зеркало из https://github.com/microsoft/NimbusML.git
Changes made by ChangeHttpURLsToHttps.py
This commit is contained in:
Mustafa Bal
Родитель
e29f6edb75
Коммит
afa5f35fe2
|
|
@ -16,7 +16,7 @@ typedef MANAGED_CALLBACK_PTR(bool, GETLABELS)(DataSourceBlock *source, int col,
|
|||
|
||||
// REVIEW: boost_python is not updated at the same speed as swig or pybind11.
|
||||
// Both have a larger audience now, see about pybind11 https://github.com/davisking/dlib/issues/293
|
||||
// It handles csr_matrix: http://pybind11-rtdtest.readthedocs.io/en/stable/advanced.html#transparent-conversion-of-dense-and-sparse-eigen-data-types.
|
||||
// It handles csr_matrix: https://pybind11-rtdtest.readthedocs.io/en/stable/advanced.html#transparent-conversion-of-dense-and-sparse-eigen-data-types.
|
||||
using namespace boost::python;
|
||||
|
||||
// The data source wrapper used for managed interop. Some of the fields of this are visible to managed code.
|
||||
|
|
|
|||
|
|
@ -45,10 +45,10 @@
|
|||
<https://en.wikipedia.org/wiki/Perceptron>`_
|
||||
|
||||
`Large Margin Classification Using the Perceptron Algorithm
|
||||
<http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.48.8200>`_
|
||||
<https://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.48.8200>`_
|
||||
|
||||
`Discriminative Training Methods for Hidden Markov Models
|
||||
<http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.6725>`_
|
||||
<https://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.6725>`_
|
||||
|
||||
|
||||
:param loss: The default is :py:class:`'hinge' <nimbusml.loss.Hinge>`. Other
|
||||
|
|
|
|||
|
|
@ -22,7 +22,7 @@
|
|||
`Field Aware Factorization Machines
|
||||
<https://www.csie.ntu.edu.tw/~r01922136/slides/ffm.pdf>`_,
|
||||
`Field-aware Factorization Machines for CTR Prediction
|
||||
<http://www.csie.ntu.edu.tw/~cjlin/papers/ffm.pdf>`_,
|
||||
<https://www.csie.ntu.edu.tw/~cjlin/papers/ffm.pdf>`_,
|
||||
`Adaptive Subgradient Methods for Online Learning and Stochastic
|
||||
Optimization
|
||||
<http://jmlr.org/papers/volume12/duchi11a/duchi11a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -33,7 +33,7 @@
|
|||
**Reference**
|
||||
|
||||
`Wikipedia: Random forest
|
||||
<http://en.wikipedia.org/wiki/Random_forest>`_
|
||||
<https://en.wikipedia.org/wiki/Random_forest>`_
|
||||
|
||||
`Quantile regression forest
|
||||
<http://jmlr.org/papers/volume7/meinshausen06a/meinshausen06a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -43,7 +43,7 @@
|
|||
**Reference**
|
||||
|
||||
`Wikipedia: Random forest
|
||||
<http://en.wikipedia.org/wiki/Random_forest>`_
|
||||
<https://en.wikipedia.org/wiki/Random_forest>`_
|
||||
|
||||
`Quantile regression forest
|
||||
<http://jmlr.org/papers/volume7/meinshausen06a/meinshausen06a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -57,7 +57,7 @@
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param optimizer: Default is ``sgd``.
|
||||
|
||||
|
|
|
|||
|
|
@ -62,7 +62,7 @@
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param optimizer: Default is ``sgd``.
|
||||
|
||||
|
|
|
|||
|
|
@ -14,7 +14,7 @@
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param optimizer: Default is ``sgd``.
|
||||
|
||||
|
|
|
|||
|
|
@ -21,7 +21,7 @@
|
|||
functions learned will step between the discretization boundaries.
|
||||
|
||||
This implementation is based on the this `paper
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
but diverges from it in several important respects: most
|
||||
significantly,
|
||||
in each round of boosting, rather than do one feature at a time, it
|
||||
|
|
@ -57,7 +57,7 @@
|
|||
`Generalized additive models
|
||||
<https://en.wikipedia.org/wiki/Generalized_additive_model>`_,
|
||||
`Intelligible Models for Classification and Regression
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
|
||||
|
||||
:param normalize: Specifies the type of automatic normalization used:
|
||||
|
|
|
|||
|
|
@ -21,7 +21,7 @@
|
|||
functions learned will step between the discretization boundaries.
