From c3b7b0c49f2e377f2851cb17cc7870e4b9698b24 Mon Sep 17 00:00:00 2001 From: Amit Sharma Date: Thu, 5 Nov 2020 22:09:46 +0530 Subject: [PATCH] more updates to notebook --- ...e-machinelearning-using-dowhy-econml.ipynb | 454 ++---------------- 1 file changed, 47 insertions(+), 407 deletions(-) diff --git a/docs/source/example_notebooks/tutorial-causalinference-machinelearning-using-dowhy-econml.ipynb b/docs/source/example_notebooks/tutorial-causalinference-machinelearning-using-dowhy-econml.ipynb index 3b0ad58f4..29195fd01 100644 --- a/docs/source/example_notebooks/tutorial-causalinference-machinelearning-using-dowhy-econml.ipynb +++ b/docs/source/example_notebooks/tutorial-causalinference-machinelearning-using-dowhy-econml.ipynb @@ -96,7 +96,7 @@ "\n", "Since there is no ground-truth test dataset available that an estimate can be compared to, causal inference requires a series of principled steps to achieve a good estimator. \n", "\n", - "Let us illustrate the four steps through a sample dataset." + "Let us illustrate the four steps through a sample dataset. This tutorial requires you to download two libraries: DoWhy and EconML. Both can be installed by the following command: `pip install dowhy econml`." ] }, { @@ -105,8 +105,11 @@ "metadata": {}, "outputs": [], "source": [ + "import dowhy\n", "from dowhy import CausalModel\n", "import dowhy.datasets\n", + "import logging\n", + "logging.getLogger(\"dowhy\").setLevel(logging.WARNING)\n", "\n", "# Load some sample data\n", "data = dowhy.datasets.linear_dataset(\n", @@ -139,9 +142,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "WARNING:dowhy.causal_model:Causal Graph not provided. DoWhy will construct a graph based on data inputs.\n", - "INFO:dowhy.causal_graph:If this is observed data (not from a randomized experiment), there might always be missing confounders. Adding a node named \"Unobserved Confounders\" to reflect this.\n", - "INFO:dowhy.causal_model:Model to find the causal effect of treatment ['v0'] on outcome ['y']\n" + "WARNING:dowhy.causal_model:Causal Graph not provided. DoWhy will construct a graph based on data inputs.\n" ] } ], @@ -195,15 +196,7 @@ "cell_type": "code", "execution_count": 4, "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:dowhy.causal_model:Model to find the causal effect of treatment ['v0'] on outcome ['y']\n" - ] - } - ], + "outputs": [], "source": [ "# I. Create a causal model from the data and given graph.\n", "model = CausalModel(\n", @@ -248,37 +241,23 @@ "name": "stdout", "output_type": "stream", "text": [ - "WARN: Do you want to continue by ignoring any unobserved confounders? (use proceed_when_unidentifiable=True to disable this prompt) [y/n] y\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:dowhy.causal_identifier:Instrumental variables for treatment and outcome:['Z1', 'Z0']\n", - "INFO:dowhy.causal_identifier:Frontdoor variables for treatment and outcome:[]\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ + "WARN: Do you want to continue by ignoring any unobserved confounders? (use proceed_when_unidentifiable=True to disable this prompt) [y/n] y\n", "Estimand type: nonparametric-ate\n", "\n", "### Estimand : 1\n", "Estimand name: backdoor1 (Default)\n", "Estimand expression:\n", " d \n", - "─────(Expectation(y|W1,W2,W4,W3,W0))\n", + "─────(Expectation(y|W1,W4,W3,W0,W2))\n", "d[v₀] \n", - "Estimand assumption 1, Unconfoundedness: If U→{v0} and U→y then P(y|v0,W1,W2,W4,W3,W0,U) = P(y|v0,W1,W2,W4,W3,W0)\n", + "Estimand assumption 1, Unconfoundedness: If U→{v0} and U→y then P(y|v0,W1,W4,W3,W0,W2,U) = P(y|v0,W1,W4,W3,W0,W2)\n", "\n", "### Estimand : 2\n", "Estimand name: iv\n", "Estimand expression:\n", - "Expectation(Derivative(y, [Z1, Z0])*Derivative([v0], [Z1, Z0])**(-1))\n", - "Estimand assumption 1, As-if-random: If U→→y then ¬(U →→{Z1,Z0})\n", - "Estimand assumption 2, Exclusion: If we remove {Z1,Z0}→{v0}, then ¬({Z1,Z0}→y)\n", + "Expectation(Derivative(y, [Z0, Z1])*Derivative([v0], [Z0, Z1])**(-1))\n", + "Estimand assumption 1, As-if-random: If U→→y then ¬(U →→{Z0,Z1})\n", + "Estimand assumption 2, Exclusion: If we remove {Z0,Z1}→{v0}, then ¬({Z0,Z1}→y)\n", "\n", "### Estimand : 3\n", "Estimand name: frontdoor\n", @@ -313,8 +292,6 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~v0+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n" ] @@ -329,11 +306,11 @@ "Estimand type: nonparametric-ate\n", "\n", "## Realized estimand\n", - "b: y~v0+W1+W2+W4+W3+W0\n", + "b: y~v0+W1+W4+W3+W0+W2\n", "Target units: ate\n", "\n", "## Estimate\n", - "Mean value: 10.764548564969608\n", + "Mean value: 9.97678676065801\n", "\n" ] } @@ -355,12 +332,10 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:dowhy.causal_estimator:INFO: Using EconML Estimator\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/deprecation.py:143: FutureWarning: The sklearn.ensemble.forest module is deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.ensemble. Anything that cannot be imported from sklearn.ensemble is now part of the private API.\n", " warnings.warn(message, FutureWarning)\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/deprecation.py:143: FutureWarning: The sklearn.ensemble.base module is deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.ensemble. Anything that cannot be imported from sklearn.ensemble is now part of the private API.