "关闭",
|
||||
"crossClassWeights": "跨类权重"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "缩放的特征值",
|
||||
"low": "低",
|
||||
"high": "高",
|
||||
"featureLabel": "特征: {0}",
|
||||
"valueLabel": "特征值: {0}",
|
||||
"importanceLabel": "重要性: {0}",
|
||||
"predictedClassTooltip": "预测类: {0}",
|
||||
"trueClassTooltip": "真实类: {0}",
|
||||
"predictedOutputTooltip": "预测输出: {0}",
|
||||
"trueOutputTooltip": "真实输出: {0}",
|
||||
"topKFeatures": "前 K 个特征:",
|
||||
"topKInfo": "前 k 个重要特征的计算方式",
|
||||
"predictedValue": "预测值",
|
||||
"predictedClass": "预测类",
|
||||
"trueValue": "True 值",
|
||||
"trueClass": "真实类",
|
||||
"noColor": "无",
|
||||
"tooManyRows": "提供的数据集大于此图表可支持的数据集"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "类: {0}",
|
||||
"sortBy": "排序依据",
|
||||
"noData": "无数据",
|
||||
"absoluteGlobal": "绝对全局",
|
||||
"absoluteLocal": "绝对本地",
|
||||
"calculatingExplanation": "正在计算解释"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "必须为数值",
|
||||
"integerError": "必须为整数",
|
||||
"prediction": "预测",
|
||||
"predictedProbability": "预测概率",
|
||||
"predictionLabel": "预测: {0}",
|
||||
"probabilityLabel": "概率: {0}",
|
||||
"noModelError": "请提供 operationalized 模型以在 ICE 绘图中浏览预测。",
|
||||
"featurePickerLabel": "功能:",
|
||||
"minimumInputLabel": "最小值:",
|
||||
"maximumInputLabel": "最大值:",
|
||||
"stepInputLabel": "步骤:",
|
||||
"loadingMessage": "正在加载数据...",
|
||||
"submitPrompt": "提交一个范围以查看 ICE 绘图",
|
||||
"topLevelErrorMessage": "参数错误",
|
||||
"errorPrefix": "出现错误: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "正在加载...",
|
||||
"perturbationLabel": "小改动:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "预测值: {0}",
|
||||
"predictedClassLabel": "预测类: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "不进行分组",
|
||||
"groupPredicted": "预测 Y",
|
||||
"groupTrue": "真实 Y",
|
||||
"groupBy": "分组依据"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "图表类型:",
|
||||
"violinText": "小提琴",
|
||||
"barText": "条形图",
|
||||
"boxText": "框",
|
||||
"beehiveText": "Swarm",
|
||||
"globalImportanceExplanation": "全局特征重要性是通过对所有点的特征重要性的绝对值求平均值(L1 规范化)计算出来的。",
|
||||
"multiclassImportanceAddendum": "计算所有类的特征重要性时，所有点都包括在内，不使用任何差异权重。因此，对许多预测为非“A 类”的点具有较大负面重要性的特征将大大提高该特征的“A 类”重要性。"
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "等于",
|
||||
"greaterThanComparison": "大于",
|
||||
"greaterThanEqualToComparison": "大于或等于",
|
||||
"lessThanComparison": "小于",
|
||||
"lessThanEqualToComparison": "小于或等于",
|
||||
"inTheRangeOf": "在以下范围内:",
|
||||
"categoricalIncludeValues": "包含的值:",
|
||||
"numericValue": "值",
|
||||
"numericalComparison": "运算",
|
||||
"minimum": "最小",
|
||||
"maximum": "最大",
|
||||
"min": "最小值: {0}",
|
||||
"max": "最大值: {0}",
|
||||
"uniqueValues": "唯一值的数目: {0} 个"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "回归错误",
|
||||
"error": "错误",
|
||||
"classificationOutcome": "分类结果",
|
||||
"truePositive": "真正",
|
||||
"trueNegative": "真负",
|
||||
"falsePositive": "假正",
|
||||
"falseNegative": "假负",
|
||||
"dataset": "数据集",
|
||||
"predictedProbabilities": "预测概率",
|
||||
"none": "计数"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "关闭",
|
||||
"defaultCustomRootName": "第 {0} 行的副本",
|
||||
"filterFeaturePlaceholder": "搜索特征"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "队列",
|
||||
"defaultLabel": "所有数据"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "了解影响整体模型预测的前 k 个重要特征(亦称为“全局解释”)。使用滑块可以按降序显示特征重要性值。最多选择三个队列，以并排查看它们的特征重要性值。单击图中的任意特征条，以查看所选特征的值如何影响模型预测。",
|
||||
"topAtoB": "前 {0}-{1} 个特征",
|
||||
"datasetCohorts": "数据集队列",
|
||||
"legendHelpText": "通过单击图例项，在绘图中切换启用/禁用队列。",
|
||||
"sortBy": "排序方式",
|
||||
"viewDependencePlotFor": "查看以下特征的相关性绘图:",
|
||||
"datasetCohortSelector": "选择数据集队列",
|
||||
"aggregateFeatureImportance": "聚合特征重要性",
|
||||
"missingParameters": "此选项卡要求提供本地特征重要性参数。",
|
||||
"weightOptions": "类重要性权重",
|
||||
"dependencePlotTitle": "依赖关系图",
|
||||
"dependencePlotHelperText": "此依赖关系图显示特征的值与特征在整个队列中的相应重要性之间的关系。",
|
||||
"dependencePlotFeatureSelectPlaceholder": "选择特征",
|
||||
"datasetRequired": "依赖关系图需要计算数据集和本地特征重要性数组。"
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "数据统计信息",
|
||||
"datapoints": "{0} 个数据点",
|
||||
"features": "{0} 个特征",
|
||||
"filters": "{0} 个筛选器",
|
||||
"binaryClassifier": "两类分类器",
|
||||
"regressor": "回归量",
|
||||
"multiclassClassifier": "多类分类器",
|
||||
"datasetCohorts": "数据集队列",
|
||||
"editCohort": "编辑队列",
|
||||
"duplicateCohort": "复制队列",
|
||||
"addCohort": "添加队列",
|
||||
"copy": "副本"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "通过研究预测值和模型性能指标值的分布情况来评估模型的性能。通过查看数据集的不同队列或子组之间的性能比较分析，可以进一步调查模型。沿着 y 值和 x 值选择筛选器可以跨越不同的维度。使用图中的齿轮图标可以更改图类型。",
|
||||
"modelStatistics": "模型统计信息",
|
||||
"cohortPickerLabel": "选择要浏览的数据集队列",
|
||||
"missingParameters": "此选项卡要求提供来自模型的预测值数组。",
|
||||
"missingTrueY": "除了预测结果之外，模型性能统计信息还要求提供 true 结果"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y 值",
|
||||
"numberOfDatapoints": "数据点数",
|
||||
"xValue": "X 值",
|
||||
"rowIndex": "行索引",
|
||||
"featureImportance": "特征重要性",
|
||||
"countTooltipPrefix": "计数: {0}",
|
||||
"count": "计数",
|
||||
"featurePrefix": "特征",
|
||||
"importancePrefix": "重要性",
|
||||
"cohort": "队列",
|
||||
"howToRead": "如何阅读此图表"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "通过沿着 X 轴、Y 轴和颜色轴选择不同的筛选器来沿着不同的维度对数据进行切片，浏览数据集统计信息。创建上述数据集队列，以使用筛选器(如预测结果、数据集特征和误差组)分析数据集统计信息。使用图表右上角的齿轮图标可以更改图表类型。",
|
||||
"colorValue": "颜色值",
|
||||
"individualDatapoints": "单个数据点",
|
||||
"aggregatePlots": "聚合绘图",
|
||||
"chartType": "图表类型",
|
||||
"missingParameters": "此选项卡要求提供评估数据集。",
|
||||
"noColor": "无"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "特性重要性为",
|
||||
"placeholder": "单击上面条形图中的特征可以显示它的相关性绘图"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "可以通过单击散点图来选择一个数据点，以在下面查看它的局部特征重要性值(局部解释)和个体条件期望(ICE)图。通过使用右侧的面板来扰乱已知数据点的特征，创建假设模拟数据点。特征重要性值基于许多近似，而不是预测的“原因”。如果因果推理没有严格的数学稳健性，不建议用户在现实生活中根据此工具制定决策。",
|
||||
"panelPlaceholder": "必须有模型，才能对新的数据点进行预测。",
|
||||
"cohortPickerLabel": "选择要浏览的数据集队列",
|
||||
"scatterLegendText": "通过单击图例项，在绘图中切换启用和禁用数据点。",
|
||||
"realPoint": "实际数据点",
|
||||
"noneSelectedYet": "尚未选择任何点",
|
||||
"whatIfDatapoints": "模拟数据点",
|
||||
"noneCreatedYet": "尚未创建任何点",
|
||||
"showLabel": "显示:",
|
||||
"featureImportancePlot": "特征重要性绘图",
|
||||
"icePlot": "个体条件期望(ICE)图",
|
||||
"featureImportanceLackingParameters": "请提供本地特征重要性，以了解每个特征对各个预测有何影响。",
|
||||
"featureImportanceGetStartedText": "选择一个点来查看特征重要性",
|
||||
"iceLackingParameters": "ICE 绘图需要运营模型才能对假设数据点进行预测。",
|
||||
"IceGetStartedText": "选择一个点或创建一个模拟点来查看 ICE 绘图",
|
||||
"whatIfDatapoint": "模拟数据点",
|
||||
"whatIfHelpText": "在绘图上选择一个点，或手动输入一个要打乱的已知数据点索引，并另存为新的模拟点。",
|
||||
"indexLabel": "要打乱的数据索引",
|
||||
"rowLabel": "第 {0} 行",
|
||||
"whatIfNameLabel": "模拟数据点名称",
|
||||
"featureValues": "特征值",
|
||||
"predictedClass": "预测类:",
|
||||
"predictedValue": "预测值:",
|
||||
"probability": "概率:",
|
||||
"trueClass": "真实类:",
|
||||
"trueValue": "真实值:",
|
||||
"trueValue.comment": "用于回归的实际标签的前缀",
|
||||
"newPredictedClass": "新的预测类:",
|
||||
"newPredictedValue": "新的预测值:",
|
||||
"newProbability": "新概率:",
|
||||
"saveAsNewPoint": "另存为新点",
|
||||
"saveChanges": "保存更改",
|
||||
"loading": "正在加载...",
|
||||
"classLabel": "类: {0}",
|
||||
"minLabel": "最小",
|
||||
"maxLabel": "最大",
|
||||
"stepsLabel": "步进",
|
||||
"disclaimer": "免责声明:这些是基于许多近似值的解释，而不是预测的“原因”。在无法从数学上进行严格、扎实的因果推理的情况下，不建议用户在现实生活中根据此工具制定决策。",
|
||||
"missingParameters": "此选项卡要求提供评估数据集。",
|
||||
"selectionLimit": "最多选择 3 个点",
|
||||
"classPickerLabel": "类",
|
||||
"tooltipTitleMany": "前 {0} 个预测类",
|
||||
"whatIfTooltipTitle": "模拟预测类",
|
||||
"tooltipTitleFew": "预测类",
|
||||
"probabilityLabel": "概率",
|
||||
"deltaLabel": "增量",
|
||||
"nonNumericValue": "值应为数值",
|
||||
"icePlotHelperText": "ICE 图演示了所选数据点的预测值是如何沿介于最小值和最大值之间的一系列特征值变化的。"
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "选择筛选器",
|
||||
"TreatAsCategorical": "视为类别",
|
||||
"addFilter": "添加筛选器",
|
||||
"addedFilters": "已添加的筛选器",
|
||||
"noAddedFilters": "尚未添加筛选器",
|
||||
"defaultFilterState": "请选择筛选器，以将参数添加到数据集队列。",
|
||||
"cohortNameLabel": "数据集队列名称",
|
||||
"cohortNamePlaceholder": "命名你的队列",
|
||||
"save": "保存",
|
||||
"delete": "删除",
|
||||
"cancel": "取消",
|
||||
"cohortNameError": "缺少队列名称",
|
||||
"placeholderName": "队列 {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "选择",
|
||||
"ditherLabel": "应抖动",
|
||||
"selectFilter": "选择轴数值",
|
||||
"selectFeature": "选择特征",
|
||||
"binLabel": "应用数据装箱",
|
||||
"TreatAsCategorical": "视为类别",
|
||||
"numOfBins": "箱数",
|
||||
"groupByCohort": "按队列分组",
|
||||
"selectClass": "选择类",
|
||||
"countHelperText": "点数直方图"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "预测概率",
|
||||
"predictedY": "预测 Y",
|
||||
"evalData": "评估数据集",
|
||||
"localFeatureImportance": "本地特征重要性",
|
||||
"inconsistentDimensions": "尺寸不一致。{0} 有尺寸 {1}，而应为 {2}",
|
||||
"notNonEmpty": "{0} 输入不是非空数组",
|
||||
"varyingLength": "尺寸不一致。{0} 有长度不同的元素",
|
||||
"notArray": "{0} 不是数组。应为尺寸 {1} 的数组",
|
||||
"errorHeader": "有些输入参数不一致，将不会被使用:",
|
||||
"datasizeWarning": "计算数据集过大，无法有效地显示在某些图表中，请添加筛选器以减小队列的大小。",
|
||||
"datasizeError": "所选队列过大，请添加筛选器以减小队列的大小。",
|
||||
"addFilters": "添加筛选器"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": "包含 {0}",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} 和另外 {1} 人"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "均方误差: {0}",
|
||||
"rSquared": "R 平方: {0}",
|
||||
"meanPrediction": "平均值预测 {0}",
|
||||
"accuracy": "准确度: {0}",
|
||||
"precision": "精准率: {0}",
|
||||
"recall": "召回率: {0}",
|
||||
"fpr": "假正率: {0}",
|
||||
"fnr": "假负率: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "了解可影响整体模型预测的前 k 个重要特征。使用滑块可按降序显示特征重要性。"
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "这些解释是什么意思?",
|
||||
"clickHere": "了解更多",
|
||||
"shapTitle": "Shapley 值",
|
||||
"shapDescription": "此解释器使用 SHAP，这是一种用于解释模型的博弈论方法，其中特征集的重要性是通过边缘化从模型中“隐藏”这些特征来衡量的。若要了解详细信息，请单击下面的链接。",
|
||||
"limeTitle": "LIME (局部可解释的与模型无关的解释)",
|
||||
"limeDescription": "此解释器使用 LIME，它对模型进行线性近似处理。我们通过执行以下操作来获取解释: 扰乱实例，获取模型预测，并将这些预测用作标签来学习局部保真的稀疏线性模型。此线性模型的权重用作“特征重要性”。若要了解详细信息，请单击下面的链接。",
|
||||
"mimicTitle": "模仿(全局代理解释)",
|
||||
"mimicDescription": "此解释器基于以下理念: 训练全局代理模型来模仿黑盒模型。全局代理模型是一种本质上可解释的模型，它被训练为尽可能准确地对任何黑盒模型的预测进行近似处理。特征重要性值是基础代理模型(LightGBM、线性回归、随机梯度下降或决策树)的基于模型的特征重要性值",
|
||||
"pfiTitle": "排列特征重要性(PFI)",
|
||||
"pfiDescription": "此解释器对整个数据集一次一个特征地随机选择数据，并计算相关性能指标的变化幅度(默认性能指标: F1 用于二元分类，含微平均的 F1 分数用于多类分类，平均绝对误差用于回归)。变化越大，相应特征就越重要。此解释器只能解释基础模型的整体行为，而不能解释单个预测。特征的特征重要性值代表了通过扰乱相应特定特征实现的模型性能增量。"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,342 +0,0 @@
|
|||
{
|
||||
"selectPoint": "選取點以查看其區域解釋",
|
||||
"defaultClassNames": "類別 {0}",
|
||||
"defaultFeatureNames": "特徵 {0}",
|
||||
"absoluteAverage": "絕對值的平均值",
|
||||
"predictedClass": "預測的類別",
|
||||
"datasetExplorer": "資料集總管",
|
||||
"dataExploration": "資料集探索",
|
||||
"aggregateFeatureImportance": "彙總特徵重要度",
|
||||
"globalImportance": "全域重要性",
|
||||
"explanationExploration": "解釋探索",
|
||||
"individualAndWhatIf": "個別特徵重要度與假設",
|
||||
"summaryImportance": "彙總重要性",
|
||||
"featureImportance": "特徵重要度",
|
||||
"featureImportanceOf": "{0} 的特徵重要度",
|
||||
"perturbationExploration": "更動探索",
|
||||
"localFeatureImportance": "區域特徵重要度",
|
||||
"ice": "ICE",
|
||||
"clearSelection": "清除選取",
|
||||
"feature": "功能:",
|
||||
"intercept": "攔截",
|
||||
"modelPerformance": "模型效能",
|
||||
"ExplanationScatter": {
|
||||
"dataLabel": "資料: {0}",
|
||||
"importanceLabel": "重要性: {0}",
|
||||
"predictedY": "預測的 Y",
|
||||
"index": "索引",
|
||||
"dataGroupLabel": "資料",
|
||||
"output": "輸出",
|
||||
"probabilityLabel": "機率: {0}",
|
||||
"trueY": "實際的 Y",
|
||||
"class": "類別: ",
|
||||
"xValue": "X 值:",
|
||||
"yValue": "Y 值:",
|
||||
"colorValue": "色彩:",
|
||||
"count": "計數"
|
||||
},
|
||||
"CrossClass": {
|
||||
"label": "跨類別加權:",
|
||||
"info": "跨類別計算的資訊",
|
||||
"overviewInfo": "多類別模型會為每個類別產生獨立的特徵重要度向量。每個類別的特徵重要度向量會呈現出哪些特徵較有可能或較不可能構成某個類別。您可以選取如何將每個類別特徵重要度向量的權重加總成單一值:",
|
||||
"absoluteValInfo": "絕對值的平均值: 顯示所有可能類別的特徵重要性總和除以類別數目",
|
||||
"predictedClassInfo": "預測的類別: 為指定點的預測類別顯示特徵重要度值",
|
||||
"enumeratedClassInfo": "列舉的類別名稱: 只顯示所有資料點中指定類別的特徵重要度值。",
|
||||
"close": "關閉",
|
||||
"crossClassWeights": "跨類別權數"
|
||||
},
|
||||
"AggregateImportance": {
|
||||
"scaledFeatureValue": "縮放的特徵值",
|
||||
"low": "低",
|
||||
"high": "高",
|
||||
"featureLabel": "特徵: {0}",
|
||||
"valueLabel": "特徵值: {0}",
|
||||
"importanceLabel": "重要性: {0}",
|
||||
"predictedClassTooltip": "預測的類別: {0}",
|
||||
"trueClassTooltip": "實際的類別: {0}",
|
||||
"predictedOutputTooltip": "預測的輸出: {0}",
|
||||
"trueOutputTooltip": "實際的輸出: {0}",
|
||||
"topKFeatures": "前 K 個特徵:",
|
||||
"topKInfo": "計算出前 k 項的方法",
|
||||
"predictedValue": "預測值",
|
||||
"predictedClass": "預測的類別",
|
||||
"trueValue": "True 值",
|
||||
"trueClass": "實際的類別",
|
||||
"noColor": "無",
|
||||
"tooManyRows": "提供的資料集超過這個圖表可支援的大小"
|
||||
},
|
||||
"BarChart": {
|
||||
"classLabel": "類別: {0}",
|
||||
"sortBy": "排序依據",
|
||||
"noData": "沒有資料",
|
||||
"absoluteGlobal": "絕對全域",
|
||||
"absoluteLocal": "絕對區域",
|
||||
"calculatingExplanation": "正在計算解釋"
|
||||
},
|
||||
"IcePlot": {
|
||||
"numericError": "必須為數值",
|
||||
"integerError": "必須為整數",
|
||||
"prediction": "預測",
|
||||
"predictedProbability": "預測的機率",
|
||||
"predictionLabel": "預測: {0}",
|
||||
"probabilityLabel": "機率: {0}",
|
||||
"noModelError": "請提供經操作化的模型，以在 ICE 繪圖中探索預測。",
|
||||
"featurePickerLabel": "功能:",
|
||||
"minimumInputLabel": "最小值:",
|
||||
"maximumInputLabel": "最大值:",
|
||||
"stepInputLabel": "步驟:",
|
||||
"loadingMessage": "正在載入資料...",
|
||||
"submitPrompt": "提交範圍以檢視 ICE 繪圖",
|
||||
"topLevelErrorMessage": "參數有誤",
|
||||
"errorPrefix": "發生的錯誤: {0}"
|
||||
},
|
||||
"PerturbationExploration": {
|
||||
"loadingMessage": "正在載入...",
|
||||
"perturbationLabel": "更動:"
|
||||
},
|
||||
"PredictionLabel": {
|
||||
"predictedValueLabel": "預測的值: {0}",
|
||||
"predictedClassLabel": "預測的類別: {0}"
|
||||
},
|
||||
"Violin": {
|
||||
"groupNone": "未分組",
|
||||
"groupPredicted": "預測的 Y",
|
||||
"groupTrue": "實際的 Y",
|
||||
"groupBy": "群組依據"
|
||||
},
|
||||
"FeatureImportanceWrapper": {
|
||||
"chartType": "圖表類型:",
|
||||
"violinText": "小提琴",
|
||||
"barText": "橫條圖",
|
||||
"boxText": "方塊",
|
||||
"beehiveText": "群集",
|
||||
"globalImportanceExplanation": "全域特徵重要度的計算方法，是先算出所有點之特徵重要性的絕對值，再計算該絕對值的平均值 (L1 正規化)。 ",
|
||||
"multiclassImportanceAddendum": "計算特徵對所有類別的重要性時，所有點皆會予以計入，而不會使用差異加權。因此，若特徵對於許多預測不屬於「A 類別」的點具有重大負面重要性，將會大幅增加該特徵的「A 類別」重要性。"
|
||||
},
|
||||
"Filters": {
|
||||
"equalComparison": "等於",
|
||||
"greaterThanComparison": "大於",
|
||||
"greaterThanEqualToComparison": "大於或等於",
|
||||
"lessThanComparison": "小於",
|
||||
"lessThanEqualToComparison": "小於或等於",
|
||||
"inTheRangeOf": "介於範圍",
|
||||
"categoricalIncludeValues": "包含的值:",
|
||||
"numericValue": "值",
|
||||
"numericalComparison": "作業",
|
||||
"minimum": "最小值",
|
||||
"maximum": "最大值",
|
||||
"min": "下限: {0}",
|
||||
"max": "上限: {0}",
|
||||
"uniqueValues": "# 個唯一值: {0}"
|
||||
},
|
||||
"Columns": {
|
||||
"regressionError": "迴歸錯誤",
|
||||
"error": "錯誤",
|
||||
"classificationOutcome": "分類結果",
|
||||
"truePositive": "確判為真",
|
||||
"trueNegative": "確判不為真",