|
||||
|
||||
This implementation is based on the this `paper
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
but diverges from it in several important respects: most
|
||||
significantly,
|
||||
in each round of boosting, rather than do one feature at a time, it
|
||||
|
|
@ -57,7 +57,7 @@
|
|||
`Generalized additive models
|
||||
<https://en.wikipedia.org/wiki/Generalized_additive_model>`_,
|
||||
`Intelligible Models for Classification and Regression
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
|
||||
|
||||
:param normalize: Specifies the type of automatic normalization used:
|
||||
|
|
|
|||
|
|
@ -10,7 +10,7 @@
|
|||
topical vectors. LightLDA is an extremely
|
||||
efficient implementation of LDA developed in MSR-Asia that
|
||||
incorporates a number of optimization techniques
|
||||
`(http://arxiv.org/abs/1412.1576) <http://arxiv.org/abs/1412.1576>`_.
|
||||
`(https://arxiv.org/abs/1412.1576) <https://arxiv.org/abs/1412.1576>`_.
|
||||
With the LDA transform, we can
|
||||
train a topic model to produce 1 million topics with 1 million
|
||||
vocabulary on a 1-billion-token document set one
|
||||
|
|
|
|||
|
|
@ -39,14 +39,14 @@
|
|||
More details about LD-SVM can be found in this paper `Local deep
|
||||
kernel
|
||||
learning for efficient non-linear SVM prediction
|
||||
<http://research.microsoft.com/en-
|
||||
<https://research.microsoft.com/en-
|
||||
us/um/people/manik/pubs/Jose13.pdf>`_.
|
||||
|
||||
|
||||
**Reference**
|
||||
|
||||
`Local deep kernel learning for efficient non-linear SVM prediction
|
||||
<http://research.microsoft.com/en-
|
||||
<https://research.microsoft.com/en-
|
||||
us/um/people/manik/pubs/Jose13.pdf>`_
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -69,14 +69,14 @@
|
|||
|
||||
**Reference**
|
||||
|
||||
`Wikipedia: L-BFGS <http://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
`Wikipedia: L-BFGS <https://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
|
||||
`Wikipedia: Logistic
|
||||
regression <http://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
regression <https://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
|
||||
`Scalable
|
||||
Training of L1-Regularized Log-Linear Models
|
||||
<http://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
<https://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
|
||||
`Test Run - L1
|
||||
and L2 Regularization for Machine Learning
|
||||
|
|
|
|||
|
|
@ -70,14 +70,14 @@
|
|||
|
||||
**Reference**
|
||||
|
||||
`Wikipedia: L-BFGS <http://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
`Wikipedia: L-BFGS <https://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
|
||||
`Wikipedia: Logistic
|
||||
regression <http://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
regression <https://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
|
||||
`Scalable
|
||||
Training of L1-Regularized Log-Linear Models
|
||||
<http://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
<https://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
|
||||
`Test Run - L1
|
||||
and L2 Regularization for Machine Learning
|
||||
|
|
|
|||
|
|
@ -29,10 +29,10 @@
|
|||
us/library/azure/dn913103.aspx>`_
|
||||
|
||||
`Estimating the Support of a High-Dimensional Distribution
|
||||
<http://research.microsoft.com/pubs/69731/tr-99-87.pdf>`_
|
||||
<https://research.microsoft.com/pubs/69731/tr-99-87.pdf>`_
|
||||
|
||||
`New Support Vector Algorithms
|
||||
<http://www.stat.purdue.edu/~yuzhu/stat598m3/Papers/NewSVM.pdf>`_
|
||||
<https://www.stat.purdue.edu/~yuzhu/stat598m3/Papers/NewSVM.pdf>`_
|
||||
|
||||
`LIBSVM: A Library for Support Vector Machines
|
||||
<https://www.csie.ntu.edu.tw/~cjlin/papers/libsvm.pdf>`_
|
||||
|
|
|
|||
|
|
@ -36,7 +36,7 @@
|
|||
|
||||
`Randomized Methods for Computing the Singular Value Decomposition
|
||||
(SVD) of very large matrices
|
||||
<http://web.stanford.edu/group/mmds/slides2010/Martinsson.pdf>`_
|
||||
<https://web.stanford.edu/group/mmds/slides2010/Martinsson.pdf>`_
|
||||
`A randomized algorithm for principal component analysis
|
||||
<https://arxiv.org/abs/0809.2274>`_,
|
||||
`Finding Structure with Randomness: Probabilistic Algorithms for
|
||||
|
|
|
|||
|
|
@ -13,14 +13,14 @@
|
|||
associated optimization problem is sparse, then Hogwild SGD achieves
|
||||
a
|
||||
nearly optimal rate of convergence. For a detailed reference, please
|
||||
refer to `http://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<http://arxiv.org/pdf/1106.5730v2.pdf>`_.