\n", " warnings.warn(message, FutureWarning)\n", - "INFO:dowhy.causal_estimator:b: y~v0+W1+W2+W4+W3+W0 | \n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", @@ -377,11 +352,11 @@ "Estimand type: nonparametric-ate\n", "\n", "## Realized estimand\n", - "b: y~v0+W1+W2+W4+W3+W0 | \n", + "b: y~v0+W1+W4+W3+W0+W2 | \n", "Target units: ate\n", "\n", "## Estimate\n", - "Mean value: 8.860891948045317\n", + "Mean value: 9.929599084950528\n", "\n" ] }, @@ -427,166 +402,68 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Refutation over 100 simulated datasets of Random Data treatment\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", + "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", + " return f(**kwargs)\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n" ] @@ -595,165 +472,68 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", + "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", + " return f(**kwargs)\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n" ] @@ -762,165 +542,68 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", + "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", + " return f(**kwargs)\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n" ] @@ -929,43 +612,14 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", - " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", - "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", - " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", - "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", - " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Using a Binomial Distribution with 1 trials and 0.5 probability of success\n", - "INFO:dowhy.causal_estimator:INFO: Using Propensity Score Stratification Estimator\n", - "INFO:dowhy.causal_estimator:b: y~placebo+W1+W2+W4+W3+W0\n", - "/home/amit/py-envs/env3.8/lib/python3.8/site-packages/sklearn/utils/validation.py:72: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", - " return f(**kwargs)\n", - "INFO:dowhy.causal_refuters.placebo_treatment_refuter:Making use of Bootstrap as we have more than 100 examples.\n", - " Note: The greater the number of examples, the more accurate are the confidence estimates\n" + " return f(**kwargs)\n" ] }, { @@ -973,9 +627,9 @@ "output_type": "stream", "text": [ "Refute: Use a Placebo Treatment\n", - "Estimated effect:10.764548564969608\n", - "New effect:0.00280382889295522\n", - "p value:0.49\n", + "Estimated effect:9.97678676065801\n", + "New effect:0.010172363182862056\n", + "p value:0.44\n", "\n" ] } @@ -1019,11 +673,7 @@ "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import math\n", - "import dowhy\n", - "from dowhy import CausalModel\n", - "import dowhy.datasets, dowhy.plotter\n", - "import logging\n", - "logging.getLogger(\"dowhy\").setLevel(logging.WARNING)" + "import dowhy.datasets, dowhy.plotter" ] }, { @@ -1043,16 +693,16 @@ "output_type": "stream", "text": [ " Action Outcome w0\n", - "0 6.598330 13.831028 -0.661638\n", - "1 16.792985 30.529184 3.343500\n", - "2 20.908634 38.745537 -3.914436\n", - "3 8.751380 13.944940 -2.428403\n", - "4 19.016263 37.987295 3.915555\n" + "0 11.436948 23.957157 2.428705\n", + "1 11.164825 24.498600 -2.601900\n", + "2 19.483392 39.550436 -3.690288\n", + "3 6.031888 13.768366 -0.999223\n", + "4 20.551930 41.252494 -3.991802\n" ] }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf0AAAGTCAYAAAAiMsuiAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAB5sUlEQVR4nO2deZxU1bXvf6uqu6FxAgQ0Igh4FRMcmkHFeK8m8UYTiYYYARWimJuYiAnqTYhNri+StHnxhQxeE6Mmol7BENQYlMHgFMUgmqDdQNprq4BR6RJlaMamq7pqvz92na5Tp8+wT9UZq9b386lPV9WZdtfZZ6+91l4DCSHAMAzDMEzlkwi7AQzDMAzDBAMLfYZhGIapEljoMwzDMEyVwEKfYRiGYaoEFvoMwzAMUyWw0GcYhmGYKqEm7AaUw6BBg8SIESPCbgbDMAzDBMarr766XQgxuJRjYy30R4wYgXXr1oXdDIZhGIYJDCL6Z6nHsnmfYRiGYaoEFvoMwzAMUyWw0GcYhmGYKoGFPsMwDMNUCSz0GYZhGKZKYKHPMAzDMFUCC32GYRiGqRJY6DMMwzBMlcBCn2EYhmGqBBb6DMMwDOMHqRRw7rnABx+E3ZIeWOgzDMMwjB80NQF//av8GxFICBF2G0pmwoQJgnPvM9WMEAJ79+7Fnj17cODAAWSzWVfHJ5NJ9OvXD4cffjgOO+wwEJFPLWWYKiOVAkaNAg4eBOrrgc2bgaOP9uTURPSqEGJCKcfGuuAOw1QzQgh8+OGH2L9/PwYOHIijjz4ayWRSWXALIZDNZrFv3z5s374dnZ2dGDJkCAt+hvGCpiYgl5Pvs1n5+c47w20T2LzPMLFl79692L9/P4477jj0798fNTU1rgQ2EaGmpgb9+/fHcccdh/3792Pv3r0+tphhqoRUCrj/fiCdlp/Tafk5Amv7LPT1RNDpgmGs2LNnDwYOHIhkMln2uZLJJAYOHIg9e/Z40DKGqXL0Wr6Gpu2HDAt9PY2NwOrV8i/DRJwDBw7g0EMP9ex8hx56KA4cOODZ+Rimalm7tqDla6TTwEsvhdMeHSz0NVIp4KGH5PtFi1jbZyJPNpv1RMvXSCaTrh0BGYYxobkZuPZaoK5Ofq6rA2bNkt+HDAt9jcZGaX4B5F/W9pkY4KXTHTvwMYxH8Jp+xNFr+Rqs7TMMwzClwGv6EUev5Wuwts8wDMOUAq/pR5wVK9x9zzAMwzBWNDcDQgBXXik/X3WV/Mxr+hFh2DDz7486Kth2MAzDMJVBRJ3DWegDhVmZ0dvy3HPDbRfDMAwTTyLqHM5CXyPC3pYMwzBMjIiwczgLfY2mJnNnvgh4WzIMwzAxIsLO4Sz0NdauBTKZ4u8i4m3JMAzDxIhly8y/f+KJYNthAgt9jfvv7/1dfT3w5JPBt4VhFPGyNHacy2wzTCxIhC9yw29BVJgxo/d3bN5nIozXaXO9TuvLMFWLlXC3ihQLkJKFPhFdQES/J6ItRNRJRDuIaAMR/ZaIPmdz3GFE9IP8vnuJqIOI/kZENxBRbantKYtUCnj99d7fs3mfiTD9+vXDvn37PDvfvn370K9fP8/OxzBVSSoF7N9f/F19vfw+jnH6RHQIEf0RwJ8BXA5gBIC+AAYCOAXA1wGYeisQ0UgALQB+mN/3UABHADgdwC8BrCWiAW7bVDZNTUCtYb4RoQIJDGPG4Ycfjp07d3qi7WezWezcuROHH364By1jmComwil4AZdCn4jqACwHcAmANICfATgDwGAAxwD4dwB3AdhhcewyAKMAdAL4NoBjAYwEMA9ADsB4AEtK+k/KIcIpExnGisMOOwyHHHII/vnPf6KjowPd3d2u1uWFEOju7kZHRwf++c9/4pBDDsFhhx3mY4sZpgqIuDypcbl/I4BPATgA4AIhxF8N21MAnrU49hoAY/LvrxJCPKLb9kMi6gTw/wB8logmCSGCy4Gr1+ZTKeCyy4AlS4Cjjw6sCQzjFiLCkCFDsHfvXuzZswcffviha60/mUyiX79+GDRoEA477DCutMcw5dLcLNPuPvhg4buZM82dxUOAVDUDIjoSwPuQpvy5QojbXF2IqBXAJwA0CyHGmWyvBbAV0mqwQgjxBadzTpgwQaxbt85NM+xJpYDx42UChWuvBe6807tzMwzDMJVPKiUd9vQT8GQSeP99zxRJInpVCDGhlGPdmPe/Ainw0wDudnMRIhoFKfAB4FGzfYQQGQCP5z+eR0TBexQ1NsobJgRw993Ahg2BN4FhGIaJMVaJeebODac9BtwIfc0j/+9CiA7tSyJKkrNNUK/Zv2Kzn7atLwqThGAwpk3M5YArrgi0CQzDMEzMsarO+vjj5t8HjBuhPz7/93UiqiOi7xNRG4AuABkiepOI/h8RDTI5drTu/Waba2yxOMZ/zGZnra2s7TMMwzDqWMXiRyQHhpLQJ6J6AJowTwN4EcCPAZwIIJl/nQDgewA2ENFphlPoJwIf2VzqQ937I1Xa5glmxRE0WNtnGIZhVGluBq68svf3+/fHquDOEbr3X4cM0/sjgFMB9AEwHMCPIMPuPgbgcSLSx/4cont/0OY6nbr3h5rtQETXENE6Ilr30Ud28wcXmBXb0Xj99UjcKIZhGCYGWCmRmUwkYvVVhb5+vzoATwCYIoTYKIRICyHeE0LcAqnpA8BxkCF6niOE+K0QYoIQYsLgwYO9OenatdbbamoicaMYhmGYGGClRHZ3RyJWX1XoG3IKYp4wj/X7FQqJeb5ocXxfm+vU6957l1/UieZmoL0dMPNHzGQicaMYhmGYGGCnREaggJuq0N8LuZYPSGHcYraTECKNgge+3vt+u+69nXqu39Yrq5+vGFPxak4XXGmPYRiGUWXlSuttEQjbUxL6QogcgDfzHzsstHyNXfm/+iTebbr3I22O1W9rs9zLa1IpmS1JnzpRM89EKGcywzAME3EaTUvPSJYtC64dFrgJ2dNS3w1wiMvXvO536757Tff+TJtjtW0HAZiUvfMJswIJGum0nBCwMx/DMAxjRyoFLFpkvT0CHvxuhL6WWeAQAKbp/4ioDwqCu0X7XgixGQUhfqnFsTUo+AE8K4Q44KJt5WFWIEEPa/sMwzCME42N1gokIOVMyLLEjdBfiULynCYiMjt2DgCtNO4fDNvuyv8dR0RfNjn2BgBDDPsGQ3Mz8NRT1tsjVCGJYRiGiShO5vtcLnRZoiz08056NwIQAC4AsJyIJhLRACI6iYh+BhmrDwCvAnjQcIrfAmjNv19IRLOI6BgiOo6IfgBAK+DzdKAV9jS+bDYPATB1qszFr6/ExzAMwzBGEg4idcCA0GWJcpW9ngOIZgP4OazL8rYA+IIQYqvJsSMBPANglMWxrwL4rBBil8X2IjyrstfSAowda76tTx/goF0+IYZhGIaBlCMtLdbbiezN/4oEVWUPACCEuANy3f5BAO9ChvJ1QKbmvQ7AmWYCP3/sFgANAG4BsBEyfn8PpJPgfwI4S1Xge8q0adbb0unQHS8YhmGYiNPSAmzZAoyy0mkhS7aHjJW2bosQ4jUAV5V47F7IZYAfOe0bGG+/bb2NSDpe3HlncO1hGIZh4sWMGcDu3fYm/gj4hrnW9CsSfVIeI7kc8MILwbWFYRiGiRctLbIqKwDssjBWz5wZ+no+wEJfYpd2gAg499zg2sIwDMPEixkznPdZuDASS8Us9AGgsxMYM8Z8mxDAffdF4mYxDMMwEUOv5dsRkXwvLPQ1xo+33tbVFYmbxTAMw0QMFS1fg9f0I4JV/WMNIXhdn2EYhunNpk1q+x1/PK/pR4bGRvP6x3pOPz2YtjAMwzDxobNTKoZCyBLtWoVWI9u3m38fMCz0AWCFQgLAiDhhMAzDMBGlqclagdy9G9iwIdj2mMBCHwCOPNJ5n4g4YTAMwzARZe1a++1TpwbTDhtY6AP2cfp6IuCEwTAMw0SU5mb7rHt2ieACoqSMfBWHkyPGmDHAP/4RTFsYhmGYeJJKAQsWWG8/5ZTg2mIBa/qAdMQ44gjr7Zych2EYhnGiqUnWazEyc2ZkqrWy0Afk7GzPHuvtHK7HMAzDOPHss+bfP/54sO2wgYU+IGdndiWGJ5RUwZBhGIapJqz8w3bvjkz0Fwt9wNnjcunSQJrBMAzDxJRUCnj9dfNtuVxkor9Y6ANyncUq9z4AHDgQXFsYhmGY+OEk1CNSw4WFPmA/QwOATCa4tjDlk0pJ58sIPGAMw1QJq1fbLxOn05HQ9lnoA/JGOMXqswCJD01NwF//GokHjGGYKsGuaBsgTfwRyPXCQh+Qa/pmYRZ6GhuDaQtTHqkUcP/98gG7/36erDEMEwzLltlvHzSIQ/YiQ3Mz0NBgv8+iRSxA4kBTkxT4AKdOZhgmOhx7bNgtAMBCv8DKlfbbs1nW9qOOpuVrVpt0mrV9hmH8xynXCxHw5JPBtccGFvoaTU1AwuHnUKnGx4SHXsvXYG2fYRi/sauuB0gHv4iMQyz0Ndau7S0wjOzfz1pjlDHzzUinI+E8wzBMBeOU6wWIzDjEQl+judnZyYK1xmjT3Cxn1MZXBJxnGIapYJqbgRNPtN+HzfsRZNo0++2sNTIMExScbyJevPWW9bZEIjIKIwt9jZYW4M037feZOZO1xkqCB1Umymj5JhobuZ9GnVTKPjFPLscZ+SKHk5YPOMdhMvGCk/gwUUObiK5fX8g3sWgR8OKL3E+jjEpkF2fkixibNjnvw458lQMn8WGiiDYRnT69ON+EENxPo4xKZBdn5IsQqZRzGl4A6OqKxEyN8QBO4sNEDf1EtLW1dyQK99NokkoB+/Y57zdrViSWh1noA+bx3WYIAbzwgv/tYfzFLIlPRNbbmCrGaRxKp4G77wY2bAiuTYwzTU1qRdkioOUDLPQlKrn3Nc4919+2MP5jNrhGZL2NqVKME1ErcjngiiuCaROjhkqOl0QiElo+wEJfosV3X3ut876s6ccPo5e+2SQvl+N7y4SHqrURkGXA2SoVHZxSuAPy3kbknrHQ10ilpInXjkSCNf04oPeAPussWUxJ7/2sTfKuvFLmxAaAujq+t0x4WFkbGxpkXx0zpvBdbS1bpaKE6r2IyD0jYRdbGHEmTJgg1q1b583JZs2S62VOv0d9PbB5M3D00d5cl/GeWbOAe+4BPv5x6RClob93LS3A2LHFx/G9ZaII99Voc/LJxeOMFQ0Nnpn4iehVIcSEUo5lTV9j7VpngQ+wB23UMJruNYuN5gGtR3/vzPIy8L1lgkYlQdSMGb2/474aHc45x377kUdGKh04C32N5mZ5c5zgVLzRwphgx86TVvPSX7/ePPsi31smKDRhP3eufYKoVEqu4RvhvhoNVJaF9+2LzHo+wEK/QColk+84EZFYSwa9E+ysX1/Q8q1Ip821/Pp6eT6+t4yfpFLSz+SUU4DVq2W2PWOCKL3239RknkNkzBjuq1FAJVwvYvldWOhrONVD1nj2Wf/bwqhhTLAzfbrzA5jLAW1tvb9ncynjJS0tQP/+vWPqGxuBl18GduyQn7UxR9//9NYrKwc/9uCPBirhekCkrDIs9DXWrlVLsMBEA82spk+w8/rr6mFPRthcynjJjBnA7t3FMfWpFPDQQ+b7a0tP48cXrFX33y/LsWoloq+9VkaZAOzBHxVUSuoCkSmrC7DQL7ByJdC3r/N+bW08w44CZmY1IhlWWQpOnrVckY9RpaWl4ETa2lrQ9hsb7a2J6TTw2mvSHAwA3d3AuHGyz5llkeRc/OGTSjlXZwUiNUFjoa/hJjlGhG5g1bJ6de/7lcup3cO6OuBjHyuUw2xuBrZsKTbFplLAxIly/VVbW+WKfIwKRm/7K66w1/I1jH03k5HHNTaaj0+8JBU+qr9/hKyILPQ13KTi5cxt4XPOOQVTZ12ddLBUDYtJp+Vgqg/f270bmDKlsE9TE/DKK3L9tbGRK/Ixaui1fI3WVuAb31DzGTJj0SI5yTWOT+k0j0Vhs3at8z4Rc7pkoa+hZWlrby9kaTOjpoYzt4WNnanTzDNfDxGQTMr3998PPPNMwTz35pvAc8/1DsPRPKwB1q4Ye6z6n0rpVSuyWTnm6DPzDRjAGUKjwMqVQJ8+9vucfnowbVGEhb6RpiYp2K3o7o6UqaYqsTJ1NjY6r68JUewxfcEFxdsvuaS3v0A2az7B4HV+xsimTebfl5v59O67gYcfLlgRdu1iy1MUUAnZW7YsmLYowkLfiIoXf4Q8MasSK1PnokXuzpNO95487N4N3HuvvW/AwYOFddYXXyw4WzHVTUtLsQlfv+y0dStw5pmlO5rmcsBXvtL7e7Y8hYtKyN6BA5EaH1joG2luLi5uYQY/ZOFyzjly8Jw1S4YxJRIybKbUNVMjTpM+IYAnnpBalhAFZyumcijFimM07eutQpqPSKkhpdr5zL5jbT88mpuBSy+13ydiEzMW+mY45VJm83546LPw3XdfIaZZJWymXMaMKYR17tkjl3o0Fi3igbeScBut0dJi3ge7u+WE0ClVqxEi2df1sflWREyoVBWpFPDoo/b7RGxixkLfiEou5Qh5YlYd+vX8dLoQ0+w1V11VSIqiOXju3FnsD6AX+ppPARN/jOmdVQZrs6I4gLQaLV+uHhmkIUQhasTpWE4sFR6zZ6vtF6GJGQt9I01N9g+ZnWc/4y9Gr/1yTKVOrFhRbOJtbJSf7Uz/rO1XBsb0zk6DtVVRHI29e0tz5Fu0yH7JKpkszjXBBIuKlq8RoYkZC30jTiV2heCB3U/s1lLdJFAqlx07gBtukCbexkbnxCoAa/uVgFPmO7P+2dRkrwy41fI1sln7SWY2KzNJ8ngUDm6edaeMnwHCQt/IypXO2vzcucG0pRqxW0t1k0CpXISQIVK5nLPGpaeceGwmfJwy3+kjNtavlxMAs+yQQbFtG49HYaH6rB95ZGQEPsBCvzdWpSz1RCzusmLQ/Ck0Jz1tUNU0mebmgrf+mDHODk5eoSLwtQQd+/ez5hVnzCaWmmlWswJoERtTp8oJ6rnnhjuoL1zY+1nRw/kk/GHYMG/3CwgW+kZUtMlDDgmmLdWGPtFFOi1L5eq1fr2D1euvB6f1q6A5FEbIYYcpAS0zp/HV3CzNuXrH0TffLDj7ffnL4bU5m5XhgkYLmSbs587luhF+oFKkLZGIlJYPACTKzRQVIhMmTBDr1q3z/sQDB8qMV1bU1fnnNV6tpFLAyJHmv2t9PbB5M/CjHwELFkhhX1cHXH458MYbMv45SmjtPfrosFvCeEUqJTU2M6tPba23Zbn79pUJoEpB3/dmzZKZ/BIJ2W7ul94yaxbwu98VR/EYGTQI+Ogjzy9NRK8KISaUcixr+mY4Zc1Kp9lU5jV26Sw1Bzmjg9XChVLgRy2igrX9yiKVAk491XqZx0uBD5Qu8LVjtUgTbSlCH2bK/dI71q61F/gAcOyxwbTFBSz0zVBZg2EvbW+xc4bSUuyaldIFys9r7jURCs9hPKCpCdi+PexWqCGEfFbmzu39vEQsSUzsaW4uFO8yo2/fSKZsZ6FvxsqVzvuwl7a36EvlmqEvehNVtDzrHDddOagk64oa2awU/GbPC2v73nLSSdbb0ulI/tYs9M1Q0eIjaLaJNSoOlAMGyL+DBpVeuMRPWIuqPJqa4um/Y7UUwVYob7FL2Z7LAS+8EFxbFIngyBkyqZRzIhaiSJptYo0+HG/WLJn29pxzZDrcREL+3b1b7rt9e3hx0XYcPMgx05VEHLV8Oxoa2ArlJSr949xzg2mLC1joG2lqco7LFgK4/vpg2lMtGPOdX3+9XOfX1vLN1vSjhhDSuZC1/crAKSV3HKiv51S9fqHSP555Jpi2uICFvpG1a9X2e/xxf9tRbegzoXV3F3Ja6z2P40A2y9p+peCUkjsO8Bq+P2havlP/iFpkEVjo90ZLznHttfb7dXUBGzYE06ZKx5jvPJOJ92DLE8L4os9ep5J8Jeqk01I4sfXJW+xCjPW0tUVOTrDQt2L1aud9rrjC/3ZUA0EW0gmCPXt4kI0rjY3y2W9oMA97iyMR9SKPNW7qLURMTrDQt2L8eOd9Xn+dB3cvWL06/munesyq7XH+8+ijd+Ldtk3W2KiEfqnVsuC+5x12XvtGNm3yrx0lwELfCpU4/JoankGXSyoF7Nwp175mzZIVqSqBZcuKhbxd9UAmGjQ2FvuOdHTYJ1+JEwcPct/zEidL8JFHFnJ2dHYG0yZFWOhboZKVL5PhmNdy0VKGCiHX9SulmFFHhyzB2tTUOzKBNa7oYRaqm8vFx4FUhQULuO95hZOmf+BAZH9rFvpWNDcXksFYMWgQh8GUg3Gg7e6O7IPimlxOTmTuvhu44YbC+h97U0cTo5YfZ444wvz7ri7ue16gEp+fyUT2t2ahb0UqZV9pD+CsfOViHGgzmcpYQ9WTywEPP1xcKIi1/ehRSWm1tSRWZvDafvmoxOd3d0f2OWehb0VTk3Oq1//5n2DaUomoZD6sVFjbjx6V4kviBHvyl49q/obu7kj+1iz0rVAJyYhYKEasUMl8WKlw/vPoUVsbdguCIZfjvlcu99+vtl9Efb5Y6FtxzjnOmj6H7JWOaubDSoGo4LDIKVGjRSoln+VKpbaWK0B6yYwZ6vtGsEYLC30rVDT92tpImm9igZb5UAhZXCeC6So9RQjuK1Glqamy+19ENc5Y4naCGMFnvmyhT0T/QkSdRCTyr5kO+x9GRD8gog1EtJeIOojob0R0AxFFx8amknyBzbTeUOmDrkYEi28wcJddLY7U10uNkxNElY/bsSqCznxeaPp3AVBKUE1EIwG0APghgFMAHArgCACnA/glgLVE5BAnFxAq5udEgk1l5ZJKSYfISh50NaphYhNH3GRXiyOa4ygniCqftWvdjVURdNotS+gT0QwA/w5gi8K+dQCWARgFoBPAtwEcC2AkgHkAcgDGA1hSTps8QzM/W8W8AvLmR6yYQuxobJSJLKqBtrbes37WvsJFS5xUyWhFdzhBVPlockE12iOC1uCShX5eI/85gCyAGxUOuQbAmPz7q4QQvxZCbBVCvCOE+CEArR7pZ4loUqnt8pz9++23T50aTDsqkVQKWLQo7FYEy/XXF39m7StcKq3YkxXpdCG2PILaZ6xIpYC9e+33qasrOE9GzBpcjqY/H8AQAL8CsF5hf61WbbMQ4hGT7b8E8JFh33BJpWSspR1m2hujRmNjdQy4evRldzk9b/isXVt5CaHMyOUKzxoniCoPleQ8EdTwNUoS+kT0rwC+CiAF4BaF/UcB+ET+46Nm+wghMgC0EfE8IupXSts8xaiVWcGzZmeMZuxqTc6TThcX4eH0vOGRSgGHHy7/XqvTM7RQ3Ur2weD+VjpO/l59+sg+FTENX8O10M972N8DgAD8pxBij8Jh43TvX7HZT9vWF4VJQng88YTafhGd0UUKoxm7knKdu0EL89S0fE7PGx5an7zmGuCuuwrfaxMxlaxrcSXCmmjkWbnSfnvEaxyUoul/D1IgPyOE+IPiMaN17zfb7Kd3CBxtuVdQOM30p06Vnr8RTMAQKVpagHvuKTZjV1Kuczdown3u3N5LG6x9BYd+aWXZsrBbEwy1tcBRRwETJ8pn8vDDiy1v7FCqhsozGuHwXFdCn4iOB3AzgDSAb7k4dJDu/UeWewEf6t6HnwzbqQ7yn/7ETlgqzJjR24ytUrq4UslmgeXLe68LsvYVHNXiwKcnkwG2bQNefhmYPr147GKHUjVUoz0inNbZraavxeT/TAjR5uI4fZH0gzb76aXsoWY7ENE1RLSOiNZ99JHd/MEjkknrbZkMO2E50dICtLYWPmvhQ/Pnh9ak0Emn5aRHy0iof0V0HbCiMC6tVCOtrYWxa/16dihVRbVmSIRTtCsLfSK6AsBnAbwD4Fa/GuSEEOK3QogJQogJgwcP9vdiqZS90Ndgs2wBo5nQLE91Og1cemmw7YoSEXbyqQoaG+W6KyPHrunT2aFUlbVrpbLnRIRTtCsJ/XxM/i/yH2cLIRzs3r3QB7vbZe+r173f5/Ia3qMSmgGwE5YevZnQKk91Lmdf87vS0aJCtAnS+vW8nhokK1ZUtpOeG9JpqfWzQ6kaK1cCfRUS0EZ4qU5V0/8BgKMAPCGEKMXrZbvuvZ16rt+2o4TreMvq1er78gy5d9z53LlATU3YrYoeDz8sBf348cCLL0qH0NWr5e/F+EsqBewLX5+INAcPSmsI0xsVX5AxYyK9VKcq9Efm/16sK6zT80Kx1/39um0j8t+1mZzL7jrGY8JBpbyuRoRndoFhjDtfvlzNFFaNfPGLhVK7b74pv1u4kDUsP2lpAUaMqO61fBWEqN7oGidUkjmde24wbSmRoErrvqZ7f6bNftq2gwDCL3CtWlxh5sxIz+wCwSzufP9+NVNYNfLPf/b+Lptlbd9Ppk2T/ZJN+zJNbHu7VGy0yWdzcyFMef9+noCa0dwsfzc77rsv0r+dqtC/EcBYm5c+V/4tuu/bAUAIsRkFIW7qwUVENQC+mP/4rBAi/Cos2g12itdnDc3c7KXP982owX3JH1paChYVRj6X119fHKY3Y0ZhQsTLleakUnJZzo50OtK/nZLQF0JsEUK0WL1QrJW/q9umH/G1lFfjiOjLJpe5ATKXv37f8GlsdNYM+AExN3vp830zaui1fU6Y4h1mUSTVziOPyOdzwQLgkEN6h9ayQ19vNAdlO3K5SC/1BmXeB4DfAtB61UIimkVExxDRcUT0AwC35bc9LYSIzoKS6tpWtT8gWslJY8x5Jecv9wstQxwnTPEGY64IppiuLvPy1qzMFKOamGfWrEgv9QYm9PNa/0WQaXjrAdwJYCtk3P8PASQBvApgWlBtUuLoo9X26+7mB0RPKgWcfTavn5bCsGFcgc9LWMsvDXZOLkY1i2PEf7MgNX0IIbYAaIBc998IGb+/B8A6AP8J4CwhxK4g2+TIOeeo7ZfJRP5mB0pjo7n2wJijhfloFhKuwOcdmzaF3YL4UVcHfOxjsq4ILzOpZ3GcOjXSWj4AkIixJjZhwgSxbt06fy8ydqw0DzpRXw9s3qxuGahkUimprVZjFb1yWL8eOPVU+fuNGiXjpTW4f3lDMsl+Jm742MeA88+XDqbf/CZw551htygcZs2Svg9OQr+2NhDnZSJ6VQgxoZRjA9X0Y0lzMzBggPN+bN4vUK1lc8vliivkXzMzImv75ZNKscBXRUs/nkpJgV/ty0wq8fmAtPhG/Ddioe9EKgXs2eO8H5v3JakU8NBDYbcinmhFOswGGF5fLQ29aVpLf8w4o5+08zKTVP6uvVZt34j/Riz0nVCtqjRoUOTXcgJB9fdiekMkfz+zSIhqT/5UKloERGMj8OijYbcm3lRzGJ+q5z4Q+ck5C3073NzovXur82EwsnZt2C2IL7kc8OyzYbeictBHQCxaxJEkXlCt2r6q5z4Q+ck5C307mprkWr0KXV3V+TAYWblSJvqII3V1aqWU/YQLFHmHMQIiSiQSwJQpYbfCPdW6zKS6pt+nj/9tKRMW+lZoWoKq0AeA3/wG2LDBvzbFgaYmoLMTGD067Ja4F6DpdPjCobUVeO65cNsQV/Tr96ohVkFRV1dcvCuXkxnx4kRDQ/UuM6mW1P2P//C/LWXCQt8KN+YcPZoHdjWiN6e2hV8ksdeELZGwfnDr62WsfBS41LQ8BeOEPoNhqc+vX6TT0WqPW+rrZcx+taLanyJebAdgoW+NqjnHiOaBXY1EbaA1kstZ39NsVmqJmtNcmBOAXbvYYuQWYwbDZ5+NjpY/c2Z0JpSlkslU7/KlG6tRxIvtACz0rdF7UKuGamhE/Kb7Qtjm1Kuukveprs5+P6tJieaZvH49cNZZwEcfed9GN1SzxagUjOv3tbXSsnPVVeGXd164EJgwwblv6klEbGju7q5ez303ykzEi+0AnJHPGbPsaE707Qts2VJd2dNUM1b5RTIp1/C7upz3nTVLTuaM7a2rA044IRrFWYhkWedq6kOlYveMJpNSgGYywbdLz6BBwPbt4bahXGprga9/vfqy8qlmZU0mgfffD+SZ5Yx8flKKyToGJh7PKXU5xCuyWfWB/aWXrBPgvP66+TFBU1tbfX2oVOyeUTf9wk+OPbZgOWxoCLs1pVGtCciam4Err3TeL5uV+SAiDgt9O0o1WedywAsv+NOmKKH3llZ9MPzEbODXPI6NSW7MEuBce210SgFXa2hUKYQ94VTh5z8vvG9uLk3wR6Fv/s//hN2CcNDKXTuxaFHkl0BY6NtRqmNaIiGFYaWj95ZuaQEefDDc9ugFfHu7rJCoeRwbK4WZfb7vvug4IqZS1RkaVQpG/xs3a+dBYYzIME46zdpdVycdALXvicJf669WXxPV3z0GyYtY6NtRqgZRDZq+0Vs6zDAzIrlOrxeS+gmJ6mcVf4CgiPjAEVmiqvU7RWTYLTdp3+dy4U9K33qr+srsplLA/v3q+0fdQieEiO1r/PjxIjBGjzbLhm7+qqsTYtas4NoWBtdeK/9P7f9V/W38eiUSQqxfL9vW3i5E377y+/p6IVpa7D+nUkKMGRP+/6B/jRkT7v2tBNrbhSAK/16Wek/1z5jV2HLttcH/D4lE5Y9veszug9nrxBMDaxKAdUKUJjdZ01dFCPV9K3091ujrEAXNKpczL02bzQLTp9t/bmqSSwG1tYXzJZPy/3QbrukVE0pyzGWAwtLN3LnF9zRsNm1yt79KtcWga120tlZfmV1V65Hb+xsWpc4WovAKTNNvb3c3G04mpfZYqajOfMN4PfNMQYtXffXtK0SfPr2/nzrV/Fx9+pjv7/VLs1ww7rj2Wqnh19SE3x8B/y1/Awb42/4jjnC2OFQy7e1q411DQ2BNAmv6PuN2fTWblVpGpRLVdVNAFjEpJcTS7P955BHzXPxdXcGs/0+d6v81Kg3NCiWETCijFVCqrwdOPNG/686aZW0V8tPyl0pJfwE/2b3b3OJQLdp+U5PzeJdI9EpTfPPSjTh+7kqMaFyB4+euxM1LN/rYSHVY6KtQiglNNcQjjug9j6MWc7xrV2khlkL0/l6IcGO829oKzl/GaINqROU3MFbG1CZt6TTw5pv+tW3BgkIZ7vp62Va9HuhXJEZTU3hLGDHwVC+bVEreWydyuaLf4ualG7Ho5XeRzY8rWSGw6OV3IyH4WeirUEoM+rBh/rQlajQ3h7fubUZdXSHj3rXXyhm49tnuZTV5GTMm3HK7Z5whhZwx2qAacfoN7Cpj+l09saurMNkMUhiuXRvexLTSfZcANS1fQxextfiV90x3sfo+SFjoq5BKyaQLKowe7e/MPmictCttoI0K2kBkDCl00pDNkvUIIR38wiy329UlJ5ALFlSfA5UelfvZ1BSuZUZbVgrS9L1yZXBJe6ZO7f18VMo4Z8Xq1er76pxvs2aWQ5vvg4SFvgpukvS0