|
||||
"falsePositive": "誤判為真",
|
||||
"falseNegative": "誤判不為真",
|
||||
"dataset": "資料集",
|
||||
"predictedProbabilities": "預測可能性",
|
||||
"none": "計數"
|
||||
},
|
||||
"WhatIf": {
|
||||
"closeAriaLabel": "關閉",
|
||||
"defaultCustomRootName": "資料列 {0} 的複本",
|
||||
"filterFeaturePlaceholder": "搜尋特徵"
|
||||
},
|
||||
"Cohort": {
|
||||
"cohort": "世代",
|
||||
"defaultLabel": "所有資料"
|
||||
},
|
||||
"GlobalTab": {
|
||||
"helperText": "探索影響整體模型預測的前 k 大重要特徵 (亦即全域解釋)。使用滑桿顯示遞減的特徵重要性度。您最多可以選取三個世代，以並排的方式查看其特徵重要度。按一下圖表中的任一特徵列，查看選取的特徵如何影響模型預測。",
|
||||
"topAtoB": "前 {0}-{1} 個特徵",
|
||||
"datasetCohorts": "資料集世代",
|
||||
"legendHelpText": "按一下圖例項目，即可在繪圖中將世代切換為開啟或關閉。",
|
||||
"sortBy": "排序依據",
|
||||
"viewDependencePlotFor": "檢視依存性繪圖:",
|
||||
"datasetCohortSelector": "選取資料集世代",
|
||||
"aggregateFeatureImportance": "彙總特徵重要度",
|
||||
"missingParameters": "此索引標籤需要提供本機特徵重要度參數。",
|
||||
"weightOptions": "類別重要性權數",
|
||||
"dependencePlotTitle": "依存性繪圖",
|
||||
"dependencePlotHelperText": "此依序性繪圖顯示跨世代的特徵值以及與特徵對應之重要性間的關聯性。",
|
||||
"dependencePlotFeatureSelectPlaceholder": "選取特徵",
|
||||
"datasetRequired": "依存性繪圖需要評估資料集與局部特徵重要度陣列。"
|
||||
},
|
||||
"CohortBanner": {
|
||||
"dataStatistics": "資料統計資訊",
|
||||
"datapoints": "{0} 個資料點",
|
||||
"features": "{0} 個特徵",
|
||||
"filters": "{0} 個篩選",
|
||||
"binaryClassifier": "二進位分類器",
|
||||
"regressor": "迴歸輸入變數",
|
||||
"multiclassClassifier": "多類別分類器",
|
||||
"datasetCohorts": "資料集世代",
|
||||
"editCohort": "編輯世代",
|
||||
"duplicateCohort": "複製世代",
|
||||
"addCohort": "新增世代",
|
||||
"copy": " 複本"
|
||||
},
|
||||
"ModelPerformance": {
|
||||
"helperText": "探索您的預測值分佈與模型效能計量的值，以評估您的模型效能。您可以查看資料集不同世代或子群組之間的效能比較分析，以進一步調查您的模型。選取 y 值及 x 值上的篩選，以穿越不同維度。選取圖表中的齒輪可以變更圖表類型。",
|
||||
"modelStatistics": "模型統計資料",
|
||||
"cohortPickerLabel": "選取要探索的資料集世代",
|
||||
"missingParameters": "此索引標籤需要提供來自模型的預測值陣列。",
|
||||
"missingTrueY": "除了預測結果以外，模型效能統計資料還需要提供真實結果"
|
||||
},
|
||||
"Charts": {
|
||||
"yValue": "Y 值",
|
||||
"numberOfDatapoints": "資料點數目",
|
||||
"xValue": "X 值",
|
||||
"rowIndex": "資料列索引",
|
||||
"featureImportance": "特徵重要度",
|
||||
"countTooltipPrefix": "計數: {0}",
|
||||
"count": "計數",
|
||||
"featurePrefix": "特徵",
|
||||
"importancePrefix": "重要度",
|
||||
"cohort": "世代",
|
||||
"howToRead": "如何閱讀此圖表"
|
||||
},
|
||||
"DatasetExplorer": {
|
||||
"helperText": "依 X、Y 與色彩軸選取不同的篩選並根據不同維度分割資料，以探索您的資料集統計資料。您可在上方建立資料集世代，以使用預測結果、資料集功能與錯誤群組等篩選，分析資料集統計資料。使用圖表右上角的齒輪圖示來變更圖表類型。",
|
||||
"colorValue": "色彩值",
|
||||
"individualDatapoints": "個別資料點",
|
||||
"aggregatePlots": "彙總繪圖",
|
||||
"chartType": "圖表類型",
|
||||
"missingParameters": "此索引標籤需要提供評估資料集。",
|
||||
"noColor": "無"
|
||||
},
|
||||
"DependencePlot": {
|
||||
"featureImportanceOf": "特徵重要度:",
|
||||
"placeholder": "按一下上方橫條圖上的特徵以顯示其依存性繪圖"
|
||||
},
|
||||
"WhatIfTab": {
|
||||
"helperText": "您可以按一下散佈圖選取資料點，以檢視其局部特徵重要度 (局部解釋) 與下方的個別條件期望 (ICE) 圖。使用右側的面板，建立假設的假設資料點，以擾動已知的資料點特徵。特徵重要度以許多近似值為基礎，而且不是預測的「原因」。若沒有因果推斷的嚴格數學加強性，不建議使用者根據這個工具做出真實生活的決策。",
|
||||
"panelPlaceholder": "需要模型才能對新的資料點進行預測。",
|
||||
"cohortPickerLabel": "選取要探索的資料集世代",
|
||||
"scatterLegendText": "按一下圖例項目，即可在繪圖中將資料點切換為開啟或關閉。",
|
||||
"realPoint": "實際資料點",
|
||||
"noneSelectedYet": "尚未選取任何項目",
|
||||
"whatIfDatapoints": "假設資料點",
|
||||
"noneCreatedYet": "尚未建立任何項目",
|
||||
"showLabel": "顯示:",
|
||||
"featureImportancePlot": "特徵重要度繪圖",
|
||||
"icePlot": "個別條件預測 (ICE) 平面圖",
|
||||
"featureImportanceLackingParameters": "提供本機特徵重要度，以了解各個特徵如何影響個別預測。",
|
||||
"featureImportanceGetStartedText": "選取一個點以檢視特徵重要度",
|
||||
"iceLackingParameters": "ICE 繪圖需要可操作的模型，才可為假設資料點做預測。",
|
||||
"IceGetStartedText": "選取一個點或建立一個假設點以檢視 ICE 繪圖",
|
||||
"whatIfDatapoint": "假設資料點",
|
||||
"whatIfHelpText": "在繪圖上選取一個點，或手動輸入已知的資料點索引以用於擾動，並另存為新的假設點。",
|
||||
"indexLabel": "要擾動的資料索引",
|
||||
"rowLabel": "資料列 {0}",
|
||||
"whatIfNameLabel": "假設資料點名稱",
|
||||
"featureValues": "特徵值",
|
||||
"predictedClass": "預測的類別: ",
|
||||
"predictedValue": "預測值: ",
|
||||
"probability": "可能性: ",
|
||||
"trueClass": "True 類別: ",
|
||||
"trueValue": "True 值: ",
|
||||
"trueValue.comment": "迴歸實際標籤的前置詞",
|
||||
"newPredictedClass": "新的預測類別: ",
|
||||
"newPredictedValue": "新的預測值: ",
|
||||
"newProbability": "新的可能性: ",
|
||||
"saveAsNewPoint": "另存為新點",
|
||||
"saveChanges": "儲存變更",
|
||||
"loading": "正在載入...",
|
||||
"classLabel": "類別: {0}",
|
||||
"minLabel": "最小值",
|
||||
"maxLabel": "最大值",
|
||||
"stepsLabel": "步驟",
|
||||
"disclaimer": "免責聲明: 這些是以許多近似值為依據的解釋，而不是預測的「原因」。若在因果推論上沒有嚴謹的數據穩定性，我們並不建議使用者根據這個工具制定實際決策。",
|
||||
"missingParameters": "此索引標籤需要提供評估資料集。",
|
||||
"selectionLimit": "最多選取 3 個點",
|
||||
"classPickerLabel": "類別",
|
||||
"tooltipTitleMany": "前 {0} 項預測類別",
|
||||
"whatIfTooltipTitle": "假設預測類別",
|
||||
"tooltipTitleFew": "預測的類別",
|
||||
"probabilityLabel": "可能性",
|
||||
"deltaLabel": "差異",
|
||||
"nonNumericValue": "值應為數值",
|
||||
"icePlotHelperText": "ICE 繪圖示範所選資料點的預測值，如何按照最小值與最大值之間的特徵值範圍發生變化。"
|
||||
},
|
||||
"CohortEditor": {
|
||||
"selectFilter": "選取篩選",
|
||||
"TreatAsCategorical": "視為類別",
|
||||
"addFilter": "新增篩選",
|
||||
"addedFilters": "新增的篩選",
|
||||
"noAddedFilters": "尚未新增任何篩選",
|
||||
"defaultFilterState": "選取篩選，以將參數新增到您的資料集世代。",
|
||||
"cohortNameLabel": "資料集世代名稱",
|
||||
"cohortNamePlaceholder": "為您的世代命名",
|
||||
"save": "儲存",
|
||||
"delete": "刪除",
|
||||
"cancel": "取消",
|
||||
"cohortNameError": "缺少世代名稱",
|
||||
"placeholderName": "世代 {0}"
|
||||
},
|
||||
"AxisConfigDialog": {
|
||||
"select": "選取",
|
||||
"ditherLabel": "應抖動",
|
||||
"selectFilter": "選取您的軸值",
|
||||
"selectFeature": "選取特徵",
|
||||
"binLabel": "對資料套用量化",
|
||||
"TreatAsCategorical": "視為類別",
|
||||
"numOfBins": "量化數目",
|
||||
"groupByCohort": "依世代分組",
|
||||
"selectClass": "選取類別",
|
||||
"countHelperText": "包含點數的長條圖"
|
||||
},
|
||||
"ValidationErrors": {
|
||||
"predictedProbability": "預測的可能性",
|
||||
"predictedY": "預測的 Y",
|
||||
"evalData": "評估資料集",
|
||||
"localFeatureImportance": "本機特徵重要度",
|
||||
"inconsistentDimensions": "維度不一致。{0} 有維度 {1}，必須是 {2}",
|
||||
"notNonEmpty": "{0} 輸入不是非空白陣列",
|
||||
"varyingLength": "維度不一致。{0} 具有不同長度的元素",
|
||||
"notArray": "{0} 不是陣列。必須是維度 {1} 的陣列",
|
||||
"errorHeader": "某些輸入參數不一致，將不予採用: ",
|
||||
"datasizeWarning": "評估資料集太大，無法有效地顯示在某些圖表中。請新增篩選縮小該世代。 ",
|
||||
"datasizeError": "選取的世代太大。請新增篩選縮小該世代。",
|
||||
"addFilters": "新增篩選"
|
||||
},
|
||||
"FilterOperations": {
|
||||
"equals": " = {0}",
|
||||
"lessThan": " < {0}",
|
||||
"greaterThan": " > {0}",
|
||||
"lessThanEquals": " <= {0}",
|
||||
"greaterThanEquals": " >= {0}",
|
||||
"includes": " 包括 {0} ",
|
||||
"inTheRangeOf": "[ {0} ]",
|
||||
"overflowFilterArgs": "{0} 及 {1} 個其他"
|
||||
},
|
||||
"Statistics": {
|
||||
"mse": "MSE: {0}",
|
||||
"rSquared": "R 平方: {0}",
|
||||
"meanPrediction": "平均預測 {0}",
|
||||
"accuracy": "正確性: {0}",
|
||||
"precision": "精確度: {0}",
|
||||
"recall": "召回率: {0}",
|
||||
"fpr": "FPR: {0}",
|
||||
"fnr": "FNR: {0}"
|
||||
},
|
||||
"GlobalOnlyChart": {
|
||||
"helperText": "探索影響您整體模型預測的前 k 項重要特徵。使用滑桿顯示遞減的特徵重要度。"
|
||||
},
|
||||
"ExplanationSummary": {
|
||||
"whatDoExplanationsMean": "這些解釋的意思為何?",
|
||||
"clickHere": "深入了解",
|
||||
"shapTitle": "Shapley 值",
|
||||
"shapDescription": "此解釋器使用 SHAP，這是一種從遊戲理論來解釋模型的方法，透過邊緣化將模型中的那些特徵「隱藏」以測量特徵集的重要性。按一下下方連結以深入了解。",
|
||||
"limeTitle": "LIME (不侷限於特定模型，利用局部可解釋性的解釋方法)",
|
||||
"limeDescription": "此解釋器使用 LIME，這提供了模型的線性近似值。若要取得解釋，請執行下列動作: 擾動該執行個體、取得模型預測，並使用這些預測作為標籤，以了解局部忠實的稀疏線性模型。此線性模型的權數會作為「特徵重要度」。按一下下方連結以深入了解。",
|
||||
"mimicTitle": "模仿 (全域代理解釋)",
|
||||
"mimicDescription": "此解釋器根據訓練全域代理模型的概念來模仿黑箱模型。全域代理模型為一種內部可解釋的模型，經過訓練能夠盡可能正確地近似任何黑箱模型的預測。特徵重要度是基礎代理模型 (LightGBM、線性迴歸或隨機梯度下降法或決策樹) 的模型特徵重要度",
|
||||
"pfiTitle": "排列特徵重要度 (PFI)",
|
||||
"pfiDescription": "此解釋器針對整個資料集的資料，隨機地一次解釋一個特徵，並計算偏好變數的效能計量 (預設效能計量: F1 用於二進位分類，具有微平均值的 F1 分數用於多元分類，而平均絕對誤差用於迴歸)。變數越大，特徵就越重要。此解釋器只能解釋基礎模型的整體行為，而不會解釋個別的預測。透過擾動該特定特徵，特徵的特徵重要度代表模型效能中的差異。"
|
||||
}
|
||||
}
|
||||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { RangeTypes } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -20,7 +21,6 @@ import {
|
|||
import { Position } from "office-ui-fabric-react/lib/utilities/positioning";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { cohortKey } from "../../cohortKey";
|
||||
import { ColumnCategories, IJointMeta, JointDataset } from "../../JointDataset";
|
||||
import { ISelectorConfig } from "../../NewExplanationDashboard";
|
||||
|
|
@ -67,7 +67,10 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
this.props.orderedGroupTitles.includes(ColumnCategories.Dataset) &&
|
||||
this.props.jointDataset.hasDataset
|
||||
) {
|
||||
previousValue.push({ key, title: localization.Columns.dataset });
|
||||
previousValue.push({
|
||||
key,
|
||||
title: localization.Interpret.Columns.dataset
|
||||
});
|
||||
return previousValue;
|
||||
}
|
||||
if (
|
||||
|
|
@ -77,7 +80,7 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
) {
|
||||
previousValue.push({
|
||||
key,
|
||||
title: localization.Columns.predictedProbabilities
|
||||
title: localization.Interpret.Columns.predictedProbabilities
|
||||
});
|
||||
return previousValue;
|
||||
}
|
||||
|
|
@ -151,7 +154,7 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
<Stack tokens={{ childrenGap: "l1" }}>
|
||||
<Stack.Item>
|
||||
<ChoiceGroup
|
||||
label={localization.AxisConfigDialog.selectFilter}
|
||||
label={localization.Interpret.AxisConfigDialog.selectFilter}
|
||||
options={this.leftItems.map((i) => ({
|
||||
key: i.key,
|
||||
text: i.title
|
||||
|
|
@ -162,12 +165,16 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
</Stack.Item>
|
||||
{this.state.selectedColumn.property === cohortKey && (
|
||||
<Stack.Item>
|
||||
<Text>{localization.AxisConfigDialog.groupByCohort}</Text>
|
||||
<Text>
|
||||
{localization.Interpret.AxisConfigDialog.groupByCohort}
|
||||
</Text>
|
||||
</Stack.Item>
|
||||
)}
|
||||
{this.state.selectedColumn.property === ColumnCategories.None && (
|
||||
<Stack.Item>
|
||||
<Text>{localization.AxisConfigDialog.countHelperText}</Text>
|
||||
<Text>
|
||||
{localization.Interpret.AxisConfigDialog.countHelperText}
|
||||
</Text>
|
||||
</Stack.Item>
|
||||
)}
|
||||
{this.state.selectedColumn.property !== cohortKey &&
|
||||
|
|
@ -177,7 +184,9 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
<ComboBox
|
||||
options={this.dataArray}
|
||||
onChange={this.setSelectedProperty}
|
||||
label={localization.AxisConfigDialog.selectFeature}
|
||||
label={
|
||||
localization.Interpret.AxisConfigDialog.selectFeature
|
||||
}
|
||||
selectedKey={this.state.selectedColumn.property}
|
||||
/>
|
||||
)}
|
||||
|
|
@ -185,7 +194,7 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
<ComboBox
|
||||
options={this.classArray}
|
||||
onChange={this.setSelectedProperty}
|
||||
label={localization.AxisConfigDialog.selectClass}
|
||||
label={localization.Interpret.AxisConfigDialog.selectClass}
|
||||
selectedKey={this.state.selectedColumn.property}
|
||||
/>
|
||||
)}
|
||||
|
|
@ -193,7 +202,9 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
RangeTypes.Integer && (
|
||||
<Checkbox
|
||||
key={this.state.selectedColumn.property}
|
||||
label={localization.AxisConfigDialog.TreatAsCategorical}
|
||||
label={
|
||||
localization.Interpret.AxisConfigDialog.TreatAsCategorical
|
||||
}
|
||||
checked={selectedMeta.treatAsCategorical}
|
||||
onChange={this.setAsCategorical}
|
||||
/>
|
||||
|
|
@ -202,14 +213,16 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
<>
|
||||
<Text variant={"small"}>
|
||||
{`${localization.formatString(
|
||||
localization.Filters.uniqueValues,
|
||||
localization.Interpret.Filters.uniqueValues,
|
||||
selectedMeta.sortedCategoricalValues?.length
|
||||
)}`}
|
||||
</Text>
|
||||
{this.props.canDither && (
|
||||
<Checkbox
|
||||
key={this.state.selectedColumn.property}
|
||||
label={localization.AxisConfigDialog.ditherLabel}
|
||||
label={
|
||||
localization.Interpret.AxisConfigDialog.ditherLabel
|
||||
}
|
||||
checked={this.state.selectedColumn.options.dither}
|
||||
onChange={this.ditherChecked}
|
||||
/>
|
||||
|
|
@ -219,20 +232,20 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
<>
|
||||
<Text variant={"small"} nowrap block>
|
||||
{localization.formatString(
|
||||
localization.Filters.min,
|
||||
localization.Interpret.Filters.min,
|
||||
minVal
|
||||
)}
|
||||
</Text>
|
||||
<Text variant={"small"} nowrap block>
|
||||
{localization.formatString(
|
||||
localization.Filters.max,
|
||||
localization.Interpret.Filters.max,
|
||||
maxVal
|
||||
)}
|
||||
</Text>
|
||||
{this.props.canBin && !this.props.mustBin && (
|
||||
<Checkbox
|
||||
key={this.state.selectedColumn.property}
|
||||
label={localization.AxisConfigDialog.binLabel}
|
||||
label={localization.Interpret.AxisConfigDialog.binLabel}
|
||||
checked={this.state.selectedColumn.options.bin}
|
||||
onChange={this.shouldBinClicked}
|
||||
/>
|
||||
|
|
@ -242,7 +255,9 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
this.state.binCount !== undefined && (
|
||||
<SpinButton
|
||||
labelPosition={Position.top}
|
||||
label={localization.AxisConfigDialog.numOfBins}
|
||||
label={
|
||||
localization.Interpret.AxisConfigDialog.numOfBins
|
||||
}
|
||||
min={AxisConfigDialog.MIN_HIST_COLS}
|
||||
max={AxisConfigDialog.MAX_HIST_COLS}
|
||||
value={this.state.binCount.toString()}
|
||||
|
|
@ -270,7 +285,9 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
this.props.canDither && (
|
||||
<Checkbox
|
||||
key={this.state.selectedColumn.property}
|
||||
label={localization.AxisConfigDialog.ditherLabel}
|
||||
label={
|
||||
localization.Interpret.AxisConfigDialog.ditherLabel
|
||||
}
|
||||
checked={this.state.selectedColumn.options.dither}
|
||||
onChange={this.ditherChecked}
|
||||
/>
|
||||
|
|
@ -288,10 +305,10 @@ export class AxisConfigDialog extends React.PureComponent<
|
|||
return (
|
||||
<Stack horizontal tokens={{ childrenGap: "l1", padding: "l1" }}>
|
||||
<PrimaryButton onClick={this.saveState}>
|
||||
{localization.AxisConfigDialog.select}
|
||||
{localization.Interpret.AxisConfigDialog.select}
|
||||
</PrimaryButton>
|
||||
<DefaultButton onClick={this.props.onCancel}>
|
||||
{localization.CohortEditor.cancel}
|
||||
{localization.Interpret.CohortEditor.cancel}
|
||||
</DefaultButton>
|
||||
</Stack>
|
||||
);
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
Callout as FabricCallout,
|
||||
CommandBarButton,
|
||||
|
|
@ -10,7 +11,6 @@ import {
|
|||
import React from "react";
|
||||
import { v4 } from "uuid";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
|
||||
import { labelWithCalloutStyles } from "./LabelWithCallout.styles";
|
||||
|
|
@ -55,7 +55,7 @@ export class LabelWithCallout extends React.Component<
|
|||
<IconButton
|
||||
id={"cross-class-weight-info"}
|
||||
iconProps={{ iconName: "Info" }}
|
||||
title={localization.calloutTitle}
|
||||
title={localization.Interpret.calloutTitle}
|
||||
onClick={this.toggleCallout}
|
||||
/>
|
||||
</>
|
||||
|
|
|
|||
|
|
@ -1,9 +1,9 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import { IExplanationModelMetadata } from "../../IExplanationContext";