|
||||
refer to `https://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<https://arxiv.org/pdf/1106.5730v2.pdf>`_.
|
||||
|
||||
|
||||
**Reference**
|
||||
|
||||
`http://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<http://arxiv.org/pdf/1106.5730v2.pdf>`_
|
||||
`https://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<https://arxiv.org/pdf/1106.5730v2.pdf>`_
|
||||
|
||||
|
||||
:param normalize: Specifies the type of automatic normalization used:
|
||||
|
|
|
|||
|
|
@ -11,7 +11,7 @@
|
|||
input time-series where each component in the spectrum corresponds to a
|
||||
trend, seasonal or noise component in the time-series. For details of the
|
||||
Singular Spectrum Analysis (SSA), refer to `this document
|
||||
<http://arxiv.org/pdf/1206.6910.pdf>`_.
|
||||
<https://arxiv.org/pdf/1206.6910.pdf>`_.
|
||||
|
||||
.. seealso::
|
||||
:py:func:`IIDChangePointDetector
|
||||
|
|
|
|||
|
|
@ -7,12 +7,12 @@
|
|||
versions of `GloVe Models
|
||||
<https://nlp.stanford.edu/projects/glove/>`_, `FastText
|
||||
<https://en.wikipedia.org/wiki/FastText>`_, and `Sswe
|
||||
<http://anthology.aclweb.org/P/P14/P14-1146.pdf>`_.
|
||||
<https://anthology.aclweb.org/P/P14/P14-1146.pdf>`_.
|
||||
|
||||
.. remarks::
|
||||
Sentiment-specific word embedding (SSWE) is a DNN featurizer
|
||||
developed
|
||||
by MSRA (`paper <http://anthology.aclweb.org/P/P14/P14-1146.pdf>`_).
|
||||
by MSRA (`paper <https://anthology.aclweb.org/P/P14/P14-1146.pdf>`_).
|
||||
It
|
||||
incorporates sentiment information into the neural network to learn
|
||||
sentiment specific word embedding. It proves to be useful in various
|
||||
|
|
|
|||
|
|
@ -24,7 +24,7 @@
|
|||
the default is to normalize features before training.
|
||||
|
||||
``SupervisedBinner`` implements the `Entropy-Based Discretization
|
||||
<http://www.aaai.org/Papers/KDD/1996/KDD96-019.pdf>`_.
|
||||
<https://www.aaai.org/Papers/KDD/1996/KDD96-019.pdf>`_.
|
||||
Partition of the data is performed recursively to select the split
|
||||
with highest entropy gain with respect to the label.
|
||||
Therefore, the final binned features will have high correlation with
|
||||
|
|
|
|||
|
|
@ -10,7 +10,7 @@
|
|||
available options are various versions of `GloVe Models
|
||||
<https://nlp.stanford.edu/projects/glove/>`_, `FastText
|
||||
<https://en.wikipedia.org/wiki/FastText>`_, and `Sswe
|
||||
<http://anthology.aclweb.org/P/P14/P14-1146.pdf>`_.
|
||||
<https://anthology.aclweb.org/P/P14/P14-1146.pdf>`_.
|
||||
|
||||
|
||||
:param model_kind: Pre-trained model used to create the vocabulary.