|
||||
import { JointDataset } from "../../JointDataset";
|
||||
|
|
@ -38,7 +38,7 @@ export class CohortBar extends React.Component<
|
|||
if (this.state.editingCohortIndex === this.props.cohorts.length) {
|
||||
cohortForEdit = {
|
||||
cohortName: localization.formatString(
|
||||
localization.CohortEditor.placeholderName,
|
||||
localization.Interpret.CohortEditor.placeholderName,
|
||||
this.state.editingCohortIndex
|
||||
),
|
||||
filterList: []
|
||||
|
|
@ -101,7 +101,7 @@ export class CohortBar extends React.Component<
|
|||
const cohorts = [...this.props.cohorts];
|
||||
cohorts.push(
|
||||
new Cohort(
|
||||
source.name + localization.CohortBanner.copy,
|
||||
source.name + localization.Interpret.CohortBanner.copy,
|
||||
this.props.jointDataset,
|
||||
[...source.filters]
|
||||
)
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { RangeTypes } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -17,7 +18,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React, { FormEvent } from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import { FilterMethods, IFilter } from "../../Interfaces/IFilter";
|
||||
import { IJointMeta, JointDataset } from "../../JointDataset";
|
||||
|
|
@ -96,8 +96,10 @@ export class CohortEditor extends React.PureComponent<
|
|||
<Stack.Item>
|
||||
<TextField
|
||||
value={this.state.cohortName}
|
||||
label={localization.CohortEditor.cohortNameLabel}
|
||||
placeholder={localization.CohortEditor.cohortNamePlaceholder}
|
||||
label={localization.Interpret.CohortEditor.cohortNameLabel}
|
||||
placeholder={
|
||||
localization.Interpret.CohortEditor.cohortNamePlaceholder
|
||||
}
|
||||
onGetErrorMessage={this._getErrorMessage}
|
||||
validateOnLoad={false}
|
||||
onChange={this.setCohortName}
|
||||
|
|
@ -106,7 +108,7 @@ export class CohortEditor extends React.PureComponent<
|
|||
<Stack.Item>
|
||||
<ChoiceGroup
|
||||
options={this.leftItems}
|
||||
label={localization.CohortEditor.selectFilter}
|
||||
label={localization.Interpret.CohortEditor.selectFilter}
|
||||
onChange={this.onFilterCategoryChange}
|
||||
selectedKey={this.state.selectedFilterCategory}
|
||||
/>
|
||||
|
|
@ -114,7 +116,7 @@ export class CohortEditor extends React.PureComponent<
|
|||
<Stack.Item>
|
||||
{!openedFilter ? (
|
||||
<Text variant={"medium"}>
|
||||
{localization.CohortEditor.defaultFilterState}
|
||||
{localization.Interpret.CohortEditor.defaultFilterState}
|
||||
</Text>
|
||||
) : (
|
||||
<CohortEditorFilter
|
||||
|
|
@ -155,14 +157,14 @@ export class CohortEditor extends React.PureComponent<
|
|||
onClick={this.deleteCohort}
|
||||
className={styles.deleteCohort}
|
||||
>
|
||||
{localization.CohortEditor.delete}
|
||||
{localization.Interpret.CohortEditor.delete}
|
||||
</DefaultButton>
|
||||
)}
|
||||
<PrimaryButton onClick={this.saveCohort}>
|
||||
{localization.CohortEditor.save}
|
||||
{localization.Interpret.CohortEditor.save}
|
||||
</PrimaryButton>
|
||||
<DefaultButton onClick={this.props.onCancel}>
|
||||
{localization.CohortEditor.cancel}
|
||||
{localization.Interpret.CohortEditor.cancel}
|
||||
</DefaultButton>
|
||||
</Stack>
|
||||
);
|
||||
|
|
@ -205,7 +207,7 @@ export class CohortEditor extends React.PureComponent<
|
|||
|
||||
private _getErrorMessage = (): string | undefined => {
|
||||
if (this.state.cohortName.length <= 0) {
|
||||
return localization.CohortEditor.cohortNameError;
|
||||
return localization.Interpret.CohortEditor.cohortNameError;
|
||||
}
|
||||
return undefined;
|
||||
};
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { RangeTypes, roundDecimal } from "@responsible-ai/mlchartlib";
|
||||
import {
|
||||
Checkbox,
|
||||
|
|
@ -16,7 +17,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import { FilterMethods, IFilter } from "../../Interfaces/IFilter";
|
||||
import { IJointMeta, JointDataset } from "../../JointDataset";
|
||||
|
|
@ -62,27 +62,27 @@ export class CohortEditorFilter extends React.Component<
|
|||
private comparisonOptions: IComboBoxOption[] = [
|
||||
{
|
||||
key: FilterMethods.Equal,
|
||||
text: localization.Filters.equalComparison
|
||||
text: localization.Interpret.Filters.equalComparison
|
||||
},
|
||||
{
|
||||
key: FilterMethods.GreaterThan,
|
||||
text: localization.Filters.greaterThanComparison
|
||||
text: localization.Interpret.Filters.greaterThanComparison
|
||||
},
|
||||
{
|
||||
key: FilterMethods.GreaterThanEqualTo,
|
||||
text: localization.Filters.greaterThanEqualToComparison
|
||||
text: localization.Interpret.Filters.greaterThanEqualToComparison
|
||||
},
|
||||
{
|
||||
key: FilterMethods.LessThan,
|
||||
text: localization.Filters.lessThanComparison
|
||||
text: localization.Interpret.Filters.lessThanComparison
|
||||
},
|
||||
{
|
||||
key: FilterMethods.LessThanEqualTo,
|
||||
text: localization.Filters.lessThanEqualToComparison
|
||||
text: localization.Interpret.Filters.lessThanEqualToComparison
|
||||
},
|
||||
{
|
||||
key: FilterMethods.InTheRangeOf,
|
||||
text: localization.Filters.inTheRangeOf
|
||||
text: localization.Interpret.Filters.inTheRangeOf
|
||||
}
|
||||
];
|
||||
public render(): React.ReactNode {
|
||||
|
|
@ -123,7 +123,7 @@ export class CohortEditorFilter extends React.Component<
|
|||
styles={FabricStyles.limitedSizeMenuDropdown}
|
||||
options={this.dataArray}
|
||||
onChange={this.props.setSelectedProperty}
|
||||
label={localization.CohortEditor.selectFilter}
|
||||
label={localization.Interpret.CohortEditor.selectFilter}
|
||||
selectedKey={this.props.openedFilter.column}
|
||||
calloutProps={FabricStyles.calloutProps}
|
||||
/>
|
||||
|
|
@ -132,7 +132,7 @@ export class CohortEditorFilter extends React.Component<
|
|||
selectedMeta.featureRange.rangeType === RangeTypes.Integer && (
|
||||
<Checkbox
|
||||
key={this.props.openedFilter.column}
|
||||
label={localization.CohortEditor.TreatAsCategorical}
|
||||
label={localization.Interpret.CohortEditor.TreatAsCategorical}
|
||||
checked={selectedMeta.treatAsCategorical}
|
||||
onChange={this.props.setAsCategorical}
|
||||
/>
|
||||
|
|
@ -141,14 +141,14 @@ export class CohortEditorFilter extends React.Component<
|
|||
<>
|
||||
<Text variant={"small"}>
|
||||
{`${localization.formatString(
|
||||
localization.Filters.uniqueValues,
|
||||
localization.Interpret.Filters.uniqueValues,
|
||||
selectedMeta.sortedCategoricalValues?.length
|
||||
)}`}
|
||||
</Text>
|
||||
<ComboBox
|
||||
key={this.props.openedFilter.column}
|
||||
multiSelect
|
||||
label={localization.Filters.categoricalIncludeValues}
|
||||
label={localization.Interpret.Filters.categoricalIncludeValues}
|
||||
selectedKey={this.props.openedFilter.arg}
|
||||
onChange={this.props.setCategoricalValues}
|
||||
options={categoricalOptions}
|
||||
|
|
@ -161,15 +161,15 @@ export class CohortEditorFilter extends React.Component<
|
|||
<>
|
||||
<Text block nowrap variant={"small"}>
|
||||
{`${localization.formatString(
|
||||
localization.Filters.min,
|
||||
localization.Interpret.Filters.min,
|
||||
minVal
|
||||
)} ${localization.formatString(
|
||||
localization.Filters.max,
|
||||
localization.Interpret.Filters.max,
|
||||
maxVal
|
||||
)}`}
|
||||
</Text>
|
||||
<ComboBox
|
||||
label={localization.Filters.numericalComparison}
|
||||
label={localization.Interpret.Filters.numericalComparison}
|
||||
selectedKey={this.props.openedFilter.method}
|
||||
onChange={this.props.setComparison}
|
||||
options={this.comparisonOptions}
|
||||
|
|
@ -182,7 +182,7 @@ export class CohortEditorFilter extends React.Component<
|
|||
<SpinButton
|
||||
labelPosition={Position.top}
|
||||
value={this.props.openedFilter.arg[0].toString()}
|
||||
label={localization.Filters.minimum}
|
||||
label={localization.Interpret.Filters.minimum}
|
||||
min={selectedMeta.featureRange.min}
|
||||
max={selectedMeta.featureRange.max}
|
||||
onIncrement={(value): void => {
|
||||
|
|
@ -208,7 +208,7 @@ export class CohortEditorFilter extends React.Component<
|
|||
<SpinButton
|
||||
labelPosition={Position.top}
|
||||
value={this.props.openedFilter.arg[1].toString()}
|
||||
label={localization.Filters.maximum}
|
||||
label={localization.Interpret.Filters.maximum}
|
||||
min={selectedMeta.featureRange.min}
|
||||
max={selectedMeta.featureRange.max}
|
||||
onIncrement={(value): void => {
|
||||
|
|
@ -235,7 +235,7 @@ export class CohortEditorFilter extends React.Component<
|
|||
) : (
|
||||
<SpinButton
|
||||
labelPosition={Position.top}
|
||||
label={localization.Filters.numericValue}
|
||||
label={localization.Interpret.Filters.numericValue}
|
||||
min={selectedMeta.featureRange.min}
|
||||
max={selectedMeta.featureRange.max}
|
||||
value={this.props.openedFilter.arg[0].toString()}
|
||||
|
|
@ -267,7 +267,7 @@ export class CohortEditorFilter extends React.Component<
|
|||
<>
|
||||
<Stack.Item>
|
||||
<PrimaryButton
|
||||
text={localization.CohortEditor.save}
|
||||
text={localization.Interpret.CohortEditor.save}
|
||||
onClick={(): void =>
|
||||
this.props.saveState(this.props.filterIndex)
|
||||
}
|
||||
|
|
@ -275,7 +275,7 @@ export class CohortEditorFilter extends React.Component<
|
|||
</Stack.Item>
|
||||
<Stack.Item>
|
||||
<DefaultButton
|
||||
text={localization.CohortEditor.cancel}
|
||||
text={localization.Interpret.CohortEditor.cancel}
|
||||
onClick={(): void => this.props.cancelFilter()}
|
||||
/>
|
||||
</Stack.Item>
|
||||
|
|
@ -283,7 +283,7 @@ export class CohortEditorFilter extends React.Component<
|
|||
) : (
|
||||
<Stack.Item>
|
||||
<PrimaryButton
|
||||
text={localization.CohortEditor.addFilter}
|
||||
text={localization.Interpret.CohortEditor.addFilter}
|
||||
onClick={(): void =>
|
||||
this.props.saveState(this.props.filters.length)
|
||||
}
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { roundDecimal } from "@responsible-ai/mlchartlib";
|
||||
import {
|
||||
IconButton,
|
||||
|
|
@ -11,7 +12,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FilterMethods, IFilter } from "../../Interfaces/IFilter";
|
||||
import { JointDataset } from "../../JointDataset";
|
||||
|
||||
|
|
@ -26,20 +26,22 @@ export class CohortEditorFilterList extends React.Component<
|
|||
ICohortEditorFilterList
|
||||
> {
|
||||
private filterMethodLabels: { [key in FilterMethods]: string } = {
|
||||
[FilterMethods.Equal]: localization.FilterOperations.equals,
|
||||
[FilterMethods.GreaterThan]: localization.FilterOperations.greaterThan,
|
||||
[FilterMethods.Equal]: localization.Interpret.FilterOperations.equals,
|
||||
[FilterMethods.GreaterThan]:
|
||||
localization.Interpret.FilterOperations.greaterThan,
|
||||
[FilterMethods.GreaterThanEqualTo]:
|
||||
localization.FilterOperations.greaterThanEquals,
|
||||
[FilterMethods.LessThan]: localization.FilterOperations.lessThan,
|
||||
localization.Interpret.FilterOperations.greaterThanEquals,
|
||||
[FilterMethods.LessThan]: localization.Interpret.FilterOperations.lessThan,
|
||||
[FilterMethods.LessThanEqualTo]:
|
||||
localization.FilterOperations.lessThanEquals,
|
||||
[FilterMethods.Includes]: localization.FilterOperations.includes,
|
||||
[FilterMethods.InTheRangeOf]: localization.FilterOperations.inTheRangeOf
|
||||
localization.Interpret.FilterOperations.lessThanEquals,
|
||||
[FilterMethods.Includes]: localization.Interpret.FilterOperations.includes,
|
||||
[FilterMethods.InTheRangeOf]:
|
||||
localization.Interpret.FilterOperations.inTheRangeOf
|
||||
};
|
||||
public render(): React.ReactNode {
|
||||
return (
|
||||
<>
|
||||
<Label>{localization.CohortEditor.addedFilters}</Label>
|
||||
<Label>{localization.Interpret.CohortEditor.addedFilters}</Label>
|
||||
{this.props.filters.length > 0 ? (
|
||||
this.props.filters.map((filter, index) => {
|
||||
return (
|
||||
|
|
@ -59,7 +61,7 @@ export class CohortEditorFilterList extends React.Component<
|
|||
) : (
|
||||
<div>
|
||||
<Text variant={"smallPlus"}>
|
||||
{localization.CohortEditor.noAddedFilters}
|
||||
{localization.Interpret.CohortEditor.noAddedFilters}
|
||||
</Text>
|
||||
</div>
|
||||
)}
|
||||
|
|
@ -89,7 +91,7 @@ export class CohortEditorFilterList extends React.Component<
|
|||
const otherValues = selectedValues.slice(0, 3).toString();
|
||||
const countOtherValues = selectedValues.length - 3;
|
||||
stringArgs = localization.formatString(
|
||||
localization.FilterOperations.overflowFilterArgs,
|
||||
localization.Interpret.FilterOperations.overflowFilterArgs,
|
||||
otherValues,
|
||||
countOtherValues.toString()
|
||||
);
|
||||
|
|
@ -104,7 +106,7 @@ export class CohortEditorFilterList extends React.Component<
|
|||
if (filter.method === FilterMethods.InTheRangeOf) {
|
||||
// example: Age [30,40]
|
||||
label = `${selectedFilter.abbridgedLabel} ${localization.formatString(
|
||||
localization.FilterOperations.inTheRangeOf,
|
||||
localization.Interpret.FilterOperations.inTheRangeOf,
|
||||
stringArgs
|
||||
)}`;
|
||||
} else {
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
CommandBarButton,
|
||||
PrimaryButton,
|
||||
|
|
@ -9,7 +10,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import {
|
||||
IExplanationModelMetadata,
|
||||
|
|
@ -29,11 +29,11 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
public render(): React.ReactNode {
|
||||
let modelType: string;
|
||||
if (this.props.metadata.modelType === ModelTypes.Binary) {
|
||||
modelType = localization.CohortBanner.binaryClassifier;
|
||||
modelType = localization.Interpret.CohortBanner.binaryClassifier;
|
||||
} else if (this.props.metadata.modelType === ModelTypes.Multiclass) {
|
||||
modelType = localization.CohortBanner.multiclassClassifier;
|
||||
modelType = localization.Interpret.CohortBanner.multiclassClassifier;
|
||||
} else {
|
||||
modelType = localization.CohortBanner.regressor;
|
||||
modelType = localization.Interpret.CohortBanner.regressor;
|
||||
}
|
||||
return (
|
||||
<Stack tokens={{ childrenGap: "l1", padding: "l1" }}>
|
||||
|
|
@ -44,7 +44,7 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
this,
|
||||
this.props.cohorts.length
|
||||
)}
|
||||
text={localization.CohortBanner.addCohort}
|
||||
text={localization.Interpret.CohortBanner.addCohort}
|
||||
iconProps={{ iconName: "Add" }}
|
||||
styles={{ label: { whiteSpace: "nowrap" } }}
|
||||
/>
|
||||
|
|
@ -52,7 +52,7 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
<Stack>
|
||||
<Stack.Item>
|
||||
<Text variant={"small"}>
|
||||
{localization.CohortBanner.dataStatistics.toUpperCase()}
|
||||
{localization.Interpret.CohortBanner.dataStatistics.toUpperCase()}
|
||||
</Text>
|
||||
</Stack.Item>
|
||||
<Stack.Item>
|
||||
|
|
@ -61,7 +61,7 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
<Stack.Item>
|
||||
<Text variant={"xSmall"}>
|
||||
{localization.formatString(
|
||||
localization.CohortBanner.datapoints,
|
||||
localization.Interpret.CohortBanner.datapoints,
|
||||
this.props.jointDataset.datasetRowCount
|
||||
)}
|
||||
</Text>
|
||||
|
|
@ -69,7 +69,7 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
<Stack.Item>
|
||||