|
||||
|
|
|
|||
|
|
@ -8432,7 +8432,7 @@ label {
|
|||
padding: 0px;
|
||||
}
|
||||
/* Flexible box model classes */
|
||||
/* Taken from Alex Russell http://infrequently.org/2009/08/css-3-progress/ */
|
||||
/* Taken from Alex Russell https://infrequently.org/2009/08/css-3-progress/ */
|
||||
/* This file is a compatability layer. It allows the usage of flexible box
|
||||
model layouts accross multiple browsers, including older browsers. The newest,
|
||||
universal implementation of the flexible box model is used when available (see
|
||||
|
|
|
|||
|
|
@ -128,8 +128,8 @@ sphinx_gallery_conf = {
|
|||
'relative': True,
|
||||
'reference_url': {
|
||||
'nimbusml': None,
|
||||
'matplotlib': 'http://matplotlib.org',
|
||||
'numpy': 'http://www.numpy.org/',
|
||||
'matplotlib': 'https://matplotlib.org',
|
||||
'numpy': 'https://www.numpy.org/',
|
||||
'scipy': 'https://www.scipy.org/'},
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -145,8 +145,8 @@ sphinx_gallery_conf = {
|
|||
'relative': True,
|
||||
'reference_url': {
|
||||
'nimbusml': None,
|
||||
'matplotlib': 'http://matplotlib.org',
|
||||
'numpy': 'http://www.numpy.org/',
|
||||
'matplotlib': 'https://matplotlib.org',
|
||||
'numpy': 'https://www.numpy.org/',
|
||||
'scipy': 'https://www.scipy.org/'},
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -44,7 +44,7 @@ class FactorizationMachineBinaryClassifier(
|
|||
`Field Aware Factorization Machines
|
||||
<https://www.csie.ntu.edu.tw/~r01922136/slides/ffm.pdf>`_,
|
||||
`Field-aware Factorization Machines for CTR Prediction
|
||||
<http://www.csie.ntu.edu.tw/~cjlin/papers/ffm.pdf>`_,
|
||||
<https://www.csie.ntu.edu.tw/~cjlin/papers/ffm.pdf>`_,
|
||||
`Adaptive Subgradient Methods for Online Learning and Stochastic
|
||||
Optimization
|
||||
<http://jmlr.org/papers/volume12/duchi11a/duchi11a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -57,7 +57,7 @@ class PcaAnomalyDetector(core, BasePredictor, ClassifierMixin):
|
|||
|
||||
`Randomized Methods for Computing the Singular Value Decomposition
|
||||
(SVD) of very large matrices
|
||||
<http://web.stanford.edu/group/mmds/slides2010/Martinsson.pdf>`_
|
||||
<https://web.stanford.edu/group/mmds/slides2010/Martinsson.pdf>`_
|
||||
`A randomized algorithm for principal component analysis
|
||||
<https://arxiv.org/abs/0809.2274>`_,
|
||||
`Finding Structure with Randomness: Probabilistic Algorithms for
|
||||
|
|
|
|||
|
|
@ -55,7 +55,7 @@ class FastForestBinaryClassifier(
|
|||
**Reference**
|
||||
|
||||
`Wikipedia: Random forest
|
||||
<http://en.wikipedia.org/wiki/Random_forest>`_
|
||||
<https://en.wikipedia.org/wiki/Random_forest>`_
|
||||
|
||||
`Quantile regression forest
|
||||
<http://jmlr.org/papers/volume7/meinshausen06a/meinshausen06a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -64,7 +64,7 @@ class FastForestRegressor(core, BasePredictor, RegressorMixin):
|
|||
**Reference**
|
||||
|
||||
`Wikipedia: Random forest
|
||||
<http://en.wikipedia.org/wiki/Random_forest>`_
|
||||
<https://en.wikipedia.org/wiki/Random_forest>`_
|
||||
|
||||
`Quantile regression forest
|
||||
<http://jmlr.org/papers/volume7/meinshausen06a/meinshausen06a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -81,7 +81,7 @@ class FastTreesBinaryClassifier(
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param feature: see `Columns </nimbusml/concepts/columns>`_.
|
||||
|
||||
|
|
|
|||
|
|
@ -83,7 +83,7 @@ class FastTreesRegressor(core, BasePredictor, RegressorMixin):
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param feature: see `Columns </nimbusml/concepts/columns>`_.
|
||||
|
||||
|
|
|
|||
|
|
@ -38,7 +38,7 @@ class FastTreesTweedieRegressor(
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param feature: see `Columns </nimbusml/concepts/columns>`_.
|
||||
|
||||
|
|
|
|||
|
|
@ -42,7 +42,7 @@ class GamBinaryClassifier(core, BasePredictor, ClassifierMixin):
|
|||
functions learned will step between the discretization boundaries.
|
||||
|
||||
This implementation is based on the this `paper
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
but diverges from it in several important respects: most
|
||||
significantly,
|
||||
in each round of boosting, rather than do one feature at a time, it
|
||||
|
|
@ -78,7 +78,7 @@ class GamBinaryClassifier(core, BasePredictor, ClassifierMixin):
|
|||
`Generalized additive models
|
||||
<https://en.wikipedia.org/wiki/Generalized_additive_model>`_,
|
||||
`Intelligible Models for Classification and Regression
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
|
||||
|
||||
:param feature: see `Columns </nimbusml/concepts/columns>`_.
|
||||
|
|
|
|||
|
|
@ -41,7 +41,7 @@ class GamRegressor(core, BasePredictor, RegressorMixin):
|
|||
functions learned will step between the discretization boundaries.
|
||||
|
||||
This implementation is based on the this `paper
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
but diverges from it in several important respects: most
|
||||
significantly,
|
||||
in each round of boosting, rather than do one feature at a time, it
|
||||
|
|
@ -77,7 +77,7 @@ class GamRegressor(core, BasePredictor, RegressorMixin):
|
|||
`Generalized additive models
|
||||
<https://en.wikipedia.org/wiki/Generalized_additive_model>`_,
|
||||
`Intelligible Models for Classification and Regression
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
|
||||
|
||||
:param feature: see `Columns </nimbusml/concepts/columns>`_.
|
||||
|
|
|
|||
|
|
@ -30,7 +30,7 @@ class LightLda(core, BaseTransform, TransformerMixin):
|
|||
topical vectors. LightLDA is an extremely
|
||||
efficient implementation of LDA developed in MSR-Asia that
|
||||
incorporates a number of optimization techniques
|
||||
`(http://arxiv.org/abs/1412.1576) <http://arxiv.org/abs/1412.1576>`_.
|
||||
`(https://arxiv.org/abs/1412.1576) <https://arxiv.org/abs/1412.1576>`_.
|
||||
With the LDA transform, we can
|
||||
train a topic model to produce 1 million topics with 1 million
|
||||
vocabulary on a 1-billion-token document set one
|
||||
|
|
|
|||
|
|
@ -31,7 +31,7 @@ class WordEmbedding(core, BaseTransform, TransformerMixin):
|
|||
available options are various versions of `GloVe Models
|
||||
<https://nlp.stanford.edu/projects/glove/>`_, `FastText
|
||||
<https://en.wikipedia.org/wiki/FastText>`_, and `Sswe
|
||||
<http://anthology.aclweb.org/P/P14/P14-1146.pdf>`_.
|
||||
<https://anthology.aclweb.org/P/P14/P14-1146.pdf>`_.
|
||||
|
||||
|
||||
:param columns: a dictionary of key-value pairs, where key is the output
|
||||
|
|
|
|||
|
|
@ -42,7 +42,7 @@ class FactorizationMachineBinaryClassifier(
|
|||
`Field Aware Factorization Machines
|
||||
<https://www.csie.ntu.edu.tw/~r01922136/slides/ffm.pdf>`_,
|
||||
`Field-aware Factorization Machines for CTR Prediction
|
||||
<http://www.csie.ntu.edu.tw/~cjlin/papers/ffm.pdf>`_,
|
||||
<https://www.csie.ntu.edu.tw/~cjlin/papers/ffm.pdf>`_,
|
||||
`Adaptive Subgradient Methods for Online Learning and Stochastic
|
||||
Optimization
|
||||
<http://jmlr.org/papers/volume12/duchi11a/duchi11a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -57,7 +57,7 @@ class PcaAnomalyDetector(
|
|||
|
||||
`Randomized Methods for Computing the Singular Value Decomposition
|
||||
(SVD) of very large matrices
|
||||
<http://web.stanford.edu/group/mmds/slides2010/Martinsson.pdf>`_
|
||||
<https://web.stanford.edu/group/mmds/slides2010/Martinsson.pdf>`_
|
||||
`A randomized algorithm for principal component analysis
|
||||
<https://arxiv.org/abs/0809.2274>`_,
|
||||
`Finding Structure with Randomness: Probabilistic Algorithms for
|
||||
|
|
|
|||
|
|
@ -54,7 +54,7 @@ class FastForestBinaryClassifier(
|
|||
**Reference**
|
||||
|
||||
`Wikipedia: Random forest
|
||||