<Text variant={"xSmall"}>
|
||||
{localization.formatString(
|
||||
localization.CohortBanner.features,
|
||||
localization.Interpret.CohortBanner.features,
|
||||
this.props.jointDataset.datasetFeatureCount
|
||||
)}
|
||||
</Text>
|
||||
|
|
@ -78,7 +78,7 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
<Stack>
|
||||
<Stack.Item>
|
||||
<Text variant={"small"}>
|
||||
{localization.CohortBanner.datasetCohorts.toUpperCase()}
|
||||
{localization.Interpret.CohortBanner.datasetCohorts.toUpperCase()}
|
||||
</Text>
|
||||
</Stack.Item>
|
||||
{this.props.cohorts.map((cohort, index) => {
|
||||
|
|
@ -98,12 +98,15 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
items: [
|
||||
{
|
||||
key: "item4",
|
||||
name: localization.CohortBanner.editCohort,
|
||||
name:
|
||||
localization.Interpret.CohortBanner.editCohort,
|
||||
onClick: (): void => this.props.editCohort(index)
|
||||
},
|
||||
{
|
||||
key: "item5",
|
||||
name: localization.CohortBanner.duplicateCohort,
|
||||
name:
|
||||
localization.Interpret.CohortBanner
|
||||
.duplicateCohort,
|
||||
onClick: (): void => this.props.cloneAndEdit(index)
|
||||
}
|
||||
]
|
||||
|
|
@ -115,7 +118,7 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
<Stack.Item>
|
||||
<Text variant={"xSmall"}>
|
||||
{localization.formatString(
|
||||
localization.CohortBanner.datapoints,
|
||||
localization.Interpret.CohortBanner.datapoints,
|
||||
cohort.filteredData.length
|
||||
)}
|
||||
</Text>
|
||||
|
|
@ -123,7 +126,7 @@ export class CohortList extends React.PureComponent<ICohortListProps> {
|
|||
<Stack.Item>
|
||||
<Text variant={"xSmall"}>
|
||||
{localization.formatString(
|
||||
localization.CohortBanner.filters,
|
||||
localization.Interpret.CohortBanner.filters,
|
||||
cohort.filters.length
|
||||
)}
|
||||
</Text>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { AccessibleChart } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -14,7 +15,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { ChartTypes } from "../../ChartTypes";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import { cohortKey } from "../../cohortKey";
|
||||
|
|
@ -76,7 +76,7 @@ export class DatasetExplorerTab extends React.PureComponent<
|
|||
<div className={classNames.missingParametersPlaceholder}>
|
||||
<div className={classNames.missingParametersPlaceholderSpacer}>
|
||||
<Text variant="large" className={classNames.faintText}>
|
||||
{localization.DatasetExplorer.missingParameters}
|
||||
{localization.Interpret.DatasetExplorer.missingParameters}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -114,12 +114,12 @@ export class DatasetExplorerTab extends React.PureComponent<
|
|||
<div className={classNames.infoWithText}>
|
||||
<Icon iconName="Info" className={classNames.infoIcon} />
|
||||
<Text variant="medium" className={classNames.helperText}>
|
||||
{localization.DatasetExplorer.helperText}
|
||||
{localization.Interpret.DatasetExplorer.helperText}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={classNames.cohortPickerWrapper}>
|
||||
<Text variant="mediumPlus" className={classNames.cohortPickerLabel}>
|
||||
{localization.ModelPerformance.cohortPickerLabel}
|
||||
{localization.Interpret.ModelPerformance.cohortPickerLabel}
|
||||
</Text>
|
||||
{cohortOptions && (
|
||||
<Dropdown
|
||||
|
|
@ -218,7 +218,7 @@ export class DatasetExplorerTab extends React.PureComponent<
|
|||
variant="large"
|
||||
className={classNames.faintText}
|
||||
>
|
||||
{localization.ValidationErrors.datasizeError}
|
||||
{localization.Interpret.ValidationErrors.datasizeError}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
ChoiceGroup,
|
||||
|
|
@ -10,7 +11,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { ChartTypes } from "../../ChartTypes";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
|
|
@ -35,11 +35,11 @@ export class SidePanel extends React.Component<ISidePanelProps> {
|
|||
private readonly chartOptions: IChoiceGroupOption[] = [
|
||||
{
|
||||
key: ChartTypes.Histogram,
|
||||
text: localization.DatasetExplorer.aggregatePlots
|
||||
text: localization.Interpret.DatasetExplorer.aggregatePlots
|
||||
},
|
||||
{
|
||||
key: ChartTypes.Scatter,
|
||||
text: localization.DatasetExplorer.individualDatapoints
|
||||
text: localization.Interpret.DatasetExplorer.individualDatapoints
|
||||
}
|
||||
];
|
||||
public render(): React.ReactNode {
|
||||
|
|
@ -48,7 +48,7 @@ export class SidePanel extends React.Component<ISidePanelProps> {
|
|||
return (
|
||||
<div className={classNames.legendAndText}>
|
||||
<Text variant={"mediumPlus"} block className={classNames.boldText}>
|
||||
{localization.DatasetExplorer.colorValue}
|
||||
{localization.Interpret.DatasetExplorer.colorValue}
|
||||
</Text>
|
||||
{this.props.chartProps.chartType === ChartTypes.Scatter && (
|
||||
<DefaultButton
|
||||
|
|
@ -90,14 +90,14 @@ export class SidePanel extends React.Component<ISidePanelProps> {
|
|||
})
|
||||
) : (
|
||||
<Text variant={"xSmall"} className={classNames.smallItalic}>
|
||||
{localization.DatasetExplorer.noColor}
|
||||
{localization.Interpret.DatasetExplorer.noColor}
|
||||
</Text>
|
||||
)}
|
||||
</div>
|
||||
|
||||
<ChoiceGroup
|
||||
id="ChartTypeSelection"
|
||||
label={localization.DatasetExplorer.chartType}
|
||||
label={localization.Interpret.DatasetExplorer.chartType}
|
||||
selectedKey={this.props.chartProps.chartType}
|
||||
options={this.chartOptions}
|
||||
onChange={this.props.onChartTypeChange}
|
||||
|
|
|
|||
|
|
@ -1,7 +1,8 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
|
||||
import { ChartTypes } from "../../ChartTypes";
|
||||
import { IGenericChartProps } from "../../IGenericChartProps";
|
||||
import { ColumnCategories, JointDataset } from "../../JointDataset";
|
||||
|
|
@ -35,7 +36,8 @@ export function buildHoverTemplate(
|
|||
": %{customdata.Color}<br>";
|
||||
}
|
||||
hovertemplate +=
|
||||
localization.Charts.rowIndex + ": %{customdata.AbsoluteIndex}<br>";
|
||||
localization.Interpret.Charts.rowIndex +
|
||||
": %{customdata.AbsoluteIndex}<br>";
|
||||
break;
|
||||
}
|
||||
case ChartTypes.Histogram: {
|
||||
|
|
@ -47,7 +49,7 @@ export function buildHoverTemplate(
|
|||
hovertemplate += yName + ": %{customdata.Y}<br>";
|
||||
}
|
||||
hovertemplate += localization.formatString(
|
||||
localization.Charts.countTooltipPrefix,
|
||||
localization.Interpret.Charts.countTooltipPrefix,
|
||||
"%{y}<br>"
|
||||
);
|
||||
break;
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
AccessibleChart,
|
||||
IPlotlyProperty,
|
||||
|
|
@ -10,7 +11,6 @@ import _ from "lodash";
|
|||
import { getTheme, Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import {
|
||||
|
|
@ -84,7 +84,7 @@ export class DependencePlot extends React.PureComponent<IDependecePlotProps> {
|
|||
<div className={classNames.secondaryChartPlacolderBox}>
|
||||
<div className={classNames.secondaryChartPlacolderSpacer}>
|
||||
<Text variant="large" className={classNames.faintText}>
|
||||
{localization.DependencePlot.placeholder}
|
||||
{localization.Interpret.DependencePlot.placeholder}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -106,7 +106,7 @@ export class DependencePlot extends React.PureComponent<IDependecePlotProps> {
|
|||
<div className={classNames.verticalAxis}>
|
||||
<div className={classNames.rotatedVerticalBox}>
|
||||
<Text variant={"medium"} block>
|
||||
{localization.DependencePlot.featureImportanceOf}
|
||||
{localization.Interpret.DependencePlot.featureImportanceOf}
|
||||
</Text>
|
||||
<Text variant={"medium"}>{yAxisLabel}</Text>
|
||||
</div>
|
||||
|
|
@ -202,7 +202,7 @@ export class DependencePlot extends React.PureComponent<IDependecePlotProps> {
|
|||
_.set(plotlyProps, "layout.yaxis.tickvals", yLabelIndexes);
|
||||
}
|
||||
const rawY: number[] = cohort.unwrap(chartProps.yAxis.property);
|
||||
const yLabel = localization.Charts.featureImportance;
|
||||
const yLabel = localization.Interpret.Charts.featureImportance;
|
||||
plotlyProps.data[0].y = rawY;
|
||||
rawY.forEach((val, index) => {
|
||||
customdata[index]["Yformatted"] = val.toLocaleString(undefined, {
|
||||
|
|
@ -216,7 +216,8 @@ export class DependencePlot extends React.PureComponent<IDependecePlotProps> {
|
|||
customdata[i]["AbsoluteIndex"] = absoluteIndex;
|
||||
});
|
||||
hovertemplate +=
|
||||
localization.Charts.rowIndex + ": %{customdata.AbsoluteIndex}<br>";
|
||||
localization.Interpret.Charts.rowIndex +
|
||||
": %{customdata.AbsoluteIndex}<br>";
|
||||
hovertemplate += "<extra></extra>";
|
||||
plotlyProps.data[0].customdata = customdata as any;
|
||||
plotlyProps.data[0].hovertemplate = hovertemplate;
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
AccessibleChart,
|
||||
IPlotlyProperty,
|
||||
|
|
@ -12,7 +13,6 @@ import memoize from "memoize-one";
|
|||
import { ComboBox, IComboBox, IComboBoxOption } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { FabricStyles } from "../FabricStyles";
|
||||
import {
|
||||
IExplanationContext,
|
||||
|
|
@ -174,7 +174,7 @@ export class EbmExplanation extends React.PureComponent<IEbmProps, IEbmState> {
|
|||
ebmObject.displayParameters &&
|
||||
ebmObject.displayParameters.yAxisLabel
|
||||
? ebmObject.displayParameters.yAxisLabel
|
||||
: localization.IcePlot.predictedProbability
|
||||
: localization.Interpret.IcePlot.predictedProbability
|
||||
}
|
||||
} as any
|
||||
};
|
||||
|
|
@ -224,7 +224,7 @@ export class EbmExplanation extends React.PureComponent<IEbmProps, IEbmState> {
|
|||
<div>
|
||||
<div>
|
||||
<ComboBox
|
||||
label={localization.feature}
|
||||
label={localization.Interpret.feature}
|
||||
selectedKey={this.state.selectedFeature}
|
||||
onChange={this.onFeatureSelect}
|
||||
options={this.featureOptions}
|
||||
|
|
|
|||
|
|
@ -1,7 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
|
||||
export interface IExplainerInfo {
|
||||
title: string;
|
||||
|
|
@ -11,28 +11,28 @@ export interface IExplainerInfo {
|
|||
|
||||
// The list of known explanations is an enumerated list of python strings that interpret outputs
|
||||
const shapExplanation: IExplainerInfo = {
|
||||
description: localization.ExplanationSummary.shapDescription,
|
||||
description: localization.Interpret.ExplanationSummary.shapDescription,
|
||||
linkUrl: "https://github.com/slundberg/shap",
|
||||
title: localization.ExplanationSummary.shapTitle
|
||||
title: localization.Interpret.ExplanationSummary.shapTitle
|
||||
};
|
||||
|
||||
const limeExplanation: IExplainerInfo = {
|
||||
description: localization.ExplanationSummary.limeDescription,
|
||||
description: localization.Interpret.ExplanationSummary.limeDescription,
|
||||
linkUrl: "https://github.com/marcotcr/lime",
|
||||
title: localization.ExplanationSummary.limeTitle
|
||||
title: localization.Interpret.ExplanationSummary.limeTitle
|
||||
};
|
||||
|
||||
const mimicExplanation: IExplainerInfo = {
|
||||
description: localization.ExplanationSummary.mimicDescription,
|
||||
description: localization.Interpret.ExplanationSummary.mimicDescription,
|
||||
linkUrl: "https://christophm.github.io/interpretable-ml-book/global.html",
|
||||
title: localization.ExplanationSummary.mimicTitle
|
||||
title: localization.Interpret.ExplanationSummary.mimicTitle
|
||||
};
|
||||
|
||||
const pfiExplanation: IExplainerInfo = {
|
||||
description: localization.ExplanationSummary.pfiDescription,
|
||||
description: localization.Interpret.ExplanationSummary.pfiDescription,
|
||||
linkUrl:
|
||||
"https://christophm.github.io/interpretable-ml-book/feature-importance.html",
|
||||
title: localization.ExplanationSummary.pfiTitle
|
||||
title: localization.Interpret.ExplanationSummary.pfiTitle
|
||||
};
|
||||
|
||||
export const explainerCalloutDictionary: { [key: string]: IExplainerInfo } = {
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
ChartBuilder,
|
||||
AccessibleChart,
|
||||
|
|
@ -21,7 +22,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import { IExplanationContext, ModelTypes } from "../../IExplanationContext";
|
||||
import { ModelExplanationUtils } from "../../ModelExplanationUtils";
|
||||
|
|
@ -132,7 +132,7 @@ export class Beehive extends React.PureComponent<
|
|||
},
|
||||
yaxis: {
|
||||
automargin: true,
|
||||
title: localization.featureImportance
|
||||
title: localization.Interpret.featureImportance
|
||||
}
|
||||
}
|
||||
};
|
||||
|
|
@ -237,7 +237,7 @@ export class Beehive extends React.PureComponent<
|
|||
_.set(
|
||||
plotlyProps,
|
||||
"layout.yaxis.title",
|
||||
`${localization.featureImportance}<br> ${localization.ExplanationScatter.class} ${explanationContext.modelMetadata.classNames[0]}`
|
||||
`${localization.Interpret.featureImportance}<br> ${localization.Interpret.ExplanationScatter.class} ${explanationContext.modelMetadata.classNames[0]}`
|
||||
);
|
||||
}
|
||||
if (selectedOption === undefined || selectedOption.key === "none") {
|
||||
|
|
@ -256,8 +256,8 @@ export class Beehive extends React.PureComponent<
|
|||
];
|
||||
_.set(plotlyProps.data[0], "marker.colorbar.tickvals", [0, 1]);
|
||||
_.set(plotlyProps.data[0], "marker.colorbar.ticktext", [
|
||||
localization.AggregateImportance.low,
|
||||
localization.AggregateImportance.high
|
||||
localization.Interpret.AggregateImportance.low,
|
||||
localization.Interpret.AggregateImportance.high
|
||||
]);
|
||||
} else {
|
||||
_.set(plotlyProps.data[0], "marker.opacity", 0.6);
|
||||
|
|
@ -435,7 +435,8 @@ export class Beehive extends React.PureComponent<
|
|||
<NoDataMessage
|
||||
explanationStrings={[
|
||||
{
|
||||
displayText: localization.AggregateImportance.tooManyRows,
|
||||
displayText:
|
||||
localization.Interpret.AggregateImportance.tooManyRows,
|
||||
format: "text"
|
||||
}
|
||||
]}
|
||||
|
|
@ -463,7 +464,7 @@ export class Beehive extends React.PureComponent<
|
|||
<div className={beehiveStyles.aggregateChart}>
|
||||
<div className={beehiveStyles.topControls}>
|
||||
<ComboBox
|
||||
label={localization.FeatureImportanceWrapper.chartType}
|
||||
label={localization.Interpret.FeatureImportanceWrapper.chartType}
|
||||
className={beehiveStyles.pathSelector}
|
||||
selectedKey={FeatureImportanceModes.Beehive}
|
||||
onChange={this.setChart}
|
||||
|
|
@ -474,7 +475,7 @@ export class Beehive extends React.PureComponent<
|
|||
/>
|
||||
{this.colorOptions.length > 1 && (
|
||||
<ComboBox
|
||||
label={localization.ExplanationScatter.colorValue}
|
||||
label={localization.Interpret.ExplanationScatter.colorValue}
|
||||
className={beehiveStyles.pathSelector}
|
||||
selectedKey={this.state.selectedColorOption}
|
||||
onChange={this.setColor}
|
||||
|
|
@ -487,14 +488,14 @@ export class Beehive extends React.PureComponent<
|
|||
<div className={beehiveStyles.sliderControl}>
|
||||
<div className={beehiveStyles.sliderLabel}>
|
||||
<span className={beehiveStyles.labelText}>
|
||||
{localization.AggregateImportance.topKFeatures}
|
||||
{localization.Interpret.AggregateImportance.topKFeatures}
|
||||
</span>
|
||||
{this.props.dashboardContext.explanationContext