<http://en.wikipedia.org/wiki/Random_forest>`_
|
||||
<https://en.wikipedia.org/wiki/Random_forest>`_
|
||||
|
||||
`Quantile regression forest
|
||||
<http://jmlr.org/papers/volume7/meinshausen06a/meinshausen06a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -64,7 +64,7 @@ class FastForestRegressor(
|
|||
**Reference**
|
||||
|
||||
`Wikipedia: Random forest
|
||||
<http://en.wikipedia.org/wiki/Random_forest>`_
|
||||
<https://en.wikipedia.org/wiki/Random_forest>`_
|
||||
|
||||
`Quantile regression forest
|
||||
<http://jmlr.org/papers/volume7/meinshausen06a/meinshausen06a.pdf>`_
|
||||
|
|
|
|||
|
|
@ -78,7 +78,7 @@ class FastTreesBinaryClassifier(
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param number_of_trees: Specifies the total number of decision trees to
|
||||
create in the ensemble. By creating more decision trees, you can
|
||||
|
|
|
|||
|
|
@ -83,7 +83,7 @@ class FastTreesRegressor(
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param number_of_trees: Specifies the total number of decision trees to
|
||||
create in the ensemble. By creating more decision trees, you can
|
||||
|
|
|
|||
|
|
@ -35,7 +35,7 @@ class FastTreesTweedieRegressor(
|
|||
<https://en.wikipedia.org/wiki/Gradient_boosting#Gradient_tree_boosting>`_
|
||||
|
||||
`Greedy function approximation: A gradient boosting machine.
|
||||
<http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
<https://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1013203451>`_
|
||||
|
||||
:param number_of_trees: Specifies the total number of decision trees to
|
||||
create in the ensemble. By creating more decision trees, you can
|
||||
|
|
|
|||
|
|
@ -42,7 +42,7 @@ class GamBinaryClassifier(
|
|||
functions learned will step between the discretization boundaries.
|
||||
|
||||
This implementation is based on the this `paper
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
but diverges from it in several important respects: most
|
||||
significantly,
|
||||
in each round of boosting, rather than do one feature at a time, it
|
||||
|
|
@ -78,7 +78,7 @@ class GamBinaryClassifier(
|
|||
`Generalized additive models
|
||||
<https://en.wikipedia.org/wiki/Generalized_additive_model>`_,
|
||||
`Intelligible Models for Classification and Regression
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
|
||||
|
||||
:param number_of_iterations: Total number of iterations over all features.
|
||||
|
|
|
|||
|
|
@ -40,7 +40,7 @@ class GamRegressor(BasePipelineItem, DefaultSignatureWithRoles):
|
|||
functions learned will step between the discretization boundaries.
|
||||
|
||||
This implementation is based on the this `paper
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_,
|
||||
but diverges from it in several important respects: most
|
||||
significantly,
|
||||
in each round of boosting, rather than do one feature at a time, it
|
||||
|
|
@ -76,7 +76,7 @@ class GamRegressor(BasePipelineItem, DefaultSignatureWithRoles):
|
|||
`Generalized additive models
|
||||
<https://en.wikipedia.org/wiki/Generalized_additive_model>`_,
|
||||
`Intelligible Models for Classification and Regression
|
||||
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
<https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.352.7619>`_
|
||||
|
||||
|
||||
:param number_of_iterations: Total number of iterations over all features.
|
||||
|
|
|
|||
|
|
@ -28,7 +28,7 @@ class LightLda(BasePipelineItem, DefaultSignature):
|
|||
topical vectors. LightLDA is an extremely
|
||||
efficient implementation of LDA developed in MSR-Asia that
|
||||
incorporates a number of optimization techniques
|
||||
`(http://arxiv.org/abs/1412.1576) <http://arxiv.org/abs/1412.1576>`_.
|
||||
`(https://arxiv.org/abs/1412.1576) <https://arxiv.org/abs/1412.1576>`_.
|
||||
With the LDA transform, we can
|
||||
train a topic model to produce 1 million topics with 1 million
|
||||
vocabulary on a 1-billion-token document set one
|
||||
|
|
|
|||
|
|
@ -28,7 +28,7 @@ class WordEmbedding(BasePipelineItem, DefaultSignature):
|
|||
available options are various versions of `GloVe Models
|
||||
<https://nlp.stanford.edu/projects/glove/>`_, `FastText
|
||||
<https://en.wikipedia.org/wiki/FastText>`_, and `Sswe
|
||||
<http://anthology.aclweb.org/P/P14/P14-1146.pdf>`_.
|
||||
<https://anthology.aclweb.org/P/P14/P14-1146.pdf>`_.
|
||||
|
||||
|
||||
:param model_kind: Pre-trained model used to create the vocabulary.
|
||||
|
|
|
|||
|
|
@ -67,10 +67,10 @@ class AveragedPerceptronBinaryClassifier(
|
|||
<https://en.wikipedia.org/wiki/Perceptron>`_
|
||||
|
||||
`Large Margin Classification Using the Perceptron Algorithm
|
||||
<http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.48.8200>`_
|
||||
<https://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.48.8200>`_
|
||||
|
||||
`Discriminative Training Methods for Hidden Markov Models
|
||||
<http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.6725>`_
|
||||
<https://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.6725>`_
|
||||
|
||||
|
||||
:param normalize: Specifies the type of automatic normalization used:
|
||||
|
|
|
|||
|
|
@ -90,14 +90,14 @@ class LogisticRegressionBinaryClassifier(
|
|||
|
||||
**Reference**
|
||||
|
||||
`Wikipedia: L-BFGS <http://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
`Wikipedia: L-BFGS <https://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
|
||||
`Wikipedia: Logistic
|
||||
regression <http://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
regression <https://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
|
||||
`Scalable
|
||||
Training of L1-Regularized Log-Linear Models
|
||||
<http://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
<https://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
|
||||
`Test Run - L1
|
||||
and L2 Regularization for Machine Learning
|
||||
|
|
|
|||
|
|
@ -91,14 +91,14 @@ class LogisticRegressionClassifier(
|
|||
|
||||
**Reference**
|
||||
|
||||
`Wikipedia: L-BFGS <http://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
`Wikipedia: L-BFGS <https://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
|
||||
`Wikipedia: Logistic
|
||||
regression <http://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
regression <https://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
|
||||
`Scalable
|
||||
Training of L1-Regularized Log-Linear Models
|
||||
<http://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
<https://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
|
||||
`Test Run - L1
|
||||
and L2 Regularization for Machine Learning
|
||||
|
|
|
|||
|
|
@ -35,14 +35,14 @@ class SgdBinaryClassifier(
|
|||
associated optimization problem is sparse, then Hogwild SGD achieves
|
||||
a
|
||||
nearly optimal rate of convergence. For a detailed reference, please
|
||||
refer to `http://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<http://arxiv.org/pdf/1106.5730v2.pdf>`_.
|
||||
refer to `https://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<https://arxiv.org/pdf/1106.5730v2.pdf>`_.
|
||||
|
||||
|
||||
**Reference**
|
||||
|
||||
`http://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<http://arxiv.org/pdf/1106.5730v2.pdf>`_
|
||||
`https://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<https://arxiv.org/pdf/1106.5730v2.pdf>`_
|
||||
|
||||
|
||||
:param normalize: Specifies the type of automatic normalization used:
|
||||
|
|
|
|||
|
|
@ -30,7 +30,7 @@ class SsaForecaster(BasePipelineItem, DefaultSignature):
|
|||
input time-series where each component in the spectrum corresponds to a
|
||||
trend, seasonal or noise component in the time-series. For details of the
|
||||
Singular Spectrum Analysis (SSA), refer to `this document
|
||||
<http://arxiv.org/pdf/1206.6910.pdf>`_.
|
||||
<https://arxiv.org/pdf/1206.6910.pdf>`_.
|
||||
|
||||
:param window_size: The length of the window on the series for building the
|
||||
trajectory matrix (parameter L).
|
||||
|
|
|
|||
|
|
@ -67,10 +67,10 @@ class AveragedPerceptronBinaryClassifier(
|
|||
<https://en.wikipedia.org/wiki/Perceptron>`_
|
||||
|
||||
`Large Margin Classification Using the Perceptron Algorithm
|
||||
<http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.48.8200>`_
|
||||
<https://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.48.8200>`_
|
||||
|
||||
`Discriminative Training Methods for Hidden Markov Models
|
||||
<http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.6725>`_
|
||||
<https://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.6725>`_
|
||||
|
||||
|
||||
:param feature: see `Columns </nimbusml/concepts/columns>`_.