|
||||
.isGlobalDerived && (
|
||||
<IconButton
|
||||
id={this._globalSortIconId}
|
||||
iconProps={{ iconName: "Info" }}
|
||||
title={localization.AggregateImportance.topKInfo}
|
||||
title={localization.Interpret.AggregateImportance.topKInfo}
|
||||
onClick={this.showGlobalSortInfo}
|
||||
styles={{
|
||||
root: { color: "rgb(0, 120, 212)", marginBottom: -3 }
|
||||
|
|
@ -504,7 +505,9 @@ export class Beehive extends React.PureComponent<
|
|||
</div>
|
||||
<Slider
|
||||
className={beehiveStyles.featureSlider}
|
||||
ariaLabel={localization.AggregateImportance.topKFeatures}
|
||||
ariaLabel={
|
||||
localization.Interpret.AggregateImportance.topKFeatures
|
||||
}
|
||||
max={Math.min(
|
||||
Beehive.maxFeatures,
|
||||
this.props.dashboardContext.explanationContext.modelMetadata
|
||||
|
|
@ -521,11 +524,11 @@ export class Beehive extends React.PureComponent<
|
|||
.modelType === ModelTypes.Multiclass && (
|
||||
<div>
|
||||
<div className={beehiveStyles.selectorLabel}>
|
||||
<span>{localization.CrossClass.label}</span>
|
||||
<span>{localization.Interpret.CrossClass.label}</span>
|
||||
<IconButton
|
||||
id={this._crossClassIconId}
|
||||
iconProps={{ iconName: "Info" }}
|
||||
title={localization.CrossClass.info}
|
||||
title={localization.Interpret.CrossClass.info}
|
||||
onClick={this.showCrossClassInfo}
|
||||
styles={{
|
||||
root: { color: "rgb(0, 120, 212)", marginBottom: -3 }
|
||||
|
|
@ -557,7 +560,7 @@ export class Beehive extends React.PureComponent<
|
|||
onClick={this.onDismiss}
|
||||
className={beehiveStyles.calloutButton}
|
||||
>
|
||||
{localization.CrossClass.close}
|
||||
{localization.Interpret.CrossClass.close}
|
||||
</DefaultButton>
|
||||
</div>
|
||||
</Callout>
|
||||
|
|
@ -621,11 +624,11 @@ export class Beehive extends React.PureComponent<
|
|||
} else {
|
||||
const calloutContent = (
|
||||
<div>
|
||||
<span>{localization.CrossClass.overviewInfo}</span>
|
||||
<span>{localization.Interpret.CrossClass.overviewInfo}</span>
|
||||
<ul>
|
||||
<li>{localization.CrossClass.absoluteValInfo}</li>
|
||||
<li>{localization.CrossClass.predictedClassInfo}</li>
|
||||
<li>{localization.CrossClass.enumeratedClassInfo}</li>
|
||||
<li>{localization.Interpret.CrossClass.absoluteValInfo}</li>
|
||||
<li>{localization.Interpret.CrossClass.predictedClassInfo}</li>
|
||||
<li>{localization.Interpret.CrossClass.enumeratedClassInfo}</li>
|
||||
</ul>
|
||||
</div>
|
||||
);
|
||||
|
|
@ -640,13 +643,16 @@ export class Beehive extends React.PureComponent<
|
|||
const calloutContent = (
|
||||
<div>
|
||||
<span>
|
||||
{localization.FeatureImportanceWrapper.globalImportanceExplanation}
|
||||
{
|
||||
localization.Interpret.FeatureImportanceWrapper
|
||||
.globalImportanceExplanation
|
||||
}
|
||||
</span>
|
||||
{this.props.dashboardContext.explanationContext.modelMetadata
|
||||
.modelType === ModelTypes.Multiclass && (
|
||||
<span>
|
||||
{
|
||||
localization.FeatureImportanceWrapper
|
||||
localization.Interpret.FeatureImportanceWrapper
|
||||
.multiclassImportanceAddendum
|
||||
}
|
||||
</span>
|
||||
|
|
@ -667,14 +673,14 @@ export class Beehive extends React.PureComponent<
|
|||
const result: IComboBoxOption[] = [
|
||||
{
|
||||
key: "none",
|
||||
text: localization.AggregateImportance.noColor
|
||||
text: localization.Interpret.AggregateImportance.noColor
|
||||
}
|
||||
];
|
||||
if (this.props.dashboardContext.explanationContext.testDataset.dataset) {
|
||||
result.push({
|
||||
data: { isCategorical: false, isNormalized: true },
|
||||
key: "normalizedFeatureValue",
|
||||
text: localization.AggregateImportance.scaledFeatureValue
|
||||
text: localization.Interpret.AggregateImportance.scaledFeatureValue
|
||||
});
|
||||
}
|
||||
if (this.props.dashboardContext.explanationContext.testDataset.predictedY) {
|
||||
|
|
@ -685,8 +691,8 @@ export class Beehive extends React.PureComponent<
|
|||
},
|
||||
key: "predictedClass",
|
||||
text: isRegression
|
||||
? localization.AggregateImportance.predictedValue
|
||||
: localization.AggregateImportance.predictedClass
|
||||
? localization.Interpret.AggregateImportance.predictedValue
|
||||
: localization.Interpret.AggregateImportance.predictedClass
|
||||
});
|
||||
}
|
||||
if (this.props.dashboardContext.explanationContext.testDataset.trueY) {
|
||||
|
|
@ -697,8 +703,8 @@ export class Beehive extends React.PureComponent<
|
|||
},
|
||||
key: "trueClass",
|
||||
text: isRegression
|
||||
? localization.AggregateImportance.trueValue
|
||||
: localization.AggregateImportance.trueClass
|
||||
? localization.Interpret.AggregateImportance.trueValue
|
||||
: localization.Interpret.AggregateImportance.trueClass
|
||||
});
|
||||
}
|
||||
return result;
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
ComboBox,
|
||||
|
|
@ -14,7 +15,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import { ModelTypes, IGlobalExplanation } from "../../IExplanationContext";
|
||||
import { ModelExplanationUtils } from "../../ModelExplanationUtils";
|
||||
|
|
@ -70,7 +70,9 @@ export class FeatureImportanceBar extends React.PureComponent<
|
|||
{this.props.chartTypeOptions &&
|
||||
this.props.chartTypeOptions.length > 1 && (
|
||||
<ComboBox
|
||||
label={localization.FeatureImportanceWrapper.chartType}
|
||||
label={
|
||||
localization.Interpret.FeatureImportanceWrapper.chartType
|
||||
}
|
||||
selectedKey={this.props.config.displayMode}
|
||||
onChange={this.setChart}
|
||||
options={this.props.chartTypeOptions}
|
||||
|
|
@ -82,12 +84,12 @@ export class FeatureImportanceBar extends React.PureComponent<
|
|||
<div className={featureImportanceBarStyles.sliderControl}>
|
||||
<div className={featureImportanceBarStyles.sliderLabel}>
|
||||
<span className={featureImportanceBarStyles.labelText}>
|
||||
{localization.AggregateImportance.topKFeatures}
|
||||
{localization.Interpret.AggregateImportance.topKFeatures}
|
||||
</span>
|
||||
<IconButton
|
||||
id={this._iconId}
|
||||
iconProps={{ iconName: "Info" }}
|
||||
title={localization.AggregateImportance.topKInfo}
|
||||
title={localization.Interpret.AggregateImportance.topKInfo}
|
||||
onClick={this.onIconClick}
|
||||
styles={{
|
||||
root: { color: "rgb(0, 120, 212)", marginBottom: -3 }
|
||||
|
|
@ -102,12 +104,14 @@ export class FeatureImportanceBar extends React.PureComponent<
|
|||
value={this.props.config.topK}
|
||||
onChange={this.setTopK}
|
||||
showValue={true}
|
||||
ariaLabel={localization.AggregateImportance.topKFeatures}
|
||||
ariaLabel={
|
||||
localization.Interpret.AggregateImportance.topKFeatures
|
||||
}
|
||||
/>
|
||||
</div>
|
||||
{this.sortOptions.length > 0 && (
|
||||
<ComboBox
|
||||
label={localization.BarChart.sortBy}
|
||||
label={localization.Interpret.BarChart.sortBy}
|
||||
selectedKey={this.state.selectedSorting}
|
||||
onChange={this.onSortSelect}
|
||||
options={this.sortOptions}
|
||||
|
|
@ -126,15 +130,19 @@ export class FeatureImportanceBar extends React.PureComponent<
|
|||
>
|
||||
<div className={featureImportanceBarStyles.calloutInfo}>
|
||||
<div>
|
||||
<span>{localization.CrossClass.overviewInfo}</span>
|
||||
<span>{localization.Interpret.CrossClass.overviewInfo}</span>
|
||||
<ul>
|
||||
<li>{localization.CrossClass.absoluteValInfo}</li>
|
||||
<li>{localization.CrossClass.predictedClassInfo}</li>
|
||||
<li>{localization.CrossClass.enumeratedClassInfo}</li>
|
||||
<li>{localization.Interpret.CrossClass.absoluteValInfo}</li>
|
||||
<li>
|
||||
{localization.Interpret.CrossClass.predictedClassInfo}
|
||||
</li>
|
||||
<li>
|
||||
{localization.Interpret.CrossClass.enumeratedClassInfo}
|
||||
</li>
|
||||
</ul>
|
||||
</div>
|
||||
<DefaultButton onClick={this.onDismiss}>
|
||||
{localization.CrossClass.close}
|
||||
{localization.Interpret.CrossClass.close}
|
||||
</DefaultButton>
|
||||
</div>
|
||||
</Callout>
|
||||
|
|
@ -201,7 +209,7 @@ export class FeatureImportanceBar extends React.PureComponent<
|
|||
const result: IDropdownOption[] = [
|
||||
{
|
||||
key: FeatureKeys.AbsoluteGlobal,
|
||||
text: localization.BarChart.absoluteGlobal
|
||||
text: localization.Interpret.BarChart.absoluteGlobal
|
||||
}
|
||||
];
|
||||
result.push(
|
||||
|
|
|
|||
|
|
@ -1,11 +1,11 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { SelectionContext } from "@responsible-ai/mlchartlib";
|
||||
import { IComboBoxOption } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { IDashboardContext } from "../../ExplanationDashboard";
|
||||
import { HelpMessageDict } from "../../Interfaces/IStringsParam";
|
||||
import { IBarChartConfig } from "../../SharedComponents/IBarChartConfig";
|
||||
|
|
@ -44,15 +44,15 @@ export class FeatureImportanceWrapper extends React.PureComponent<
|
|||
? [
|
||||
{
|
||||
key: FeatureImportanceModes.Box,
|
||||
text: localization.FeatureImportanceWrapper.boxText
|
||||
text: localization.Interpret.FeatureImportanceWrapper.boxText
|
||||
},
|
||||
{
|
||||
key: FeatureImportanceModes.Beehive,
|
||||
text: localization.FeatureImportanceWrapper.beehiveText
|
||||
text: localization.Interpret.FeatureImportanceWrapper.beehiveText
|
||||
},
|
||||
{
|
||||
key: FeatureImportanceModes.Violin,
|
||||
text: localization.FeatureImportanceWrapper.violinText
|
||||
text: localization.Interpret.FeatureImportanceWrapper.violinText
|
||||
}
|
||||
]
|
||||
: [];
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
ChartBuilder,
|
||||
AccessibleChart,
|
||||
|
|
@ -20,7 +21,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import { IExplanationContext, ModelTypes } from "../../IExplanationContext";
|
||||
import { ModelExplanationUtils } from "../../ModelExplanationUtils";
|
||||
|
|
@ -236,7 +236,7 @@ export class Violin extends React.PureComponent<
|
|||
},
|
||||
yaxis: {
|
||||
automargin: true,
|
||||
title: localization.featureImportance
|
||||
title: localization.Interpret.featureImportance
|
||||
}
|
||||
} as any
|
||||
};
|
||||
|
|
@ -276,7 +276,7 @@ export class Violin extends React.PureComponent<
|
|||
},
|
||||
yaxis: {
|
||||
automargin: true,
|
||||
title: localization.featureImportance
|
||||
title: localization.Interpret.featureImportance
|
||||
}
|
||||
} as any
|
||||
};
|
||||
|
|
@ -337,7 +337,7 @@ export class Violin extends React.PureComponent<
|
|||
<div className={violinStyles.aggregateChart}>
|
||||
<div className={violinStyles.topControls}>
|
||||
<ComboBox
|
||||
label={localization.FeatureImportanceWrapper.chartType}
|
||||
label={localization.Interpret.FeatureImportanceWrapper.chartType}
|
||||
className={violinStyles.pathSelector}
|
||||
selectedKey={this.props.config.displayMode}
|
||||
onChange={this.setChart}
|
||||
|
|
@ -350,7 +350,7 @@ export class Violin extends React.PureComponent<
|
|||
.modelType !== ModelTypes.Regression &&
|
||||
this.groupByOptions.length > 1 && (
|
||||
<ComboBox
|
||||
label={localization.Violin.groupBy}
|
||||
label={localization.Interpret.Violin.groupBy}
|
||||
className={violinStyles.pathSelector}
|
||||
selectedKey={this.state.groupBy}
|
||||
onChange={this.onGroupSelect}
|
||||
|
|
@ -363,12 +363,12 @@ export class Violin extends React.PureComponent<
|
|||
<div className={violinStyles.sliderControl}>
|
||||
<div className={violinStyles.sliderLabel}>
|
||||
<span className={violinStyles.labelText}>
|
||||
{localization.AggregateImportance.topKFeatures}
|
||||
{localization.Interpret.AggregateImportance.topKFeatures}
|
||||
</span>
|
||||
<IconButton
|
||||
id={this._globalSortIconId}
|
||||
iconProps={{ iconName: "Info" }}
|
||||
title={localization.CrossClass.info}
|
||||
title={localization.Interpret.CrossClass.info}
|
||||
ariaLabel="Info"
|
||||
onClick={this.showGlobalSortInfo}
|
||||
styles={{
|
||||
|
|
@ -395,12 +395,12 @@ export class Violin extends React.PureComponent<
|
|||
<div>
|
||||
<div className={violinStyles.selectorLabel}>
|
||||
<span className={violinStyles.selectorSpan}>
|
||||
{localization.CrossClass.label}
|
||||
{localization.Interpret.CrossClass.label}
|
||||
</span>
|
||||
<IconButton
|
||||
id={this._crossClassIconId}
|
||||
iconProps={{ iconName: "Info" }}
|
||||
title={localization.CrossClass.info}
|
||||
title={localization.Interpret.CrossClass.info}
|
||||
ariaLabel="Info"
|
||||
onClick={this.showCrossClassInfo}
|
||||
styles={{
|
||||
|
|
@ -433,7 +433,7 @@ export class Violin extends React.PureComponent<
|
|||
className={violinStyles.calloutButton}
|
||||
onClick={this.onDismiss}
|
||||
>
|
||||
{localization.CrossClass.close}
|
||||
{localization.Interpret.CrossClass.close}
|
||||
</DefaultButton>
|
||||
</div>
|
||||
</Callout>
|
||||
|
|
@ -503,7 +503,10 @@ export class Violin extends React.PureComponent<
|
|||
return [];
|
||||
}
|
||||
const result: IDropdownOption[] = [
|
||||
{ key: GroupByOptions.None, text: localization.Violin.groupNone }
|
||||
{
|
||||
key: GroupByOptions.None,
|
||||
text: localization.Interpret.Violin.groupNone
|
||||
}
|
||||
];
|
||||
if (
|
||||
this.props.dashboardContext.explanationContext.testDataset &&
|
||||
|
|
@ -512,7 +515,7 @@ export class Violin extends React.PureComponent<
|
|||
) {
|
||||
result.push({
|
||||
key: GroupByOptions.PredictedY,
|
||||
text: localization.Violin.groupPredicted
|
||||
text: localization.Interpret.Violin.groupPredicted
|
||||
});
|
||||
}
|
||||
if (
|
||||
|
|
@ -522,7 +525,7 @@ export class Violin extends React.PureComponent<
|
|||
) {
|
||||
result.push({
|
||||
key: GroupByOptions.TrueY,
|
||||
text: localization.Violin.groupTrue
|
||||
text: localization.Interpret.Violin.groupTrue
|
||||
});
|
||||
}
|
||||
return result;
|
||||
|
|
@ -552,11 +555,11 @@ export class Violin extends React.PureComponent<
|
|||
} else {
|
||||
const calloutContent = (
|
||||
<div>
|
||||
<span>{localization.CrossClass.overviewInfo}</span>
|
||||
<span>{localization.Interpret.CrossClass.overviewInfo}</span>
|
||||
<ul>
|
||||
<li>{localization.CrossClass.absoluteValInfo}</li>
|
||||
<li>{localization.CrossClass.predictedClassInfo}</li>
|
||||
<li>{localization.CrossClass.enumeratedClassInfo}</li>
|
||||
<li>{localization.Interpret.CrossClass.absoluteValInfo}</li>
|
||||
<li>{localization.Interpret.CrossClass.predictedClassInfo}</li>
|
||||
<li>{localization.Interpret.CrossClass.enumeratedClassInfo}</li>
|
||||
</ul>
|
||||
</div>
|
||||
);
|
||||
|
|
@ -571,13 +574,16 @@ export class Violin extends React.PureComponent<
|
|||
const calloutContent = (
|
||||
<div>
|
||||
<span>
|
||||
{localization.FeatureImportanceWrapper.globalImportanceExplanation}
|
||||
{
|
||||
localization.Interpret.FeatureImportanceWrapper
|
||||
.globalImportanceExplanation
|
||||
}
|
||||
</span>
|
||||
{this.props.dashboardContext.explanationContext.modelMetadata
|
||||
.modelType === ModelTypes.Multiclass && (
|
||||
<span>
|
||||
{
|
||||
localization.FeatureImportanceWrapper
|
||||
localization.Interpret.FeatureImportanceWrapper
|
||||
.multiclassImportanceAddendum
|
||||
}
|
||||
</span>
|
||||
|
|
|
|||
|
|
@ -2,12 +2,12 @@
|
|||
// Licensed under the MIT License.