|
||||
|
|
|
|||
|
|
@ -91,14 +91,14 @@ class LogisticRegressionBinaryClassifier(
|
|||
|
||||
**Reference**
|
||||
|
||||
`Wikipedia: L-BFGS <http://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
`Wikipedia: L-BFGS <https://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
|
||||
`Wikipedia: Logistic
|
||||
regression <http://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
regression <https://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
|
||||
`Scalable
|
||||
Training of L1-Regularized Log-Linear Models
|
||||
<http://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
<https://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
|
||||
`Test Run - L1
|
||||
and L2 Regularization for Machine Learning
|
||||
|
|
|
|||
|
|
@ -92,14 +92,14 @@ class LogisticRegressionClassifier(
|
|||
|
||||
**Reference**
|
||||
|
||||
`Wikipedia: L-BFGS <http://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
`Wikipedia: L-BFGS <https://en.wikipedia.org/wiki/L-BFGS>`_
|
||||
|
||||
`Wikipedia: Logistic
|
||||
regression <http://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
regression <https://en.wikipedia.org/wiki/Logistic_regression>`_
|
||||
|
||||
`Scalable
|
||||
Training of L1-Regularized Log-Linear Models
|
||||
<http://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
<https://research.microsoft.com/apps/pubs/default.aspx?id=78900>`_
|
||||
|
||||
`Test Run - L1
|
||||
and L2 Regularization for Machine Learning
|
||||
|
|
|
|||
|
|
@ -34,14 +34,14 @@ class SgdBinaryClassifier(core, BasePredictor, ClassifierMixin):
|
|||
associated optimization problem is sparse, then Hogwild SGD achieves
|
||||
a
|
||||
nearly optimal rate of convergence. For a detailed reference, please
|
||||
refer to `http://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<http://arxiv.org/pdf/1106.5730v2.pdf>`_.
|
||||
refer to `https://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<https://arxiv.org/pdf/1106.5730v2.pdf>`_.
|
||||
|
||||
|
||||
**Reference**
|
||||
|
||||
`http://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<http://arxiv.org/pdf/1106.5730v2.pdf>`_
|
||||
`https://arxiv.org/pdf/1106.5730v2.pdf
|
||||
<https://arxiv.org/pdf/1106.5730v2.pdf>`_
|
||||
|
||||
|
||||
:param feature: see `Columns </nimbusml/concepts/columns>`_.
|
||||
|
|
|
|||
|
|
@ -31,7 +31,7 @@ class SsaForecaster(core, BaseTransform, TransformerMixin):
|
|||
input time-series where each component in the spectrum corresponds to a
|
||||
trend, seasonal or noise component in the time-series. For details of the
|
||||
Singular Spectrum Analysis (SSA), refer to `this document
|
||||
<http://arxiv.org/pdf/1206.6910.pdf>`_.
|
||||
<https://arxiv.org/pdf/1206.6910.pdf>`_.
|
||||
|
||||
:param columns: see `Columns </nimbusml/concepts/columns>`_.
|
||||
|
||||
|
|
|
|||
|
|
@ -148,7 +148,7 @@ setup(
|
|||
|
||||
# Although 'package_data' is the preferred approach, in some case
|
||||
# you may need to place data files outside of your packages. See:
|
||||
# http://docs.python.org/3.4/distutils/setupscript.html#installing
|
||||
# https://docs.python.org/3.4/distutils/setupscript.html#installing
|
||||
# -additional-files # noqa
|
||||
# In this case, 'data_file' will be installed into
|
||||
# '<sys.prefix>/my_data'
|
||||
|
|
|
|||
|
|
@ -148,7 +148,7 @@ setup(
|
|||
|
||||
# Although 'package_data' is the preferred approach, in some case
|
||||
# you may need to place data files outside of your packages. See:
|
||||
# http://docs.python.org/3.4/distutils/setupscript.html#installing
|
||||
# https://docs.python.org/3.4/distutils/setupscript.html#installing
|
||||
# -additional-files # noqa
|
||||
# In this case, 'data_file' will be installed into
|
||||
# '<sys.prefix>/my_data'
|
||||
|
|
|
|||
Загрузка…
Ссылка в новой задаче