|
||||
|
||||
import { PartialRequired } from "@responsible-ai/core-ui";
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { IPlotlyProperty, AccessibleChart } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import { getTheme, Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { ChartTypes } from "../../ChartTypes";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import { JointDataset } from "../../JointDataset";
|
||||
|
|
@ -164,9 +164,10 @@ export class FeatureImportanceBar extends React.PureComponent<
|
|||
baseSeries.layout.barmode = "group";
|
||||
let hovertemplate = this.props.unsortedSeries[0].unsortedFeatureValues
|
||||
? "%{text}: %{customdata.Yvalue}<br>"
|
||||
: localization.Charts.featurePrefix + ": %{text}<br>";
|
||||
: localization.Interpret.Charts.featurePrefix + ": %{text}<br>";
|
||||
hovertemplate +=
|
||||
localization.Charts.importancePrefix + ": %{customdata.Yformatted}<br>";
|
||||
localization.Interpret.Charts.importancePrefix +
|
||||
": %{customdata.Yformatted}<br>";
|
||||
hovertemplate += "%{customdata.Name}<br>";
|
||||
hovertemplate += "<extra></extra>";
|
||||
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { Dictionary } from "lodash";
|
||||
import {
|
||||
ComboBox,
|
||||
|
|
@ -15,7 +16,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { ChartTypes } from "../../ChartTypes";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
|
|
@ -144,7 +144,7 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
<div className={classNames.missingParametersPlaceholder}>
|
||||
<div className={classNames.missingParametersPlaceholderSpacer}>
|
||||
<Text variant="large" className={classNames.faintText}>
|
||||
{localization.GlobalTab.missingParameters}
|
||||
{localization.Interpret.GlobalTab.missingParameters}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -173,7 +173,7 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
<div className={classNames.infoWithText}>
|
||||
<Icon iconName="Info" className={classNames.infoIcon} />
|
||||
<Text variant="medium" className={classNames.helperText}>
|
||||
{localization.GlobalTab.helperText}
|
||||
{localization.Interpret.GlobalTab.helperText}
|
||||
</Text>
|
||||
</div>
|
||||
<div
|
||||
|
|
@ -182,12 +182,12 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
>
|
||||
<Slider
|
||||
label={localization.formatString(
|
||||
localization.GlobalTab.topAtoB,
|
||||
localization.Interpret.GlobalTab.topAtoB,
|
||||
1,
|
||||
this.state.topK
|
||||
)}
|
||||
className={classNames.startingK}
|
||||
ariaLabel={localization.AggregateImportance.topKFeatures}
|
||||
ariaLabel={localization.Interpret.AggregateImportance.topKFeatures}
|
||||
max={this.props.jointDataset.localExplanationFeatureCount}
|
||||
min={1}
|
||||
step={1}
|
||||
|
|
@ -199,14 +199,16 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
<div className={classNames.rightJustifiedContainer}>
|
||||
{this.explainerCalloutInfo && (
|
||||
<LabelWithCallout
|
||||
label={localization.ExplanationSummary.whatDoExplanationsMean}
|
||||
label={
|
||||
localization.Interpret.ExplanationSummary.whatDoExplanationsMean
|
||||
}
|
||||
calloutTitle={this.explainerCalloutInfo.title}
|
||||
type="button"
|
||||
>
|
||||
<Text block>{this.explainerCalloutInfo.description}</Text>
|
||||
{this.explainerCalloutInfo.linkUrl && (
|
||||
<Link href={this.explainerCalloutInfo.linkUrl} target="_blank">
|
||||
{localization.ExplanationSummary.clickHere}
|
||||
{localization.Interpret.ExplanationSummary.clickHere}
|
||||
</Link>
|
||||
)}
|
||||
</LabelWithCallout>
|
||||
|
|
@ -215,7 +217,9 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
<div className={classNames.globalChartWithLegend}>
|
||||
<FeatureImportanceBar
|
||||
jointDataset={this.props.jointDataset}
|
||||
yAxisLabels={[localization.GlobalTab.aggregateFeatureImportance]}
|
||||
yAxisLabels={[
|
||||
localization.Interpret.GlobalTab.aggregateFeatureImportance
|
||||
]}
|
||||
sortArray={this.state.sortArray}
|
||||
chartType={this.state.chartType}
|
||||
unsortedX={this.props.metadata.featureNamesAbridged}
|
||||
|
|
@ -244,7 +248,7 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
<div className={classNames.missingParametersPlaceholder}>
|
||||
<div className={classNames.missingParametersPlaceholderSpacer}>
|
||||
<Text variant="large" className={classNames.faintText}>
|
||||
{localization.GlobalTab.datasetRequired}
|
||||
{localization.Interpret.GlobalTab.datasetRequired}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -253,11 +257,15 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
<div>
|
||||
<div className={classNames.rightJustifiedContainer}>
|
||||
<LabelWithCallout
|
||||
label={localization.Charts.howToRead}
|
||||
calloutTitle={localization.GlobalTab.dependencePlotTitle}
|
||||
label={localization.Interpret.Charts.howToRead}
|
||||
calloutTitle={
|
||||
localization.Interpret.GlobalTab.dependencePlotTitle
|
||||
}
|
||||
type="button"
|
||||
>
|
||||
<Text>{localization.GlobalTab.dependencePlotHelperText}</Text>
|
||||
<Text>
|
||||
{localization.Interpret.GlobalTab.dependencePlotHelperText}
|
||||
</Text>
|
||||
</LabelWithCallout>
|
||||
</div>
|
||||
<div
|
||||
|
|
@ -280,12 +288,14 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
{featureOptions && (
|
||||
<ComboBox
|
||||
id="DependencePlotFeatureSelection"
|
||||
label={localization.GlobalTab.viewDependencePlotFor}
|
||||
label={
|
||||
localization.Interpret.GlobalTab.viewDependencePlotFor
|
||||
}
|
||||
options={featureOptions}
|
||||
allowFreeform={false}
|
||||
autoComplete="on"
|
||||
placeholder={
|
||||
localization.GlobalTab
|
||||
localization.Interpret.GlobalTab
|
||||
.dependencePlotFeatureSelectPlaceholder
|
||||
}
|
||||
selectedKey={
|
||||
|
|
@ -300,7 +310,9 @@ export class GlobalExplanationTab extends React.PureComponent<
|
|||
)}
|
||||
{cohortOptions && (
|
||||
<Dropdown
|
||||
label={localization.GlobalTab.datasetCohortSelector}
|
||||
label={
|
||||
localization.Interpret.GlobalTab.datasetCohortSelector
|
||||
}
|
||||
options={cohortOptions}
|
||||
selectedKey={this.state.selectedCohortIndex}
|
||||
onChange={this.setSelectedCohort}
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { Dictionary } from "lodash";
|
||||
import {
|
||||
ChoiceGroup,
|
||||
|
|
@ -13,7 +14,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { ChartTypes } from "../../ChartTypes";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
|
|
@ -54,9 +54,12 @@ export class SidePanel extends React.Component<
|
|||
private chartOptions: IChoiceGroupOption[] = [
|
||||
{
|
||||
key: ChartTypes.Bar,
|
||||
text: localization.FeatureImportanceWrapper.barText
|
||||
text: localization.Interpret.FeatureImportanceWrapper.barText
|
||||
},
|
||||
{ key: ChartTypes.Box, text: localization.FeatureImportanceWrapper.boxText }
|
||||
{
|
||||
key: ChartTypes.Box,
|
||||
text: localization.Interpret.FeatureImportanceWrapper.boxText
|
||||
}
|
||||
];
|
||||
public constructor(props: ISidePanelProps) {
|
||||
super(props);
|
||||
|
|
@ -69,8 +72,10 @@ export class SidePanel extends React.Component<
|
|||
const classNames = globalTabStyles();
|
||||
return (
|
||||
<Stack className={classNames.legendAndSort}>
|
||||
<Label>{localization.GlobalTab.datasetCohorts}</Label>
|
||||
<Text variant={"small"}>{localization.GlobalTab.legendHelpText}</Text>
|
||||
<Label>{localization.Interpret.GlobalTab.datasetCohorts}</Label>
|
||||
<Text variant={"small"}>
|
||||
{localization.Interpret.GlobalTab.legendHelpText}
|
||||
</Text>
|
||||
<InteractiveLegend
|
||||
items={this.props.cohortSeries.map((row, rowIndex) => {
|
||||
return {
|
||||
|
|
@ -82,7 +87,7 @@ export class SidePanel extends React.Component<
|
|||
})}
|
||||
/>
|
||||
<Dropdown
|
||||
label={localization.GlobalTab.sortBy}
|
||||
label={localization.Interpret.GlobalTab.sortBy}
|
||||
selectedKey={this.props.sortingSeriesIndex}
|
||||
options={this.props.cohortSeries.map((row, rowIndex) => ({
|
||||
key: rowIndex,
|
||||
|
|
@ -91,7 +96,7 @@ export class SidePanel extends React.Component<
|
|||
onChange={this.onSortChange}
|
||||
/>
|
||||
<ChoiceGroup
|
||||
label={localization.DatasetExplorer.chartType}
|
||||
label={localization.Interpret.DatasetExplorer.chartType}
|
||||
selectedKey={this.props.chartType}
|
||||
options={this.chartOptions}
|
||||
onChange={this.onChartTypeChange}
|
||||
|
|
@ -101,16 +106,22 @@ export class SidePanel extends React.Component<
|
|||
this.state.weightOptions && (
|
||||
<div>
|
||||
<LabelWithCallout
|
||||
calloutTitle={localization.CrossClass.crossClassWeights}
|
||||
label={localization.GlobalTab.weightOptions}
|
||||
calloutTitle={
|
||||
localization.Interpret.CrossClass.crossClassWeights
|
||||
}
|
||||
label={localization.Interpret.GlobalTab.weightOptions}
|
||||
>
|
||||
<Text>{localization.CrossClass.overviewInfo}</Text>
|
||||
<Text>{localization.Interpret.CrossClass.overviewInfo}</Text>
|
||||
<ul>
|
||||
<li>
|
||||
<Text>{localization.CrossClass.absoluteValInfo}</Text>
|
||||
<Text>
|
||||
{localization.Interpret.CrossClass.absoluteValInfo}
|
||||
</Text>
|
||||
</li>
|
||||
<li>
|
||||
<Text>{localization.CrossClass.enumeratedClassInfo}</Text>
|
||||
<Text>
|
||||
{localization.Interpret.CrossClass.enumeratedClassInfo}
|
||||
</Text>
|
||||
</li>
|
||||
</ul>
|
||||
</LabelWithCallout>
|
||||
|
|
|
|||
|
|
@ -1,10 +1,10 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { IDropdownOption, Icon, Slider, Text } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { ChartTypes } from "../../ChartTypes";
|
||||
import { IExplanationModelMetadata } from "../../IExplanationContext";
|
||||
import { ModelExplanationUtils } from "../../ModelExplanationUtils";
|
||||
|
|
@ -42,7 +42,7 @@ export class GlobalOnlyChart extends React.PureComponent<
|
|||
? [
|
||||
{
|
||||
colorIndex: 0,
|
||||
name: localization.BarChart.absoluteGlobal,
|
||||
name: localization.Interpret.BarChart.absoluteGlobal,
|
||||
unsortedAggregateY:
|
||||
this.props.globalImportance?.map((classArray) => classArray[0]) ||
|
||||
[]
|
||||
|
|
@ -69,7 +69,7 @@ export class GlobalOnlyChart extends React.PureComponent<
|
|||
);
|
||||
this.classOptions.unshift({
|
||||
key: FeatureKeys.AbsoluteGlobal,
|
||||
text: localization.BarChart.absoluteGlobal
|
||||
text: localization.Interpret.BarChart.absoluteGlobal
|
||||
});
|
||||
this.state = {
|
||||
sortArray: ModelExplanationUtils.buildSortedVector(
|
||||
|
|
@ -87,20 +87,20 @@ export class GlobalOnlyChart extends React.PureComponent<
|
|||
<div className={classNames.infoWithText}>
|
||||
<Icon iconName="Info" className={classNames.infoIcon} />
|
||||
<Text variant="medium" className={classNames.helperText}>
|
||||
{localization.GlobalOnlyChart.helperText}
|
||||
{localization.Interpret.GlobalOnlyChart.helperText}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={classNames.globalChartControls}>
|
||||
<Text variant="medium" className={classNames.sliderLabel}>
|
||||
{localization.formatString(
|
||||
localization.GlobalTab.topAtoB,
|
||||
localization.Interpret.GlobalTab.topAtoB,
|
||||
+1,
|
||||
+this.state.topK
|
||||
)}
|
||||
</Text>
|
||||
<Slider
|
||||
className={classNames.startingK}
|
||||
ariaLabel={localization.AggregateImportance.topKFeatures}
|
||||
ariaLabel={localization.Interpret.AggregateImportance.topKFeatures}
|
||||
max={this.featureDimension}
|
||||
min={1}
|
||||
step={1}
|
||||
|
|
@ -112,7 +112,9 @@ export class GlobalOnlyChart extends React.PureComponent<
|
|||
<div className={classNames.globalChartWithLegend}>
|
||||
<FeatureImportanceBar
|
||||
jointDataset={undefined}
|
||||
yAxisLabels={[localization.GlobalTab.aggregateFeatureImportance]}
|
||||
yAxisLabels={[
|
||||
localization.Interpret.GlobalTab.aggregateFeatureImportance
|
||||
]}
|
||||
sortArray={this.state.sortArray}
|
||||
chartType={ChartTypes.Bar}
|
||||
unsortedX={this.props.metadata.featureNamesAbridged}
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
IPlotlyProperty,
|
||||
AccessibleChart,
|
||||
|
|
@ -11,7 +12,6 @@ import memoize from "memoize-one";
|
|||
import { IComboBoxOption } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { Cohort } from "../Cohort";
|
||||
import { FabricStyles } from "../FabricStyles";
|
||||
import { IExplanationModelMetadata } from "../IExplanationContext";
|
||||
|
|
@ -110,7 +110,7 @@ export class GlobalViolinPlot extends React.PureComponent<
|
|||
},
|
||||
yaxis: {
|
||||
automargin: true,
|
||||
title: localization.featureImportance
|
||||
title: localization.Interpret.featureImportance
|
||||
}
|
||||
} as any
|
||||
};
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
AccessibleChart,
|
||||
ICategoricalRange,
|
||||
|
|
@ -21,7 +22,6 @@ import {
|
|||
import { Data } from "plotly.js";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { FabricStyles } from "../FabricStyles";
|
||||
import { IExplanationContext, ModelTypes } from "../IExplanationContext";
|
||||
import { HelpMessageDict } from "../Interfaces/IStringsParam";
|
||||
|
|
@ -128,8 +128,8 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
automargin: true,
|
||||
title:
|
||||
modelType === ModelTypes.Regression
|
||||
? localization.IcePlot.prediction
|
||||
: localization.IcePlot.predictedProbability
|
||||
? localization.Interpret.IcePlot.prediction
|
||||
: localization.Interpret.IcePlot.predictedProbability
|
||||
}
|
||||
} as any
|
||||
};
|
||||
|
|
@ -192,7 +192,10 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
const result = [];
|
||||
if (modelType !== ModelTypes.Regression) {
|
||||
result.push(
|
||||
localization.formatString(localization.BarChart.classLabel, className)
|
||||
localization.formatString(
|
||||
localization.Interpret.BarChart.classLabel,
|
||||
className
|
||||
)
|
||||
);
|
||||
}
|
||||
if (!Number.isNaN(+xValue)) {
|
||||
|
|
@ -206,14 +209,14 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
if (modelType === ModelTypes.Regression) {
|
||||
result.push(
|
||||
localization.formatString(
|
||||
localization.IcePlot.predictionLabel,
|
||||
localization.Interpret.IcePlot.predictionLabel,
|
||||
yData[index].toLocaleString(undefined, { minimumFractionDigits: 3 })
|
||||
)
|
||||
);
|
||||
} else {
|
||||
result.push(
|
||||
localization.formatString(
|
||||
localization.IcePlot.probabilityLabel,
|
||||
localization.Interpret.IcePlot.probabilityLabel,
|
||||
yData[index].toLocaleString(undefined, { minimumFractionDigits: 3 })
|
||||
)
|
||||
);
|
||||
|
|
@ -286,7 +289,7 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
<ComboBox
|
||||
options={this.featuresOption}
|
||||
onChange={this.onFeatureSelected}
|
||||
label={localization.IcePlot.featurePickerLabel}
|
||||
label={localization.Interpret.IcePlot.featurePickerLabel}
|
||||
ariaLabel="feature picker"
|
||||
selectedKey={
|
||||
this.state.rangeView
|
||||
|
|
@ -315,21 +318,21 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
{this.state.rangeView.type !== RangeTypes.Categorical && (
|
||||
<div className={iCEPlotStyles.parameterSet}>
|
||||
<TextField
|
||||
label={localization.IcePlot.minimumInputLabel}
|
||||
label={localization.Interpret.IcePlot.minimumInputLabel}
|
||||
styles={FabricStyles.textFieldStyle}
|
||||
value={this.state.rangeView.min?.toString()}
|
||||
onChange={this.onMinRangeChanged}
|
||||
errorMessage={this.state.rangeView.minErrorMessage}
|
||||
/>
|
||||
<TextField
|
||||
label={localization.IcePlot.maximumInputLabel}
|
||||
label={localization.Interpret.IcePlot.maximumInputLabel}
|
||||
styles={FabricStyles.textFieldStyle}
|
||||
value={this.state.rangeView.max?.toString()}
|
||||
onChange={this.onMaxRangeChanged}
|
||||
errorMessage={this.state.rangeView.maxErrorMessage}
|
||||
/>
|
||||
<TextField
|
||||
label={localization.IcePlot.stepInputLabel}
|
||||
label={localization.Interpret.IcePlot.stepInputLabel}
|
||||
styles={FabricStyles.textFieldStyle}
|
||||
value={this.state.rangeView.steps?.toString()}
|
||||
onChange={this.onStepsRangeChanged}
|
||||
|
|
@ -343,7 +346,7 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
</div>
|
||||
{this.state.abortController !== undefined && (
|
||||
<div className={iCEPlotStyles.loading}>
|
||||
{localization.IcePlot.loadingMessage}
|
||||
{localization.Interpret.IcePlot.loadingMessage}
|
||||
</div>
|
||||
)}
|
||||
{this.state.errorMessage && (
|
||||
|
|
@ -351,9 +354,11 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
)}
|
||||
{plotlyProps === undefined &&
|
||||
this.state.abortController === undefined && (
|
||||
<div>{localization.IcePlot.submitPrompt}</div>
|
||||
<div>{localization.Interpret.IcePlot.submitPrompt}</div>
|
||||
)}
|
||||
{hasError && <div>{localization.IcePlot.topLevelErrorMessage}</div>}
|
||||
{hasError && (
|
||||
<div>{localization.Interpret.IcePlot.topLevelErrorMessage}</div>
|
||||
)}
|
||||
{plotlyProps !== undefined &&
|
||||
this.state.abortController === undefined &&
|
||||
!hasError && (
|
||||
|
|
@ -436,8 +441,8 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
) {
|
||||
rangeView.minErrorMessage =
|
||||
rangeView.type === RangeTypes.Integer
|
||||
? localization.IcePlot.integerError
|
||||
: localization.IcePlot.numericError;
|
||||
? localization.Interpret.IcePlot.integerError
|
||||
: localization.Interpret.IcePlot.numericError;
|
||||
this.setState({ rangeView });
|
||||
} else {
|
||||
rangeView.minErrorMessage = undefined;
|
||||
|
|
@ -463,8 +468,8 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
) {
|
||||
rangeView.maxErrorMessage =
|
||||
rangeView.type === RangeTypes.Integer
|
||||
? localization.IcePlot.integerError
|
||||
: localization.IcePlot.numericError;
|
||||
? localization.Interpret.IcePlot.integerError
|
||||
: localization.Interpret.IcePlot.numericError;
|
||||
this.setState({ rangeView });
|
||||
} else {
|
||||
rangeView.maxErrorMessage = undefined;
|
||||
|
|
@ -485,7 +490,7 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
}
|
||||
rangeView.steps = val;
|
||||
if (!Number.isInteger(val)) {
|
||||
rangeView.stepsErrorMessage = localization.IcePlot.integerError;
|
||||
rangeView.stepsErrorMessage = localization.Interpret.IcePlot.integerError;
|
||||
this.setState({ rangeView });
|
||||
} else {
|
||||
rangeView.stepsErrorMessage = undefined;
|
||||
|
|
@ -573,7 +578,7 @@ export class ICEPlot extends React.Component<IIcePlotProps, IIcePlotState> {
|
|||
if (error.name === "PythonError") {
|
||||
this.setState({
|
||||
errorMessage: localization.formatString(
|
||||
localization.IcePlot.errorPrefix,
|
||||
localization.Interpret.IcePlot.errorPrefix,
|
||||
error.message
|
||||
)
|
||||
});
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { AccessibleChart, IPlotlyProperty } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -14,7 +15,6 @@ import {
|
|||
import { Transform } from "plotly.js";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { ChartTypes } from "../../ChartTypes";
|
||||
import { Cohort } from "../../Cohort";
|
||||
import { cohortKey } from "../../cohortKey";
|
||||
|
|
@ -177,7 +177,7 @@ export class ModelPerformanceTab extends React.PureComponent<
|
|||
const x = new Array(rawY.length).fill(1);
|
||||
plotlyProps.data[0].text = rawY.map((index) => yLabels?.[index] || "");
|
||||
plotlyProps.data[0].hoverinfo = "all";
|
||||
plotlyProps.data[0].hovertemplate = ` ${yAxisName}:%{y}<br> ${localization.Charts.count}: %{x}<br>`;
|
||||
plotlyProps.data[0].hovertemplate = ` ${yAxisName}:%{y}<br> ${localization.Interpret.Charts.count}: %{x}<br>`;
|
||||
plotlyProps.data[0].y = rawY;
|
||||
plotlyProps.data[0].x = x;
|
||||
plotlyProps.data[0].marker = {};
|
||||
|
|
@ -226,7 +226,7 @@ export class ModelPerformanceTab extends React.PureComponent<
|
|||
<div className={classNames.missingParametersPlaceholder}>
|
||||
<div className={classNames.missingParametersPlaceholderSpacer}>
|
||||
<Text variant="large" className={classNames.faintText}>
|
||||
{localization.ModelPerformance.missingParameters}
|
||||
{localization.Interpret.ModelPerformance.missingParameters}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -254,13 +254,13 @@ export class ModelPerformanceTab extends React.PureComponent<
|
|||
<div className={classNames.infoWithText}>
|
||||
<Icon iconName="Info" className={classNames.infoIcon} />
|
||||
<Text variant="medium" className={classNames.helperText}>
|
||||
{localization.ModelPerformance.helperText}
|
||||
{localization.Interpret.ModelPerformance.helperText}
|
||||
</Text>
|
||||
</div>
|
||||
{cohortOptions && (
|
||||
<div className={classNames.cohortPickerWrapper}>
|
||||
<Text variant="mediumPlus" className={classNames.cohortPickerLabel}>
|
||||
{localization.ModelPerformance.cohortPickerLabel}
|
||||
{localization.Interpret.ModelPerformance.cohortPickerLabel}
|
||||
</Text>
|
||||
<Dropdown
|
||||
styles={{ dropdown: { width: 150 } }}
|
||||
|
|
@ -345,7 +345,10 @@ export class ModelPerformanceTab extends React.PureComponent<
|
|||
variant="large"
|
||||
className={classNames.faintText}
|
||||
>
|
||||
{localization.ModelPerformance.missingTrueY}
|
||||
{
|
||||
localization.Interpret.ModelPerformance
|
||||
.missingTrueY
|
||||
}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
|
|
|
|||
|
|
@ -2,6 +2,7 @@
|
|||
// Licensed under the MIT License.
|
||||
|
||||
import { isTwoDimArray } from "@responsible-ai/core-ui";
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
IPlotlyProperty,
|
||||
RangeTypes,
|
||||
|
|
@ -20,7 +21,6 @@ import {
|
|||
import { Data } from "plotly.js";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import {
|
||||
ModelTypes,
|
||||
|
|
@ -75,13 +75,13 @@ export class MultiICEPlot extends React.PureComponent<
|
|||
selectedClass: number
|
||||
): string {
|
||||
if (metadata.modelType === ModelTypes.Regression) {
|
||||
return localization.IcePlot.prediction;
|
||||
return localization.Interpret.IcePlot.prediction;
|
||||
}
|
||||
return (
|
||||
localization.IcePlot.predictedProbability +
|
||||
localization.Interpret.IcePlot.predictedProbability +
|
||||
"<br>" +
|
||||
localization.formatString(
|
||||
localization.WhatIfTab.classLabel,
|
||||
localization.Interpret.WhatIfTab.classLabel,
|
||||
metadata.classNames[selectedClass]
|
||||
)
|
||||
);
|
||||
|
|
@ -112,8 +112,8 @@ export class MultiICEPlot extends React.PureComponent<
|
|||
: [singleRow];
|
||||
const predictionLabel =
|
||||
metadata.modelType === ModelTypes.Regression
|
||||
? localization.IcePlot.prediction
|
||||
: localization.IcePlot.predictedProbability +
|
||||
? localization.Interpret.IcePlot.prediction
|
||||
: localization.Interpret.IcePlot.predictedProbability +
|
||||
": " +
|
||||
metadata.classNames[selectedClass];
|
||||
const hovertemplate = `%{customdata.Name}<br>${featureName}: %{x}<br>${predictionLabel}: %{customdata.Yformatted}<br><extra></extra>`;
|
||||
|
|
@ -250,7 +250,7 @@ export class MultiICEPlot extends React.PureComponent<
|
|||
},
|
||||
spinButtonWrapper: { maxWidth: "68px" }
|
||||
}}
|
||||
label={localization.WhatIfTab.minLabel}
|
||||
label={localization.Interpret.WhatIfTab.minLabel}
|
||||
value={this.state.rangeView.min?.toString()}
|
||||
onIncrement={this.onMinRangeChanged.bind(this, 1)}
|
||||
onDecrement={this.onMinRangeChanged.bind(this, -1)}
|
||||
|
|
@ -270,7 +270,7 @@ export class MultiICEPlot extends React.PureComponent<
|
|||
},
|
||||
spinButtonWrapper: { maxWidth: "68px" }
|
||||
}}
|
||||
label={localization.WhatIfTab.maxLabel}
|
||||
label={localization.Interpret.WhatIfTab.maxLabel}
|
||||
value={this.state.rangeView.max?.toString()}
|
||||
onIncrement={this.onMaxRangeChanged.bind(this, 1)}
|
||||
onDecrement={this.onMaxRangeChanged.bind(this, -1)}
|
||||
|
|
@ -290,7 +290,7 @@ export class MultiICEPlot extends React.PureComponent<
|
|||
},
|
||||
spinButtonWrapper: { maxWidth: "68px" }
|
||||
}}
|
||||
label={localization.WhatIfTab.stepsLabel}
|
||||
label={localization.Interpret.WhatIfTab.stepsLabel}
|
||||
value={this.state.rangeView.steps?.toString()}
|
||||
onIncrement={this.onStepsRangeChanged.bind(this, 1)}
|
||||
onDecrement={this.onStepsRangeChanged.bind(this, -1)}
|
||||
|
|
@ -302,7 +302,7 @@ export class MultiICEPlot extends React.PureComponent<
|
|||
)}
|
||||
{hasOutgoingRequest && (
|
||||
<div className={classNames.placeholder}>
|
||||
<Text>{localization.IcePlot.loadingMessage}</Text>
|
||||
<Text>{localization.Interpret.IcePlot.loadingMessage}</Text>
|
||||
</div>
|
||||
)}
|
||||
{this.state.errorMessage && (
|
||||
|
|
@ -312,12 +312,12 @@ export class MultiICEPlot extends React.PureComponent<
|
|||
)}
|
||||
{plotlyProps === undefined && !hasOutgoingRequest && (
|
||||
<div className={classNames.placeholder}>
|
||||
<Text>{localization.IcePlot.submitPrompt}</Text>
|
||||
<Text>{localization.Interpret.IcePlot.submitPrompt}</Text>
|
||||
</div>
|
||||
)}
|
||||
{hasError && (
|
||||
<div className={classNames.placeholder}>
|
||||
<Text>{localization.IcePlot.topLevelErrorMessage}</Text>
|
||||
<Text>{localization.Interpret.IcePlot.topLevelErrorMessage}</Text>
|
||||
</div>
|
||||
)}
|
||||
{plotlyProps !== undefined && !hasOutgoingRequest && !hasError && (
|
||||
|
|
@ -523,7 +523,7 @@ export class MultiICEPlot extends React.PureComponent<
|
|||
if (error.name === "PythonError") {
|
||||
this.setState({
|
||||
errorMessage: localization.formatString(
|
||||
localization.IcePlot.errorPrefix,
|
||||
localization.Interpret.IcePlot.errorPrefix,
|
||||
error.message
|
||||
)
|
||||
});
|
||||
|
|
|
|||
|
|
@ -1,11 +1,11 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { ICategoricalRange } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { IExplanationContext } from "../IExplanationContext";
|
||||
import { HelpMessageDict } from "../Interfaces/IStringsParam";
|
||||
import { FeatureEditingTile } from "../SharedComponents/FeatureEditingTile";
|
||||
|
|
@ -108,7 +108,7 @@ export class PerturbationExploration extends React.Component<
|
|||
)}
|
||||
{this.state.abortController && !hasErrors && (
|
||||
<div className={perturbationExplorationStyles.loadingMessage}>
|
||||
{localization.PerturbationExploration.loadingMessage}
|
||||
{localization.Interpret.PerturbationExploration.loadingMessage}
|
||||
</div>
|
||||
)}
|
||||
{this.state.errorMessage && (
|
||||
|
|
@ -119,7 +119,7 @@ export class PerturbationExploration extends React.Component<
|
|||
)}
|
||||
{hasErrors && (
|
||||
<div className={perturbationExplorationStyles.loadingMessage}>
|
||||
{localization.IcePlot.topLevelErrorMessage}
|
||||
{localization.Interpret.IcePlot.topLevelErrorMessage}
|
||||
</div>
|
||||
)}
|
||||
{!hasErrors &&
|
||||
|
|
@ -127,7 +127,10 @@ export class PerturbationExploration extends React.Component<
|
|||
this.state.abortController === undefined && (
|
||||
<div className={perturbationExplorationStyles.labelGroup}>
|
||||
<div className={perturbationExplorationStyles.labelGroupLabel}>
|
||||
{localization.PerturbationExploration.perturbationLabel}
|
||||
{
|
||||
localization.Interpret.PerturbationExploration
|
||||
.perturbationLabel
|
||||
}
|
||||
</div>
|
||||
<div className={perturbationExplorationStyles.flexFull}>
|
||||
<PredictionLabel
|
||||
|
|
@ -263,7 +266,7 @@ export class PerturbationExploration extends React.Component<
|
|||
if (error.name === "PythonError") {
|
||||
this.setState({
|
||||
errorMessage: localization.formatString(
|
||||
localization.IcePlot.errorPrefix,
|
||||
localization.Interpret.IcePlot.errorPrefix,
|
||||
error.message
|
||||
) as string
|
||||
});
|
||||
|
|
|
|||
|
|
@ -1,12 +1,12 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { AccessibleChart, IPlotlyProperty } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import { ComboBox, IComboBox, IComboBoxOption } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import { NoDataMessage } from "../../SharedComponents/NoDataMessage";
|
||||
|
||||
|
|
@ -70,7 +70,7 @@ export class DataExploration extends React.PureComponent<IScatterProps> {
|
|||
<ComboBox
|
||||
options={dropdownOptions}
|
||||
onChange={this.onXSelected}
|
||||
label={localization.ExplanationScatter.xValue}
|
||||
label={localization.Interpret.ExplanationScatter.xValue}
|
||||
ariaLabel="x picker"
|
||||
selectedKey={this.plotlyProps.data[0].xAccessor}
|
||||
useComboBoxAsMenuWidth={true}
|
||||
|
|
@ -81,7 +81,7 @@ export class DataExploration extends React.PureComponent<IScatterProps> {
|
|||
<ComboBox
|
||||
options={dropdownOptions}
|
||||
onChange={this.onColorSelected}
|
||||
label={localization.ExplanationScatter.colorValue}
|
||||
label={localization.Interpret.ExplanationScatter.colorValue}
|
||||
ariaLabel="color picker"
|
||||
selectedKey={initialColorOption}
|
||||
useComboBoxAsMenuWidth={true}
|
||||
|
|
@ -94,7 +94,7 @@ export class DataExploration extends React.PureComponent<IScatterProps> {
|
|||
<ComboBox
|
||||
options={dropdownOptions}
|
||||
onChange={this.onYSelected}
|
||||
label={localization.ExplanationScatter.yValue}
|
||||
label={localization.Interpret.ExplanationScatter.yValue}
|
||||
ariaLabel="y picker"
|
||||
selectedKey={this.plotlyProps.data[0].yAccessor}
|
||||
useComboBoxAsMenuWidth={true}
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import { AccessibleChart, IPlotlyProperty } from "@responsible-ai/mlchartlib";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
|
|
@ -13,7 +14,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { FabricStyles } from "../../FabricStyles";
|
||||
import { ModelTypes } from "../../IExplanationContext";
|
||||
import { LoadingSpinner } from "../../SharedComponents/LoadingSpinner";
|
||||
|
|
@ -94,7 +94,7 @@ export class ExplanationExploration extends React.PureComponent<
|
|||
<ComboBox
|
||||
options={dropdownOptions}
|
||||
onChange={this.onXSelected}
|
||||
label={localization.ExplanationScatter.xValue}
|
||||
label={localization.Interpret.ExplanationScatter.xValue}
|
||||
ariaLabel="x picker"
|
||||
selectedKey={this.plotlyProps.data[0].xAccessor}
|
||||
useComboBoxAsMenuWidth={true}
|
||||
|
|
@ -105,7 +105,7 @@ export class ExplanationExploration extends React.PureComponent<
|
|||
<ComboBox
|
||||
options={dropdownOptions}
|
||||
onChange={this.onColorSelected}
|
||||
label={localization.ExplanationScatter.colorValue}
|
||||
label={localization.Interpret.ExplanationScatter.colorValue}
|
||||
ariaLabel="color picker"
|
||||
selectedKey={initialColorOption}
|
||||
useComboBoxAsMenuWidth={true}
|
||||
|
|
@ -118,7 +118,7 @@ export class ExplanationExploration extends React.PureComponent<
|
|||
<ComboBox
|
||||
options={dropdownOptions}
|
||||
onChange={this.onYSelected}
|
||||
label={localization.ExplanationScatter.yValue}
|
||||
label={localization.Interpret.ExplanationScatter.yValue}
|
||||
ariaLabel="y picker"
|
||||
selectedKey={this.plotlyProps.data[0].yAccessor}
|
||||
useComboBoxAsMenuWidth={true}
|
||||
|
|
@ -129,12 +129,12 @@ export class ExplanationExploration extends React.PureComponent<
|
|||
<div className={scatterStyles.selector}>
|
||||
<div className={scatterStyles.selectorLabel}>
|
||||
<div className={scatterStyles.labelText}>
|
||||
{localization.CrossClass.label}
|
||||
{localization.Interpret.CrossClass.label}
|
||||
</div>
|
||||
<IconButton
|
||||
id={this.iconId}
|
||||
iconProps={{ iconName: "Info" }}
|
||||
title={localization.CrossClass.info}
|
||||
title={localization.Interpret.CrossClass.info}
|
||||
onClick={this.onIconClick}
|
||||
styles={{
|
||||
root: { color: "rgb(0, 120, 212)", marginBottom: -3 }
|
||||
|
|
@ -162,18 +162,22 @@ export class ExplanationExploration extends React.PureComponent<
|
|||
>
|
||||
<div className={scatterStyles.calloutInfo}>
|
||||
<div>
|
||||
<span>{localization.CrossClass.overviewInfo}</span>
|
||||
<span>{localization.Interpret.CrossClass.overviewInfo}</span>
|
||||
<ul>
|
||||
<li>{localization.CrossClass.absoluteValInfo}</li>
|
||||
<li>{localization.CrossClass.predictedClassInfo}</li>
|
||||
<li>{localization.CrossClass.enumeratedClassInfo}</li>
|
||||
<li>{localization.Interpret.CrossClass.absoluteValInfo}</li>
|
||||
<li>
|
||||
{localization.Interpret.CrossClass.predictedClassInfo}
|
||||
</li>
|
||||
<li>
|
||||
{localization.Interpret.CrossClass.enumeratedClassInfo}
|
||||
</li>
|
||||
</ul>
|
||||
</div>
|
||||
<DefaultButton
|
||||
onClick={this.onDismiss}
|
||||
className={scatterStyles.calloutButton}
|
||||
>
|
||||
{localization.CrossClass.close}
|
||||
{localization.Interpret.CrossClass.close}
|
||||
</DefaultButton>
|
||||
</div>
|
||||
</Callout>
|
||||
|
|
|
|||
|
|
@ -2,6 +2,7 @@
|
|||
// Licensed under the MIT License.
|
||||
|
||||
import { PartialRequired2 } from "@responsible-ai/core-ui";
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
AccessorMappingFunctionNames,
|
||||
ChartBuilder,
|
||||
|
|
@ -17,7 +18,6 @@ import {
|
|||
IDropdownOption
|
||||
} from "office-ui-fabric-react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import { IDashboardContext } from "../../ExplanationDashboard";
|
||||
import {
|
||||
IExplanationContext,
|
||||
|
|
@ -79,7 +79,7 @@ export class ScatterUtils {
|
|||
result.push({
|
||||
itemType: DropdownMenuItemType.Header,
|
||||
key: "Header0",
|
||||
text: localization.featureImportance
|
||||
text: localization.Interpret.featureImportance
|
||||
});
|
||||
explanationContext.modelMetadata.featureNames.forEach(
|
||||
(featureName, index) => {
|
||||
|
|
@ -87,7 +87,7 @@ export class ScatterUtils {
|
|||
data: { isCategorical: false, isFeatureImportance: true },
|
||||
key: `LocalExplanation[${index}]`,
|
||||
text: localization.formatString(
|
||||
localization.ExplanationScatter.importanceLabel,
|
||||
localization.Interpret.ExplanationScatter.importanceLabel,
|
||||
featureName
|
||||
)
|
||||
});
|
||||
|
|
@ -102,7 +102,7 @@ export class ScatterUtils {
|
|||
result.push({
|
||||
itemType: DropdownMenuItemType.Header,
|
||||
key: "Header1",
|
||||
text: localization.ExplanationScatter.dataGroupLabel
|
||||
text: localization.Interpret.ExplanationScatter.dataGroupLabel
|
||||
});
|
||||
explanationContext.modelMetadata.featureNames.forEach(
|
||||
(featureName, index) => {
|
||||
|
|
@ -114,7 +114,7 @@ export class ScatterUtils {
|
|||
key: `TrainingData[${index}]`,
|
||||
text: includeFeatureImportance
|
||||
? localization.formatString(
|
||||
localization.ExplanationScatter.dataLabel,
|
||||
localization.Interpret.ExplanationScatter.dataLabel,
|
||||
featureName
|
||||
)
|
||||
: featureName
|
||||
|
|
@ -124,7 +124,7 @@ export class ScatterUtils {
|
|||
result.push({
|
||||
data: { isCategorical: false },
|
||||
key: "Index",
|
||||
text: localization.ExplanationScatter.index
|
||||
text: localization.Interpret.ExplanationScatter.index
|
||||
});
|
||||
result.push({
|
||||
itemType: DropdownMenuItemType.Divider,
|
||||
|
|
@ -134,7 +134,7 @@ export class ScatterUtils {
|
|||
result.push({
|
||||
itemType: DropdownMenuItemType.Header,
|
||||
key: "Header2",
|
||||
text: localization.ExplanationScatter.output
|
||||
text: localization.Interpret.ExplanationScatter.output
|
||||
});
|
||||
if (explanationContext.testDataset.predictedY) {
|
||||
result.push({
|
||||
|
|
@ -149,7 +149,7 @@ export class ScatterUtils {
|
|||
: undefined
|
||||
},
|
||||
key: "PredictedY",
|
||||
text: localization.ExplanationScatter.predictedY
|
||||
text: localization.Interpret.ExplanationScatter.predictedY
|
||||
});
|
||||
}
|
||||
if (explanationContext.testDataset.probabilityY) {
|
||||
|
|
@ -162,7 +162,7 @@ export class ScatterUtils {
|
|||
data: { isCategorical: false },
|
||||
key: `ProbabilityY[${index}]`,
|
||||
text: localization.formatString(
|
||||
localization.ExplanationScatter.probabilityLabel,
|
||||
localization.Interpret.ExplanationScatter.probabilityLabel,
|
||||
className
|
||||
)
|
||||
});
|
||||
|
|
@ -181,7 +181,7 @@ export class ScatterUtils {
|
|||
: undefined
|
||||
},
|
||||
key: "TrueY",
|
||||
text: localization.ExplanationScatter.trueY
|
||||
text: localization.Interpret.ExplanationScatter.trueY
|
||||
});
|
||||
}
|
||||
return result;
|
||||
|
|
@ -343,8 +343,8 @@ export class ScatterUtils {
|
|||
},
|
||||
key: colorAccessor,
|
||||
text: hasPredictedY
|
||||
? localization.ExplanationScatter.predictedY
|
||||
: localization.ExplanationScatter.index
|
||||
? localization.Interpret.ExplanationScatter.predictedY
|
||||
: localization.Interpret.ExplanationScatter.index
|
||||
};
|
||||
const modelData = exp.modelMetadata;
|
||||
const colorbarTitle = ScatterUtils.formatItemTextForAxis(
|
||||
|
|
@ -360,15 +360,15 @@ export class ScatterUtils {
|
|||
colorAccessor
|
||||
];
|
||||
props.data[0].datapointLevelAccessors!["text"].mapArgs = [
|
||||
localization.ExplanationScatter.index,
|
||||
localization.Interpret.ExplanationScatter.index,
|
||||
modelData.featureNames[maxIndex],
|
||||
localization.ExplanationScatter.predictedY
|
||||
localization.Interpret.ExplanationScatter.predictedY
|
||||
];
|
||||
|
||||
_.set(
|
||||
props,
|
||||
"layout.xaxis.title.text",
|
||||
localization.ExplanationScatter.index
|
||||
localization.Interpret.ExplanationScatter.index
|
||||
);
|
||||
_.set(props, "layout.yaxis.title.text", modelData.featureNames[maxIndex]);
|
||||
|
||||
|
|
@ -424,15 +424,15 @@ export class ScatterUtils {
|
|||
];
|
||||
props.data[0].datapointLevelAccessors!["text"].mapArgs = [
|
||||
localization.formatString(
|
||||
localization.ExplanationScatter.dataLabel,
|
||||
localization.Interpret.ExplanationScatter.dataLabel,
|
||||
modelData.featureNames[maxIndex]
|
||||
),
|
||||
localization.formatString(
|
||||
localization.ExplanationScatter.importanceLabel,
|
||||
localization.Interpret.ExplanationScatter.importanceLabel,
|
||||
modelData.featureNames[maxIndex]
|
||||
),
|
||||
localization.formatString(
|
||||
localization.ExplanationScatter.dataLabel,
|
||||
localization.Interpret.ExplanationScatter.dataLabel,
|
||||
modelData.featureNames[secondIndex]
|
||||
)
|
||||
];
|
||||
|
|
@ -440,11 +440,11 @@ export class ScatterUtils {
|
|||
const yAxisLabel =
|
||||
modelData.modelType === ModelTypes.Binary
|
||||
? localization.formatString(
|
||||
localization.ExplanationScatter.importanceLabel,
|
||||
localization.Interpret.ExplanationScatter.importanceLabel,
|
||||
modelData.featureNames[maxIndex]
|
||||
) + ` : ${modelData.classNames[0]}`
|
||||
: localization.formatString(
|
||||
localization.ExplanationScatter.importanceLabel,
|
||||
localization.Interpret.ExplanationScatter.importanceLabel,
|
||||
modelData.featureNames[maxIndex]
|
||||
);
|
||||
_.set(props, "layout.yaxis.title.text", yAxisLabel);
|
||||
|
|
@ -452,7 +452,7 @@ export class ScatterUtils {
|
|||
props,
|
||||
"layout.xaxis.title.text",
|
||||
localization.formatString(
|
||||
localization.ExplanationScatter.dataLabel,
|
||||
localization.Interpret.ExplanationScatter.dataLabel,
|
||||
modelData.featureNames[maxIndex]
|
||||
)
|
||||
);
|
||||
|
|
@ -618,7 +618,7 @@ export class ScatterUtils {
|
|||
) {
|
||||
// Add the first class's name to the text for binary case, to clarify
|
||||
const className = modelMetadata.classNames[0];
|
||||
return `${item.text}<br> ${localization.ExplanationScatter.class} : ${className}`;
|
||||
return `${item.text}<br> ${localization.Interpret.ExplanationScatter.class} : ${className}`;
|
||||
}
|
||||
return item.text;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import _ from "lodash";
|
||||
import {
|
||||
IDropdownOption,
|
||||
|
|
@ -11,7 +12,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { FabricStyles } from "../FabricStyles";
|
||||
import {
|
||||
IExplanationContext,
|
||||
|
|
@ -109,7 +109,7 @@ export class SinglePointFeatureImportance extends React.PureComponent<
|
|||
<div className={singlePointFeatureImportanceStyles.topControls}>
|
||||
<Slider
|
||||
className={singlePointFeatureImportanceStyles.featureSlider}
|
||||
label={localization.AggregateImportance.topKFeatures}
|
||||
label={localization.Interpret.AggregateImportance.topKFeatures}
|
||||
max={Math.min(
|
||||
30,
|
||||
this.props.explanationContext.modelMetadata.featureNames
|
||||
|
|
@ -123,7 +123,7 @@ export class SinglePointFeatureImportance extends React.PureComponent<
|
|||
/>
|
||||
{this.sortOptions.length > 1 && (
|
||||
<ComboBox
|
||||
label={localization.BarChart.sortBy}
|
||||
label={localization.Interpret.BarChart.sortBy}
|
||||
selectedKey={this.state.selectedSorting}
|
||||
onChange={this.onSortSelect}
|
||||
options={this.sortOptions}
|
||||
|
|
@ -209,7 +209,7 @@ export class SinglePointFeatureImportance extends React.PureComponent<
|
|||
const result: IDropdownOption[] = [
|
||||
{
|
||||
key: FeatureKeys.AbsoluteGlobal,
|
||||
text: localization.BarChart.absoluteGlobal
|
||||
text: localization.Interpret.BarChart.absoluteGlobal
|
||||
}
|
||||
];
|
||||
// if (!this.props.explanationContext.testDataset.predictedY) {
|
||||
|
|
@ -221,7 +221,7 @@ export class SinglePointFeatureImportance extends React.PureComponent<
|
|||
) {
|
||||
result.push({
|
||||
key: FeatureKeys.AbsoluteLocal,
|
||||
text: localization.BarChart.absoluteLocal
|
||||
text: localization.Interpret.BarChart.absoluteLocal
|
||||
});
|
||||
}
|
||||
if (
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
IPlotlyProperty,
|
||||
PlotlyMode,
|
||||
|
|
@ -11,7 +12,6 @@ import memoize from "memoize-one";
|
|||
import { IComboBoxOption } from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../Localization/localization";
|
||||
import { Cohort } from "../Cohort";
|
||||
import { IExplanationModelMetadata, ModelTypes } from "../IExplanationContext";
|
||||
import { JointDataset } from "../JointDataset";
|
||||
|
|
@ -68,7 +68,7 @@ export class SwarmFeaturePlot extends React.PureComponent<
|
|||
_.set(
|
||||
plotlyProps,
|
||||
"layout.yaxis.title",
|
||||
`${localization.featureImportance}<br> ${localization.ExplanationScatter.class} ${metadata.classNames[0]}`
|
||||
`${localization.Interpret.featureImportance}<br> ${localization.Interpret.ExplanationScatter.class} ${metadata.classNames[0]}`
|
||||
);
|
||||
}
|
||||
if (selectedOption === undefined || selectedOption.key === "none") {
|
||||
|
|
@ -87,8 +87,8 @@ export class SwarmFeaturePlot extends React.PureComponent<
|
|||
];
|
||||
_.set(plotlyProps.data[0], "marker.colorbar.tickvals", [0, 1]);
|
||||
_.set(plotlyProps.data[0], "marker.colorbar.ticktext", [
|
||||
localization.AggregateImportance.low,
|
||||
localization.AggregateImportance.high
|
||||
localization.Interpret.AggregateImportance.low,
|
||||
localization.Interpret.AggregateImportance.high
|
||||
]);
|
||||
} else {
|
||||
_.set(plotlyProps.data[0], "marker.opacity", 0.6);
|
||||
|
|
@ -146,7 +146,7 @@ export class SwarmFeaturePlot extends React.PureComponent<
|
|||
},
|
||||
yaxis: {
|
||||
automargin: true,
|
||||
title: localization.featureImportance
|
||||
title: localization.Interpret.featureImportance
|
||||
}
|
||||
} as any
|
||||
};
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
DirectionalHint,
|
||||
IconButton,
|
||||
|
|
@ -11,7 +12,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import {
|
||||
IExplanationModelMetadata,
|
||||
ModelTypes
|
||||
|
|
@ -117,25 +117,25 @@ export class CustomPredictionLabels extends React.Component<
|
|||
<div className={classNames.tooltipWrapper}>
|
||||
<div className={classNames.tooltipTitle}>
|
||||
<Text variant="large">
|
||||
{localization.WhatIfTab.whatIfTooltipTitle}
|
||||
{localization.Interpret.WhatIfTab.whatIfTooltipTitle}
|
||||
</Text>
|
||||
</div>
|
||||
<div className={classNames.tooltipTable}>
|
||||
<div className={classNames.tooltipColumn}>
|
||||
<Text className={classNames.boldText}>
|
||||
{localization.WhatIfTab.classPickerLabel}
|
||||
{localization.Interpret.WhatIfTab.classPickerLabel}
|
||||
</Text>
|
||||
{tooltipClasses}
|
||||
</div>
|
||||
<div className={classNames.tooltipColumn}>
|
||||
<Text block className={classNames.boldText}>
|
||||
{localization.WhatIfTab.probabilityLabel}
|
||||
{localization.Interpret.WhatIfTab.probabilityLabel}
|
||||
</Text>
|
||||
{tooltipProbs}
|
||||
</div>
|
||||
<div className={classNames.tooltipColumn}>
|
||||
<Text block className={classNames.boldText}>
|
||||
{localization.WhatIfTab.deltaLabel}
|
||||
{localization.Interpret.WhatIfTab.deltaLabel}
|
||||
</Text>
|
||||
{tooltipDeltas}
|
||||
</div>
|
||||
|
|
@ -160,13 +160,13 @@ export class CustomPredictionLabels extends React.Component<
|
|||
<div>
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.newPredictedClass}
|
||||
{localization.Interpret.WhatIfTab.newPredictedClass}
|
||||
</Text>
|
||||
<Text variant="small">{predictedClassName}</Text>
|
||||
</div>
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.newProbability}
|
||||
{localization.Interpret.WhatIfTab.newProbability}
|
||||
</Text>
|
||||
<Text variant="small">
|
||||
{predictedProb.toLocaleString(undefined, {
|
||||
|
|
@ -190,10 +190,14 @@ export class CustomPredictionLabels extends React.Component<
|
|||
</div>
|
||||
<div>
|
||||
<div>
|
||||
<Text variant="small">{localization.WhatIfTab.loading}</Text>
|
||||
<Text variant="small">
|
||||
{localization.Interpret.WhatIfTab.loading}
|
||||
</Text>
|
||||
</div>
|
||||
<div>
|
||||
<Text variant="small">{localization.WhatIfTab.loading}</Text>
|
||||
<Text variant="small">
|
||||
{localization.Interpret.WhatIfTab.loading}
|
||||
</Text>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
|
|
@ -206,12 +210,12 @@ export class CustomPredictionLabels extends React.Component<
|
|||
].toLocaleString(undefined, {
|
||||
maximumFractionDigits: 3
|
||||
})
|
||||
: localization.WhatIfTab.loading;
|
||||
: localization.Interpret.WhatIfTab.loading;
|
||||
return (
|
||||
<div className={classNames.customPredictBlock}>
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.newPredictedValue}
|
||||
{localization.Interpret.WhatIfTab.newPredictedValue}
|
||||
</Text>
|
||||
<Text variant="small">{predictedValueString}</Text>
|
||||
</div>
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
// Copyright (c) Microsoft Corporation.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
import { localization } from "@responsible-ai/localization";
|
||||
import {
|
||||
DirectionalHint,
|
||||
IconButton,
|
||||
|
|
@ -11,7 +12,6 @@ import {
|
|||
} from "office-ui-fabric-react";
|
||||
import React from "react";
|
||||
|
||||
import { localization } from "../../../Localization/localization";
|
||||
import {
|
||||
IExplanationModelMetadata,
|
||||
ModelTypes
|
||||
|
|
@ -77,8 +77,8 @@ export class ExistingPredictionLabels extends React.Component<
|
|||
});
|
||||
const tooltipTitle =
|
||||
predictedProbs.length > WhatIfConstants.MAX_CLASSES_TOOLTIP
|
||||
? localization.WhatIfTab.tooltipTitleMany
|
||||
: localization.WhatIfTab.tooltipTitleFew;
|
||||
? localization.Interpret.WhatIfTab.tooltipTitleMany
|
||||
: localization.Interpret.WhatIfTab.tooltipTitleFew;
|
||||
const tooltipProps: ITooltipProps = {
|
||||
onRenderContent: () => (
|
||||
<div className={classNames.tooltipWrapper}>
|
||||
|
|
@ -88,13 +88,13 @@ export class ExistingPredictionLabels extends React.Component<
|
|||
<div className={classNames.tooltipTable}>
|
||||
<div className={classNames.tooltipColumn}>
|
||||
<Text className={classNames.boldText}>
|
||||
{localization.WhatIfTab.classPickerLabel}
|
||||
{localization.Interpret.WhatIfTab.classPickerLabel}
|
||||
</Text>
|
||||
{tooltipClasses}
|
||||
</div>
|
||||
<div className={classNames.tooltipColumn}>
|
||||
<Text block className={classNames.boldText}>
|
||||
{localization.WhatIfTab.probabilityLabel}
|
||||
{localization.Interpret.WhatIfTab.probabilityLabel}
|
||||
</Text>
|
||||
{tooltipProbs}
|
||||
</div>
|
||||
|
|
@ -120,7 +120,7 @@ export class ExistingPredictionLabels extends React.Component<
|
|||
{trueClass !== undefined && (
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.trueClass}
|
||||
{localization.Interpret.WhatIfTab.trueClass}
|
||||
</Text>
|
||||
<Text variant="small">
|
||||
{
|
||||
|
|
@ -133,13 +133,13 @@ export class ExistingPredictionLabels extends React.Component<
|
|||
)}
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.predictedClass}
|
||||
{localization.Interpret.WhatIfTab.predictedClass}
|
||||
</Text>
|
||||
<Text variant="small">{predictedClassName}</Text>
|
||||
</div>
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.probability}
|
||||
{localization.Interpret.WhatIfTab.probability}
|
||||
</Text>
|
||||
<Text variant="small">
|
||||
{predictedProb.toLocaleString(undefined, {
|
||||
|
|
@ -158,7 +158,7 @@ export class ExistingPredictionLabels extends React.Component<
|
|||
{trueClass !== undefined && (
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.trueClass}
|
||||
{localization.Interpret.WhatIfTab.trueClass}
|
||||
</Text>
|
||||
<Text variant="small">
|
||||
{
|
||||
|
|
@ -172,7 +172,7 @@ export class ExistingPredictionLabels extends React.Component<
|
|||
{predictedClass !== undefined && (
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.predictedClass}
|
||||
{localization.Interpret.WhatIfTab.predictedClass}
|
||||
</Text>
|
||||
<Text variant="small">{predictedClassName}</Text>
|
||||
</div>
|
||||
|
|
@ -196,7 +196,7 @@ export class ExistingPredictionLabels extends React.Component<
|
|||
{trueValue !== undefined && (
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.trueValue}
|
||||
{localization.Interpret.WhatIfTab.trueValue}
|
||||
</Text>
|
||||
<Text variant="small">{trueValue}</Text>
|
||||
</div>
|
||||
|
|
@ -204,7 +204,7 @@ export class ExistingPredictionLabels extends React.Component<
|
|||
{predictedValue !== undefined && (
|
||||
<div>
|
||||
<Text className={classNames.boldText} variant="small">
|
||||
{localization.WhatIfTab.predictedValue}
|
||||
{localization.Interpret.WhatIfTab.predictedValue}
|
||||
</Text>
|
||||
<Text variant="small">
|
||||
{predictedValue.toLocaleString(undefined, {
|
||||
|
|
|
|||
Некоторые файлы не были показаны из-за слишком большого количества измененных файлов Показать больше
Загрузка…
Ссылка в новой задаче