tVXWoOykXYVdWc8q457Rg79UzSto72gzurvD0SKjhNX91E9Kn3023AmanqDuU1OTrDkRBA8/LItRVdL45sQ556jvu2JFz9ukxUTM6vsgYaGfZ2nzVpx923MY2bgCZ9/2HJY2by1sdDvwV8qgbKVdaQPt+vXBVtarr5eOevrPZjW+zUIKS9W89BaA9vZwTf1AdTlQadjdT/2k1MysHxZBmL613yVI68bLL1fO+KbCs8+q73vssT1vJ44aYLrL5WeGv+zLQh9S4M99bCO2dnRCANja0Ym5j20sCP7mZuDII9VPWCnZ+Ky0K22g1ce7B0EmAzz6aOFzd7f5AGRmffBC82pqioYmWW3avtX9bGwsTEoXLAA2bw6nfRp6f5IgTN9hWdnuvts+u2Aloeok2bdvjwKytHkrXnt3d69dThhyCG6dfIqXrSsJFvoA5q9qQ2emeDDvzGQx7wldiVU3jnljxnjUshCx0q407T6XK04RGgTd3cVe9pmMudarktTELVHyXagGByo9Vvdz+fKC0AsyhTKR+TMe9H0JK3RWnwir0lFNuKOrrGomTwDg7Q/3F1uQQ4KFPqRmb0ZHZ6Y0bf/hh+NvfjXTIrq7gU9+sqDt1tbKwS/MdSozbd/KKa8czSts3wWNMWOqw4FKj9n9bG+X+dDDEHpEcnnL6z7mljBDZ1tbq0Pb7+wEnnrKeT/dEmi7hTwRkBOCsGGh70DRTXKj7cc9OY+ZFpHJAAcOFNYQ02n58IfpkRpUjW8rraqhIdiSwm+/XXhfzbH7YU7ColhQS5sABBk+Wy3a/rRpavvll93697NeErCaEAQJC30Him5SczPQp4/agQ8+GO/B2FiulCga9byNNDQEo11ZWQ9WruwdznnVVcAXvuBPOwYOLPSrao7dDzsrZBRLZwe9BPX66/Ee41RoaVHPeJhf3rHTgfrWhi9yw29BxOlXZ/DW/upX1Q40ZGiKLfq0pmFo9EbnqKjFCTc29tY4Fy2S681+kEpJK5LbPASVhjYJa28HPvax4K8fJZ8KfYGhIK0ftbWVMcbZMWOG2n61tT3lvXd3WkdTdGbCXyJkoe/AgbTOISOVAu67T/3gKA0MpRK0xzpRsTUlyiFqVkmb/P69Fi4sHuCryZvfuKTR2Ci/CwKiQnrdsCebejSLz/LlwVo/Kt2hNJWSy5cq6JYZj+lf72OjyoeFvgNFuq2bjF+DBkVrYCiFMOKAzfLdR1WohbWunM1Kwe9FHoK4oV/SSKWAhx4K7tpR1Gz1Fp8DB+TSUl1dMNc+4oj4j3F2uLnXY8b0/BZzLhjtU4O8gYU+gKE2M7OiZey1a9UHeV2ihtgSllAzXjOqGkWY2fr8yEMQdYxLGtdfH6wVKor90JhLY8WK4LT9gweDuU5YuEnME0UfDwtY6AP49EmDLbcRUBy2p+KpTWSeKS5uhO0sNXNmdNbuzTBz7nOTxMlLoiiQvEYv4Lq7ZenjIJk4MVrPtVkujf375fdBhPB1dVV22J6b6oW6Z2/uY9H+TVjoA/jLGx9ZbssJQ9ieLr+yJULEP2QP6O3Bnwi4uyxcGC+TdSoF7NsX3PW0mP0oT4y8wijgglhySiQKfT6RiF4KWrvMk83N0smxb19/26BPi11JuFnP15n2lzZvjYSznh0s9OEcO1m0fdAgtZPGPWRPj+bAGLSpP5uN1+QpyCpvtbXSpFgtsfphLDXlcoVran/vuy86v7VT5skgfrM334zO7+ElTU3qIcq6rH0/XKY4UQgRFvpwjp08ol5n5lEd1HO5eAksOxobg01zqidO2v7q1cEJJs1bWHNsa2ysbOEf1FKTFh567bXmDnG6dKuh45R5Mqjf7PrrK6/vrV6tFqK8fr3M2pdn1wFn+RB2Kl4W+nCOndytT8e7ZYv6iR9/vIxWRQiVJQ2/iKq2b6ZhuynDWS7JJPDAAwXHtkWLgBdfjI5A8hrNXO23Z7qmJVsJzDjlRTAuz/nF449XXpIo1We5hOWNsFPxstBXQACF4juqGfkAYM+eeAwOdqRS0jkIkKVsUyk5mASZnW/ZsuCupYpZNrwgvfmz2eIqh9msHNytBFKclwH0yWf81Fx1a7NFaW2NE424RUr4namvqytekyEn3ORj0afFBqCScC/sVLws9BXp0LIsbd6s7tAWt8HBDLPyujNmBJudz03NgyCwyoYXdP7z1tbeQtCu3HBctbGmJmnFWLjQ3+uYhV35UbExaILyh6iE8Q6Q/4Pq5JKoaKKTVRgWw07ew0LfLWZpV+2IWmEON5iFBC1YoO7VWg769LtR80o3mwhphF2C16zccJxT9mpalxD+Cy6z38aPio1BE9TafjoN3HVX/MP41q5VV2ryy49Lm7fi7NueQ87hsASFn7yHhb4iA7TKSW7Xt83qbscFMw3BD4c+IuCZZ4rDi6KaZc5sIqRvZxRK8BonInaTlKgTZERE3H4bVbSJSxCx+0LEv/qey0l7158ex9zHNlqWaNdz1qiBmDx2aKkt8wQW+pAJeJyYdGq+qIdbU/Of/uS6PZEhKA1BCOkQE4csc3ax0cYJQRCMNtEa9OZnp0lKlHEKFa2p8fZ6cTPbu6W5GTjxRP+vo6++F0dfEtVSunnE/v04dNd2pX1f3ryrlBZ5Cgt9ANMnDnfcZ8WGfFGPlSvdObFlMvHq8HqMps0rr/TPgW/Xrnisndqt8Yah5evvT3u79DrWnC0B+0lK1HHS8ru7vb3ezJnxMtu7JZWScfV+U1NT6F9x8yUp4TdK5HL49kuLlfbNhlGp1AALfQC3Tj7FcZ+e+MumJnfpGQHgmmtKaFXESKWA3//eewc+bRJhVUI3aoOw3RpvGGmL33yzsIZqFVEQh8mUGUHmPQDilROiFIISvJpfyfr18fMlaWx0fUhdrhtTNj6DwfuctfhkkFFPFrDQd0spA3uYce5e0djovWYFFCYRcTI7W9HcbJ3UxU8mTrQeYOPsiHbOOcH+lnGxgJRC0A6mmQwwdWrBByguv22JY3VCqGn7E0cNKOn8XsJCH+oZkpY2by0M7G4I27GrXFpaZFphv4nLwGBHGNp+ZydwySXxddazIozfMu4TTyuCXnrq7pZWqDhN6vU5SVzSJ9uN8VvfcNzvnR3hxugDLPQBqGdI+v5jG0qbMRtiOWPHl78czHXiYna2Q69ZB+EtrbF5czyd9ewoZYJdLgcPlmTijTxhV8wEoj8ZLXFilE7W4MGxF2LS1Xc47ht2Yh6AhT4AKIVaAMCBTK60jlFbG+3Obscf/iAFip8MGhQvs7MqmtBy6wPiBVEfYFUII+eBEDKlcdwnTEZWrnSXTdQPoj6pL3FiVKeo5QPhJ+YBWOgDcOdc0fHcavcdI+qd3Y4rr/T/Gkcd5f81wmLt2uDizPXEuc9phJXzIJutPG0/yHwHZuhTeEcVrb6DqjxIJHqcjxffs1TpkLAT8wAs9AG4C6OYNPMO94IwrqFATz8dzEBhlv60UmhuDs7MTyT7WpytJvq47jBN0pXgfKtn7dpwfYviYnm6/nr1CCWd0+yE4wYqHRJ2Yh6AhT4AoH+9uvl1a0en+wEhigVjVCihglRJPPNMMNcJC6MHvV8e/kLEt69p6MMOm5uDsTSZsX9/ZZn4tT7Y3l6c+TIo4mJ5euIJd/vna12EXTnPDSz04S7fTJIIOPpodxdwu3+YaJrWM88Au3cHc80w1rzDwu+sfXEWVsYaAevXAw89FE5bIqKZajndRzauwNm3PVd+Lfagl0z0uTeiaHnSW5ZSKfdpxjMZ4KWXlPzCwo/Ql7DQB9BxQN2EnRXCffxwnMzXmqY1dWpw19Sn7VREdTD0fNAsF78H3XQ6EsKqJIw1AqZPl3/DIAKa6dLmrT053QWklfHGJS24eenG0k8a9JJJ1KNItPGusbG0ZbhkEnjySaVdw8/FJyERgbSApTJhwgSxbt26ss9z9m3PKXvwD963E39fcI0M7VGlb19gy5boa/ypFDBqlLv/zQvq6oCvfQ24806l3bXBsDNTEAj1tUn85BKZWXHeE62FUsgm9K+vxbyLx/Ssry1t3or5q9rQ3tGJY/rXY84Fo/1bexs7VuY98JO49Dc9YfU9MxoaQtdKlzZvxXceXm/pbzRj4nClTKJKXHWVf3k4XD7bgeJRn9s85Up8ZpSzkjS0fz3WNH6mrGtpENGrQogJpRzLmj6kR2VtUs34MnvNYveaWhS1L7NCGH5qobW1clZshkutav6qtiKBDwCdmSz+608bMeeR9bYCHwA6OjOY88h6jP3RUxjRuAI3Lmkp0qbmPrbRP4tAEBXPotjfnLDqe2PG+FskpqEhsGyFTlYnbbvWJ+0cjB96+d3S+qjxuU+l/F1CSadl0aQoavsejXcf+9MflFLwRsFzH2ChD0B6VM6/9LRC+VwbxrW3uTeP5XLR6vipFDB+PPDiiwXh4PdacybT21SrhfG4HGitElzsT2eRcSporTUnJ3rqKRiP6Mxk/XfM8TPxTC4HvPCCP+f2Cyuzc2ur90VixowJbJ1ZL8hvMEwu5zyyvkdw6035gLMpWEA9qVgRxvoMjY3+L6FEMeGRh+OdagreKHjuAyz0e5g8diiaf3C+436Trr4Dm6d8xb33dZS0r8bGgrDV1tzCiIku0VkqiAQXKpmzyvYXWLu2xNYpMGZMvEqa6iMc9NENCR+GqLfeCqTkq1GQG8nkBG5c0tIzITBar5xwnd3N6Cj5wQfBhSYuXRrMdVTxcLxTTcEbFTwuSF0ddK1eU5q2HwXty2jO0xKRPPJI8DHRJTpLzblgNG5Y0uJ9e3Q4TSyMfgXasgDgYkbf3Czvx8iR7r2GnVi6tCecKJLrqVYYNTA/JqInnFCs8Xr0+xh9Qw6kux0FeTkeVa4nv0ZHyaYmYNgwYMeOMlqhyO7dcpIRFT+TchwaU6me/8ONP1hUYE3fgMrK/udn3lFazOu2beFrXkZzXjot044eOCA/W627+0GJDlOTxw51lVuhFPZ3dReZXvUa/c1LN+I7D6839StwbXL1K1NaOh2vkqYaQVicWluBBQs8/X3MPO13uYgKckt9bdLdGrFxMqXVZ3jySWlZ8cOiYuT66/2/hirlLK/p/o9PnzRY6RC/xys3sNA3oDrzXtq8VQ5QbtbDtm8P18Rv5bSj/x/8Xt/TO06VsZ467+IxqK/1b4LS0ZnB3Mc24ualG3sN5oteftfSycqVyTWVkr4efgq5iMSbKxNUSJl2DZe/j9WSzvcf2+DaPF8OOSEw74lW9aUls8mUZuXTTP5+8/DD0ZqAlrq8pluqWPK3d5UOmXfxmNKu5QMs9A2o5uH/4bLW0vKq33tveB3f7STFSzTnqTIdp7RB98YlLQAEEj5mvOjMZLHo5XddDeauTK5B5ENPp4G77gI2bPD3Ola4XTvXNDA3GbPKwUVFQjNtfu5jGzH9d2tlMa4A6erOoaMzox5xYjaZSqeB5cuD9eWZOze4azmh5dp3m6EwnQY++ABLm7dC9bZHxYkPYKHfC9U8/LsOZHDbt3/u/gLpNDBuXDiC30/HMTvq6swTFLkUCMZBtzOTg6KzfiC4NrkGlQ9dCOCKK/y/jhn6tXOV+61ZP/zMH2JcwlLU9q1CRdds2ull60qiM5PFDUtarLV+Yypo7TVsWLC+PI8/Hty1VGhqkr4vbpk7F/OeaPW+PQHAQt+AStiexr/88selOeKkUuHMeM1ywAexlmflsGcMH3LAbNCNEtqavrIXvz4fut9lT1tbg59oGr3F5851vt9NTf4LIaO1S8GhdGnz1lg4bBm1fscIE82yoqsY52tfDNJnSIW1a0sT+suWOeYD0fDTGlkKnJHPwMf/z5PoVLTZvPbfl2Pgwb2lXSiZBN5/P1xv1iCyw2m1Cl57rfh/1WfDqq8HNm92/C1GNq6ITCpLO7TsgMomvVmzgHvu8VfrTyRkCWPjffCTWbOkw1w6LZMz5XJS4Nrd75NPlhMUv9F5YJuheeJv7egEkb+GBz9IEJAT0jFZ3/RefVP/HPbtC3ziE7KP+AWRHHNOPdW/a7ghlQKOOcb1YbuPH43TLlW39L5z2yTX17CDM/J5iKrAB4DU4Wqem6ZEwcGqudn/SnpCmFs2TMKHnLSSIOLzvcCVF79eG/aTXE5eKygPaqO3uD45k1nf10z/E0oax9xRV2f77PVKlBMzgQ+gZ9nLMfGU/jlMp/0V+ID8MYOq3mmGcYmphKRBAsCK/v/ibbsChIV+GcycMq88zTPscKpUCnj00eLv/DK/LVxYnPrTED6UXXAffv7gC0VOUjcsacEn/s+TPZOAT5802FePfS9R9uIPOinSww8DZ53lf7+z+7/MnOeammSGyEWL/G2Xdn0bc37Ul5HKpadvBpETwYw33wzPsdS4pFhC0iAC8KX1Tyul3gWC80lVhYW+ATdr+rPXLC5P6Iet7V9/fW81xi/vfv3/aiIQst3d+PoLvcMJD2RyRaFycRmMla0SQVc9A4CXX7bWtL2aDDj9X/r+oHfe8zO6xCHts2ZpisPafTn09M0wsnBqXHRR8Bkj9Va1++6Tk999+0o6VSLbrZR6F4iepYiFvoFbLhqDpKLnxbj2tvJ+wLDLdz7xRLDX07Q7E4FQF7NUlna48uLXHKncpnUuF2MtCJdOlY7onUbNigvp+34QzntAIUOhCU4pcyuFor4ZxoRT4913i2t/BEFjYyHzZTotJ78lTnr6iJzyeDU0YsuS7MhnwtLmrcppXk/athlPPjBbKZNfLxQd2HyhpUU68vlJIlH8UFmU2VzavBU3LmmJhZOeCn2ThDd+fKH6AUE4VBrRO/YJ4dqp0jP8SkNshcX/Vw0a/lCrstFmzpPGZ9cvgupvqZQMT/TIktSVrMHo7y5V2tfTMsh52JHPY9wkUvjV8p+VfiEbzcMX9CbcGTP8v55x0LCwbMxf1VYxAh8ADmYFpv/ORU4EY7GZINA79o0f3zsne1AEpeVrWPx/lS7w+9UmzAV+KgW8/nrvA4Iy+wfV3zysJigAbOmv7vH/lzc+8uS6XsFCv0xG7vmgNC0fkB7NQZr39SbcTZv8v14yWVhDNUm9W8lrqCUnbAk6gdLDD8t7ZMzJHtRa6+rVwS56mkw8lzZvLf0ZjgkHMjn855KW3nH6TU0ynFJPXR0waFAwDUunC+vrfvY5D6sJEoDjOtTb6roaos8oC30iqieiLxHR3UT0KhF1EFGGiD4iomeJaBYROS5eENFRRPRTInqDiA4Q0XYieoGIriaKjp+jSksSBNRcdlnpF+nTx/d63j0YE6UsW+b/NW1m8Tcv3dhTX7xS0cIOXZXg1bR+v5P12BGktj9+vL/nr6srJJ2xmHje+HDlLC3ZkQN6Z5GzSs977LH2Phleoq2v+9XnUqmSHfbMEADO+eYC5f2jFmrsRtPfBuAxAN8AMA7AEZCleQcB+AyAOwG8RkQnWJ2AiCYA2ABgDoDRAOoBHAngHAD3AVhJRCGOdgWmnznccZ+cQHkzyHwOZ88x88Q2xsVPner9dc0w0aoafvgUFr2sVqgizmzt6MScR9djziPre+Vrd8zaN21aIG00JUgHU7/rudv8L0ubt2LOI+sj513tJ72yyFml59UrIytX+pu5UxuX/LIweVzjQgDKnvsEuEvNHQBu7uRhALoALAZwGYDjAQwE0ADgLsjf4iQATxHRocaDiWgQgCcADAGwA8AMAB/LH6N5dn0OwK9K+D88R8XxYvC+ncD+/eVdSGV26zacyuiJbVZWc5dajGlJWIRGaR7SqukrK4FMViBjKBCglLwnCEuMEYeQNs9Jpcp/fqwYNQo45xx5DYv/Zf6qtl73phoYNXcFbl66Uf2Axkb/s0UC/lmYVq/2tP0JQNlzXyBaxXYAd0L/TgDHCSGuEEIsEUJsFkLsEkKsF0LMAqClXBsBYJbJ8Y2QQj4H4AtCiIeEEB8IIdqEEN+CnDgAwH8QkbeujiXiVAN59prFyGbL6ExC9A6dMsNNOJXRjP/BB8HH41o4KFZ60hM3OK7zBVETwcjBg8HWhPCzX27e7PjMRG2tNShyAlj08rsY84M/qy05+W2N0WcE9EPbP+ecQkhsmc9V16GHY8RNyzHp6juU9o9auB7gQugLIb4lhNhms8vPITV4APi8fgMR1QL4Wv7j40KIl02OvwVAJt+mb6i2y0/GHHOY7fZx7W1IZsr0PE6n7YW5mRC3wyS9beDxuBYOitU6yJrR3y4JlJ8asB1CBGth8LtfOjwzRzhM6iud/elsz5LTnEfWmwv+oPuilbZfavIoj7MO1u3bo7xvFE37gIfe+0KIbgBv5T8a4xn+DdIHAAAMeV97jv8IwAv5jxd71a5SWdq81dEDe9LVd2DETcsL62Dt7e4vlMsBL7xgvd1MiFthZsa//37gyScLlbSuukqW9vWThgZTk2rUHFrCZN/BbmvtKsxMaQcOBOO5n0oBhx8uzfB+ks1K87RBYNy8tLqWmZzI5IR5qdhSS8+WikcVOYuO8/BZ2nzUCOV9o2jaB7wP2Tsq/9c4HdJLmVdsjte2Dcv7AISGcsEUPbNnl3Yxs1rzgLUQtxqUzTq4Nuhp1oJFi/wtqtHQgKX3LTf1Vv/0SWUUKKowMjlh3ceCssyMHt07E2CQcdOrV0szvJ+k07LP67K/LW3eioeqwJHULR2dmd6m/lJLz5ZK375SSdFjtHauX6+u9Xv8LK055hPK+0bRtA94KPSJaCyAkfmPRvO9ZuPIAfinzWm2mBwTCm5M0Wff9hyefOq13sVrVDHT9FOp4qQpGnaDslX4zfLlxdYCvzjiCCy9b3lPOlOjt3rUklSEjWUfa26WVqOaGn8b0NZm3l/8jtNvaQEefNC/8xvJZgEhcPC39+L0by3Cdx5eXxUheqXQK7qkuVkK4qDo6uo9vhmtndOnq2v9xuiESy8tq3lnb1Uv+xxF0z7graY/P/9XAPitYZumte/KLwNY8aHu/ZFeNawU3Jiit3Z04uCc75U+kJiVE21qKk6aomEXTmUWftPeLmNUg9AcOztNnfU0b3Ve0y9GABjRuAJjf/RUb1N/Y2OwGpYevzNFhhSOSCKHb7+0GNlqitErgc5MttjU39npf6y+ht65OZUCJk7sbe1sbVX3cdJjVlXUJXWnnaq0X7/aRCRN+4BHQp+I5gA4L//xLiGEMR7kkPzfgw6n0kuFXmF/+WtdQ0TriGjdRx/5pzm6naV9qu2V0rN66cvOAgVzFlAcRmUWQ+uExzGqtqTTSL9v7tegaf5Mb3YdyGDOozpHqlQKeKh3xcHA8CtTZCoFnHqqLK3qJ1pCnvb2Ii21T7YbUzY+o1wStZrp6MwUT0RXrgxO49e0/aYm4JVXrBUWt0tRjY1lN23Y6qeU9juQCcknR4GyhT4RXQDgJ/mPGwF8t9xz2iGE+K0QYoIQYsLgwf6tEU8eOxQzJjon6AFkvP5h6QOlXyyXkx1S81CdO7f8XOjauZ57LlCnsMZ1jwR2rUoik9Wt8Tc1+bsM40Qy2Xtd1QsaG4GNLuLDS0VbotA/R3kSeW2fcabI5yRI51Ih5JKnpvhYXde4FGXm4a//zoPQQ5FOK00ak9FJLtuLsoQ+EY0H8AiAJID3AEwSQpjZcLWYD6epot6m7l3exBKZcNxA1Cadb97sNYuREGU+EIsWyUHqxRfl+3JzoWverjU1gcZ8T25ehWFduwO7XiWxtaMTZ3xrIbp+p57i0xdUokTchk+lUrJfB0U2K31ZDFpinwoq4ew3Wzs6pbZvdCj2m759ge3bCxPfRMJ6DNNHZ8ydW1jr1ytQf/2r3MeD0MNMIqk0aYzyElLJ0oCITgTwJGSmvo8AnC+EeM9i9+35v/2JyM47Sa+677DcKyDmr2pDJut888a1t5VvMslm5aAoRG8tz622n0rJdbFcrrD+FRTpNL7+Qoim6Zjz7TWLgXInkF7gFCVi5khlNRlIpeSacMD9EMOGAUJg6WvvY+RNyzEi/1JNrMIANy5pwStfvSH4e7dtW2FZMpez1/aXLy8oS9pav16B0qKWPLCe1SkuEUXVcx8oUegT0TAAT0MK6T0APieEsJs+a3aiJAA7m/lI3fsSYua8RdXxTIvX/9zMO8pbt7bqlG5zoQe5jm8gCYEz3/1HKNeuBMa1t6FPNiQHPj1WznypFLBggRxIFyzoXd/BbDLQ2Ah8+CECZdYsoLkZS5u3Vlzp5iARAA59bV2wyb2cJhj6Ikrt7VKD1ytLZgpUNuvZmKiyRBRVz32gBKFPRIMhBf5wSMe7i4QQToHf+u1n2uynbXtPCLHdZr9AcJtM5r+XzXfeSZW+fYsd+FSd9/RafoBoV+tK1uCV4ScHeu1KQptAtg7xOWmNE5mMXFc1K9ykCQB9eJVV5sigzfoaL73UU+uhkis5BsGkq+/A8Y0rsPS194Pz4rdDrwSZ+Rqk0776xKgsEUXVcx9wKfSJ6HAAqyBj6DMALhVCrFY49EUA2kKvaaBkPhmPlqXmCTft8gu3s7XhHdu8q8vtlJ5Xj96sGoCWr9eauiiBbkr0dCT2kPaGSVffgcfOvKh38pygIAI+8QmZQEfzem5pAe66q3i/3/wG2LDBOnNkUA5gJuVzudaDd2SFkPH79y2X2T2D7JfHH29eBTBgX4POZF1FLBEpC30i6gtgGYCxkIrdV4QQK1WOFUJkANyb//hFIjrDZLd5AOogZco9qu3yE7eztY9/9zGcft2DOFjjwQPhlJ5XT1OTXL8aNw549tlAtfw+IoekYA9pr6mvTeIzHZuDNavqqakpxDQvWiQnlDNmmO87ZYp55shnngHuCehRzlsXtNLNIxpXsIbvMZ2ZLH754AvBOvUBwKZNcmJpJKAJpQDwxpHD8fHvPqa0f4Qd9wEoCn0iSgJYAln3HpBheSuI6FCLVz+T09wGIAW5rr+CiK4goqOI6AQiugPAdfn97jWJ8w+NAXaFUUyYvWYxKOeBdlFba52eV4822xVCvn/vPd8zuRn7tPEze0iXz7jhR2DSzDsw8qblOPsnz+LJVa8G24BMBj2F5rNZ4BvfkE6hZrz5prnz6dSpgU1As91ZLLno67hhSUtkc+qfffxAvHPbpLCbURZfe34RurtDsJ5Mndr7u4DSVROAE3e8p2y9nH6mWqh3WKhq+sNQXATnFwD22rxeN54gv0Z/MWTWvUEAHgLwAYA3AXw7v9ufde8jwS0XjUFCceY2eN9OTPnHs+jjhdDPZNxX1QNk9qwQMrl1jB6D6b99qSLMX1FgzaadRamMO77/A2RrQzL1A8ATDituxiWldBrYFdwSTzKTxsUtT0V6WemdHdLyoOLZPaBfLfrVhlBe2YFx7W2o6Q5hUtXW1lvb19JVn3NOwf+pvR0480ygn5neWTrdiqF6AHDr5EhUhrck0F4lhFgH4FQAP4MU9p0AdkKu+X8VwIVCiK4g2+TE5LFD8YupDUr7eqblaziF6rW0AHffHZ4JWKOhAc8vXuVYlTAKWFluopxMAwBOfe9/yy/jHCTJpLRWBUjUl5W0aKA5F4xGfW2yaJvW+4b2r8ft0xrQ/IPz8X8vORW1qhpHQPSqLJpMOh/kFZdcIv8afZj0ESNaFr8DB4AhQzzLIliXqxzrpZIdWAjxDnpbcUtCCLENwJz8KxZMHjsUNyxpcdxvXHubN1q+hlWoXioFXHaZ7PRhJoHo21daFgB8Z2752a7KhQD0rU2g0yIF5tD+9ZhzwWjMfWxjkYNXfW0SXx4/FH98dWtkHb8mXX0HCMAWvXk4kQj3/tuRzfrjQZ1IWC4ZRH1ZSYsG0nyFtHoUx+T7pdGHyLhf/3612HUg/KWLnglyKiUndkFlj9QqMmqCvrERWLKkEDFyzTUycknDwzDR1iGjKsZ66XMZr8qhf32t41rhzCnz8OI9X0Pf7jI1sj59gHfeAY4+2ny75rgX1oBfXy8fwHz7pv9uLRRyGPnKIXVJtP7ocxjZaD350A+sZgPuhOMGYv6qNmzt6ARR4ecd0K8Wt1w0BgAw74nW0NaMe4WQ9ukDHHQqZ1FhmAj8OAzICSqOBpo8dqiSo7Bxv5uXbsRDL78bat6BrBA4+7bn8Ps1d+G4roANs+vXF5cJ1zL1adX3fIhcyoEwc8oPlfbtXx+sdasUWOgromL9nb1mMciLbGpdXXIWu2WLnMkKITV77b3muBcW2ayMFHjtNSxNZSNh1v/xl+Q62jH96029tvvX1/YMnlYDrspArG0fYTO58Iv9Xd1Y2ry10MbOTrnEM3Zs4G0x5aij5PrqI8HVX+isqVMekMOiX20C//eSUz2J3b518ilFk9Ow2NrRicOf+XOw45AQUrDrQ0M1K4NWfc+PywL49kuL8YPzZ9nulyBg3sVjfGmDl7DQV0TFrOZpNrUnngB275ZavRCFdat9++SkIEzSaSCVwubZN2Hu6MtDbQoBmD5xeM+AamW+9/JhvHlpOMElHZ0ZzH1MXrtHgFiF0ZUKUdGyjSu2bSu7dKlbtHV8pwHZt+sTkBPSGiQEsLszY2mu9wptcnr2bc+FJvhP2rYZ/Q/my6Mkk/IVhG+RT4LdjiQEpmx8Br/65OX46NABpvsQgF9MbYh0Uh4NFvqKJIkciyhMuvoONK26EzNanizfAWLPHjmjve8+KfRzOeDee8N32gPkA57N4pg//QGHfuOz6LR4EPxGM7vrHzTV9dJSuXnpRix6+V1PzlUKnZks5q9qk/9PKuX9IChEaQJff3yAhLmOnyTCpp9cGMq1AfMJblD8+vHbCh/88t+IEE6TS73iEXVY6CuiUjVp8L6dmLLxGW88HvVmK82cFQWBD0BksyAAFJKWNdRBkKuul5bC4lesakoFh1b9bPLvfhx2U0JBAHjniKPx6W/e67ivn4RdSa3HuvVIC4Is337Sts04fle7d9lHY4DT5PKPr27FhOMGxkLwRy8QNKKoxNbOXrMYtV4XSwk4h74d2hCnPexhpNydMXE41jR+JrSHK+yBXmPuYxuxd+WqsJsRCgRgxO4PcNK2LaG2IwqV1CaPHYohhwfbjiItPwwGDCik4vUx1FbLxKeSe0SzwMUBFvqKmMXWGhnX3oZkBdfzMnu8go6N/ssbHwV2LTOs4vmTRIEKgc5MFu37uiu4tzlz15/CtXREpZKaajVQLxi8b2f4Wv6uXcBZZwHTpvm6nOQ2E1+Q96EcWOgrMnnsUPzkEvtMS5OuvsO73PsxIeg11bAfrMvPHGb5fZBtO2nbZpy4492qMrHq0bT9MDPwRcWU67YaaDnMXrM4GhPNl1+W6Z99JpNIKCs1Qd6HcmCh7wKVh9yzsL0Y0DpkVOApd8N+sG6dfApmTBzeo/EniTBj4nDcOvkU9HdZp6EcPC3jHGP8tDLZZcGNgmlfw8wKWV+bxAlDDvH8WuPa28yFRsQzWpZKn1xWeQkzKpYfJ9iRz2M8DduLIGEmQqmvTUbiwbp18imm+bWDXO73tIxzTCHAVyvTW/93EpY2bzUNAY1CP9SwiljxY41Z/+w3rboT01v+jDePHIaTdobv4OoXKmGhdUmKjOXHCRb6HqM9FPKBeLJiTCkCQNuRwwMX+Eki5ITwPfbZC3YHmKnvS1/5GZ58YHbVCf6gJp1aZjW/Q0C9wixi5UaF1OGlohUXS0Bg9I7wQliDQGUJMxN2SlIXsND3CUszWEyRTi3vYvC+XZYJKrymvjaJn1xySuQGWCussgH6QbWZ99OUxB8aLgg0PHRk44oeIb+m8TOBXdcr/OyP1bCMKQB8fuav8MZRIx337RvBiohWxKelEcGqSpuRmVPmVZxDX4YSWPbA7ECcp4b2r4+VwAfUIjy8otrM+3UiizPf+weW/L4xkP7X0ZnpKWs897GNWNq81fdreo1f/bGnhHh+GbOS++HtipPrzkwuNn2Ehb5LOhSrXFXiTLiPyOGo/bt8dZ46+/iBeOe2SaHG4peKFuERhJPXx7/7GEbctBw7+h7m+7WigFxeGobT32sNvHxunGKw9fjVHytxbDPDbcheXPoIC32XqHqPV6JDXw4EAjBl4zM4adtmz7WuGROH46Gvn+XZ+cJg8tihWNP4Gbxz2yTcPq3B16pbg/ftxCHdIddhCJCL2tYgkc+DHnSoXtihoqWi9UcvcRrb2gYNrxgrZyaZVJ5kxqWPsNB3iarJrCdmPxn9UosqdFMCmXwZy4TI4fZlP/NM6+pXm8A7t00y9YiPM5PHDsW8i8f4ZvKvFo0LKDYhB50QCgg/VDRKTLr6jp4sdSNuWl6Um6Szpg4bjv6XiumXbvKQxKWPsNB3iUqSHo3ZaxajJhtO7XWvSYoc+uRk2FKfbDdG73jXM62rriYZm/Uwt8xf1eZbQZRKtCbZEVb656iF6JWCqi9SKcxesxiUHxsSuSz+/e2/xb5f5gA8OPZC5TwkceojLPRLYPLYoUrOK2e811ox4RFW/68XWpdWMrYSBb+fJj9N46q2LJCAN/0umSDH5Zc4OpSacctF3tZ5H7xvJ5b8vhEnbdssnfo0hSCXxWEH9yMHwiMnnycrcsaQBIApG55WmljGrY+w0C+R6ROHO+6z8ejjo5Gy0ge81rri6izlRBAmv9lrFoNirlm5xYv0zz+fchpabjkft09rMM1od/u0hlg6lJoxeexQzJg43DNP+9lrFuP091px+7Kf9Wj5GkkIJCDwpda/QMS45G5ttttxYklA7PoIC/0SUVl//vSmdRURzqKl2x1x03IsbPg8upLF9guv1ljj4gjjhiDC+Ma1t6GPyRpqpU44NdNrOUl6iAqJd/Re7oT4aW6q3Dr5FPxyWkOPN7+WStptiLk+Mc+JO97r0fI1tDEvKXKxHv9qIBwnlnFZx9dTKdbnSLL9kAE48uDesJtRFp01dZg55Yc9n83Wkb0quhPHB8gJs4xunz5pMFZsSGGXYvinE5OuvgP/+7NLUJ9NF31PkII/rgOvsZSzRgLlp98Vojj5jllGu0rE6v8c0bhC+Rx6B9JMIoGkEKgxmXTGqd/p+1pXsgZLTj1fKRFUXNbx9bDQL4PaBJCxcVL927AxOGHHewBEbE0qxrzT5WhX9bUJ/G/T52ORz9xLzAZazVLkZrC14+PffazXd83/fRkGHNznyfmDRJuoaEKjM1mL5R//N3z5H3/BorGf9ywrnz75DhCdqnlRxpiYp08ua2tR6krW4Kl/mYgvtP010pMAfdu0JctfffJy2+yjhHj2mbjKokgwf0qD5Ta9CSzKnd0JL0vnalmrqsWcqoKfiXzi2vOMra7LZvCl1ud9i9GvVH8SN6j2Q7MwUbte1ifbjc++tbaMlvmP2aRFZckyrstnLPTLwE5IVUIMdWdNHU6/bqGnBU60wVVLGrIlptn3vGLOBaN9Ec2D9+1Ev8xBH84cPEnI9WHAvxj9SvQnUWVp81bs71JzBHUTJrqj72HIgdAnl4309NOsbSrKTjKm5YTZvF8GViFmlZKbWqWkpFuqeXA1Y/LYoVj3z51Y9LK3lcpkjojK8eg3Ros4mV7dUon+JCqYLbXZoaoADN63Ey/e8zUkIGKnEXdTAmfN+h/H/pUNspa2h7CmXwZWJkEzLT+O3cNL075GtQ6udtw6+RTMUAgBdcO49jbUxLLXOVOutm8Wnlep/iRO+JU8Sj8Gxk3pUe1fQdTY8AMW+mVgpbWamcDi1vHbBg1XzkalSjUPrk7cOvkU16FTdky6+g48OubTFSn2y5mMav4j7E8i8cPyZrR0xgH9c5IAHH1HCPH03AfYvF8WVvWq9YKyadWdmLbhadTl4vMAZEF4ZdjJnp5zqC40ijGn22MXkDjnidA8+Dtr6nDONxZ4YsrXJp3VEp6ngtUYVg5x9GfqFRbqsLRZX5uIbR9iTb8M5lwwGrVJ+2F1XHtbrAQ+IDNqeWXWr7TMZn7i5dJHpVTgS4gcvvfCA2VXdKx2jd4KP5JHVUJNCCdr0gG7WO2Iw5p+GWgDyA1LWiz3mXT1HXj1vy+PVZKe1iGjPDHrs3bvjjkXjLbtS26Io7alR++496XWv5TkVNq/vhbzLh7D/c8GY/KoI+prsT/djUy2YPDWkjypYhw7mlbdiakbn47FRCAH4MzrFnrqJBo1WNMvE6cBJS4al9uqUnbUJoi1+xLw8reKq7bVTQl0JYo1z6TIKa2z6ulfX4uWW87n/qeAPny25ZbzMf/S04p8Hsr1C4lDX9T+x0yyRsmJz6lQU5Rhoe8BdvGaZhpXFhQ5BysvUpsCcq1r/pTTeLANGX0Fvm6Kz2Oe0JVw1iDdNpUBOQFg3sXeVpWrJvSTAC/ySOj7YlRtT24KiNUmKNb9Kz6jQYS5/MxhltvMZrnJiOVKEwDeHjjUE5P+wEP6sMAvA681iJuev78nsU0cSKBQ4MlYMli1ouMvpjVwH/SI+avaPFNQZq9ZjEy+WFfUlB49TpPLuCs1LPQ94NbJp6CfRbyVNsvVvx45+TOR6vQE4PidW3HSti1ln4uT75THvIvHoDbh3ZQwDh78AuixRugLPJlZyZwG5CRRrAfkqOHV8xynhGVOTnxx71/syOcRnYrenIP37cTkfzwfyU7/6yduw79//Z6yzsHJd8pDG1DmPdGKjs7yq/BtPzT6lR4JhTS7yVwOyx64HhfN/O+SKjrGNUtaVPEqpC/KjqUCQNuRw/G5r/0m7KYEAgt9j1B9OL73wgOoieDKFgEYsStV1jk4+Y43aHHkS5u39nhVJ4hKEmhvDTwWJ25/N5KTTD1a++py3Thq/07MeeGBkpab4polLarMuWA0blzSUtHOfATguI4PlPYd0C++DnwabN73CJV418H7duJLrdHU8gGgO5F0HQvNmc38Q+9QlStRg/33TX+PbH+zggBc0voX130xgfhmSYsqk8cOxfSJwz1z5htx03K0DhnlSdu8onXIKNPS1GbcclF8Hfg0WOh7xOSxQ/Hl8fYCb/aaxUiEbOISAJaNPtt05l6bzbjOac6V8oKh1GWTaLmMqlNKfv0j+tVyH/SBWyefgl9Oa/DMiqJNAKIg/Hf0PUzZolSXrAx/ERb6HvKXNz6y3KY5skThBz//rZdNRUES7sL24lpaMo6Umjnt4999rEfDWtjweagUOQ1iVXxXn0N72tRl4tHtNi4fADoOlO8DwZijWZ28XD6ZdPUdoQv+Dw4frLzvTy89zceWBEcUZFDFYOfpGhVHFgJQmzOvqiUAfOfCG5XPxU5TwTF57NCiQjGlcMZ7rUgqiPQgpnIikXD06Har7bMTqf98+iR1IanCpKvvQNuRwwKPZhIA3jhyuCu/kUrQ8gEW+p5iN+hEyZHFblC/fdl85fOw01Sw6Nf4b5/W4Pr4vw0bExkX0vrMQXzvhQdsJ8JuqumxE2kw2FkzS2Hwvp04Ycd7gS9CEYDjdqWUazpU0ljHQt9D7AadSVffgdOvexCvDDsZp1+3MJTZrRME4MQd7yo9BDzIho/bcP5x7W2ReeATIofzNq0znQhryXlUU0L3r69lJ9KA8DoPx+w1iwP3O9FSjj966r/j9PdaHa1JlTbWccieh0weOxRzH9tgGbM/e83ink52XMcHgc9utXKldtfVck/bFTapSxL61CRw45IWzF/VxkV1QmD+qjbkXM4aJ119B1bcPxtjPtzsT6Nc0CfbjbcPG4Rxs39f1nlu5+x7geJ1Kd4z3ldbcvKSBICpLauARAIJCEzZ+Ax+9cnLLYvsfHl8ZZVijsrEv2I4aCHwC458spMt+/i/9Sos4jd6Yd9ZU4e2QcN77eNkUj1hyCFIJhLo6MxAANja0Ym5j23E0uat3jeYsaRUjWvS1XcUOc8FgXFI70rW4MGxF5ad9pmz7wWP16V4/3bsGKQTweuefUQWtXkrUzJn7zvi9ZJG2LCm7zFWM2G9I18il8WXWp9HTYiOfQmRwyvDTsYF/6Geher2aQ2Yv6oNnZliR8DOTBbzV7XxABwg5WhcQfuXGC1Lbtbq7bCrecH4g7EUb6lJozTGtbehLhe8rxMBPRaGuly3rbZfaanFWdP3GLOqVEYv5T65bOhFUEoZeCePHWr5AFTagxF1ytG4wqzA100JnH7dwiItf/C+ncoOVRr9ahO4dfIpfjSRcUDvUPrzqaeVpfnPnDIPB5PWWe68NvwLw18NO22/fwVk4dPDQt9jJo8d2qtDmYXrhRHhLgA8fPJ5rpyk9Cxt3moZocDhUsGiD+Erle+98EDgk8+kyGHOCw8Ufaf3dVFFtdYF4y/l9sPZaxajxsbq5PU4SYa/GnU5ayWo0iKTWej7gPEBsDKn5gIW/QTgvLf/VvLxmtOecWZfad6tcaHchCnnvR18il5jHzT6uqhq+zzJjA6Txw7FnAtGuy4Lrd37moAc+eyuoq/uaGS3B4WvogQLfR8wCkaz8roLGz6PTNLcLObnI+AmA5WR9o7OXkliOOd++JQy4Rq8byf6dXcVfddZU4e3BxxjaQL1in7dXT3CvcjXRTEZD4Fz7EeJpc1bMfexja6rQppZQLso4Vu/s5vg2vW9SptgsiOfD2gC8IfLWrHLIjWonTOVl9qXAHDGdQ/i10/8FN+6+CbLsBQVtM6vVYFjosHksUNt+5oZZgNuMtuN43e1W5pAVdFCQ60+awPsrz55WbGvS9beoUpr0/SJw7n/RQgz514VTEsni1wo+UvsfJwqbYLJQt8nNME49kdPmQ7G2nr64H07sfY3M33z5D9Q2xffe/6BnjVTu/h7O9iEH21uuWgM5j62UXnwNRtw6zwYcI0CHiaftQHWbOKhTQiM/ZQgJ52cEyJ6lBM+qmfwvp145c4rAzM/CwCPnHwevjfJOvX4jAqcYLLQ95lbLhqD7zyyHlmLTCpmzlQi/0rAfBBVoXXIKLx2zGhMb/kzvvT680pJKIwkiZATggfbGGAMpTqivtbW3Go24P7tzivLtjJZHS8ALBx7YZEwX3H/7N6anonGlSTCpp9cWGbLGL/wKmHP7DWLkUnWoE+2u+Rxzw1mPk71tQkczOQqesxjoe8zk8cOxbp/7sSil9813W7mTKXPmue24+cA/PHk8/DTc6/Ci/d8DQkIUN791EqLMqO+Nslr9THDuOxy9m3PKQ/Gs9csRjo/4HYla/DCiLH47CbvHP0IvSs4qkaPcDx+tJlzwWhTK9OAfrUQAqaTz8H7dhYtOToVX/ILo49Tn5ok/rfp8wFdPRzYkS8A7DI6pcpwrDMjAeC8TX/H7DWLkchJC4L2AGlrpk4e0uycVxmoLsf0yiOR7cZnN/3d07a0DhlVUga+uiRxPH7EMXPuvX1aA5p/cD7mXTzGNI7fGKapH6+CIAf0yhcByAlKpWcXZaEfAHZrXl7VlO6mBA7m01nWdx3AlI3PmGa6UvGQrlSzVrUxeexQnH38QMf9rMo+e6VttbksYaqRoMqpYV7p6BP2rGn8TM/4oU0IBugS3JiFaZaamU9bCnWLVmPEjPmr2ko4Y3xgoR8ATiEf+qxUnclavD3wGNcdOSlyPQ9Nn1w3+mTN13NVMvFVeqevJh76+lmYMXE4yEaCmzn1eSHwc5AD8ivDT3Z9bP/6WvxiKhfTqQQmjx2K5h+cj9unNYDIPExz5pR5OFhT5/rcpfZTu3Gw0rOL8pp+AHz6pMGWa/pAcVaq2mwGx+9sd92Z9X4AVjM5zcSqpT21CuGr9E5fbdw6+RTcOvkU3Lx0o2k/NKbE9cqDWjvHlI3P4PenfQ4/fPa3tmGjCQKuOHM4m/MrlMljh+LXD602DdOszxzsmQh0g5B0UXDX7VjZmazDx7/7mOX2SovLN8KafgDYrekP3rcTUzY+05OVSj8L82KF62CyFqdft7Ao7a5T2tNK7/TVyq2Ti82sZsxeo54K14ocCI+M+UxPJb+EyOH2ZT+z7HMEGRq1+SeTWOBXOFc9u7B3mGa+AJk2EagxCPyD6J2wx84SKgDLfP6tQ0bZCvxqCE1mTT8AzDRnzXv13cOH9Mo9rdfYyw1dqc11F3nsG9fTjCF8tQmq+E5fzdxy0RjcuKTFctAc197mShMw65/diQS+9HqhimSfbDdG73hXJtZpfhK/P+3zeOOokUXnqLTypYw5Z257s3eYZi5rq+D0MdlqNyYSgD7ZDD4381dF/UyFanBgZk0/AMw0Z03bPu/tV2xzT5e7tpoUAlM3PG2a9jQpcvjPl//Qs2//+lrMn3JaxXf6amby2KGYPnG4Zb/Sp4w+/boH0e3QA816bp1NFckEBG5fNr/X97ykVB28vuIFfPzmJ4tSkrcOGWUriPRLl24w62d2DO1fXxVjHwv9ADDm4tdr24enO5FO+Gtw0bR9Y2hWXbYbl7/+HN65YTzeuW0SWm45vyo6fbVz6+RT8MtpDY4D6ew1i3tqjhvJQXrlWw0gVln5CMDoHe/ipG1birbzklJ1YFaVTz/RXNjw+Z5lIY1SvPMJwHG7Usr7V4NZX4OFfgAYO7rRe9UsVKXbw1uTFMIy7SmyWaCpybNrMfGhJmkv9s94r9U6wx4RXhl+co+mZkbrkFGmgzhQrIVV04DL2FeH9CqSJEuER077rNK+1ZaXhNf0A6InW1oqheztzyGZ79gJyOpm53xjQc/a+uB9O/HiPV9DTXe6Z820K5FEQgC1ondu9bYjh+P6i76LlQ/MRgIyZv+sWf/Ty0vaLO0p0mngpZe8/4eZSDN/VRsyWXsdauPRx+OEHe+aTj+1ZaNfffJy2xh8sz5HAE7c8R4G79uFumOP4bwQVYpZJr9JV9+BplV3YurGpy0LkmnY+Ttpio4Tt0+rvrBQ1vSDpqkJSVE82CZyWSx7YHbRunuvbHq5LGpEFo+cfF6R9tSVrMErw0/GnY/f1rNvUuQw54UHel160tV34PjGFYAQxa/mZj/+UybCqKyhf3rTOtsBQls2smPS1XeYavuZZBKN6x4pSuTCVBdmpn7AugJp65BRRcsAWZ3IzxLhwbEXFvkKqCSEqsa+x5p+0KxdK7VrHX1yWRy1f1dRuVEzkz8B+NI/nkN3Pkc6ID2jp65/Cn1y3UXrppe0/gXzz53ZS9vPilJWyJhKQ18kxZgHXfvukO4uANIStfXwwfiXncXpSZNC4Mz3/uF4LdMSqtlunNexyYt/hYkxmgV0ZOOKnrV7J2FtDHMGii1PqgXFzJYXqgHW9IOmublHwz77J8/i9OsexMGaOhCAqRuexsr7v22bgzoJgVpjSVSTCYKVtl+tHZ0pZs4Fo3smiWZ5G4x+J92U7KWtdyVr8Mow52x7eketETctx8ibluPmP21A/zecJwxMdeDGkVOfzExPbbYbyx643rG2CFDdocks9ENkzgWjiwbX2mw3Bh3YbZuDmoBeHtUJmHtL9y4byQ5TjGTy2KEQMM+DblaA58Qd7ymVwXViaP96/HJaAyfhYYr49EmDlR32xrW3mYY5JyFw1P6djktOAHBo35qqNO0DbN4PlclHJ3BQN7hqwrwzWYtzvnkfPjp0gKlTi+bA0llTh+Un/Su+/I/nTGdvHx4xGEP716O9o7Oi60MzpTG0fz2+uap3HnQSoleURyaZxJb+x2BkR3tP+d0lp56vVKZZf701jZ/x9H9g4s/S5q3446tblUPzzMz/mvNz3+60adIxIx0HzGuTVAOs6YdJUxNqTaa3tdmCg5RdCEtC5HCehbNV65BRuHDmHaaVrxgGAG4e1980D/oZ77WaavV6bV+1TLMeTsDDmDF/VVuRB38pmBXxsaOa80Kwph8ma9cimUn3+rpGlyLXWAxFm80CcuDNZQ7i9OsWms5qef2esePzS+9F1jDprCOg9jOfQsOY6ejoLGhDZhYnbXBV1fareaBlrFGZDBKkhTNJ1MsZ2Ww5ShZ5+jx++Ow9vYo8VfsyJ2v6YaI59V17LVBXXFbSbLZqllzHalZb7R2bUcBk0pnMpDHqrQ1ouUWWQq3NJ/Cx8sC3WtOvTRTPJrg/MlZYTQaTRCAU/EDeuW0SNv3kwl5r/1bj4q9WyCJP3/3bEvSvr+05VzUl4jEjNE2fiCYD+CaAsQCOANAO4M8Afi6EqK5YHrMwPsOAmiSyHHhPb3+jZ5+sEBjK6/eMCg75GbT+88NlrUoxzxpa/5u/qo39SRhHzJL01NcmLYWzPtwUsJ6QnrD9XQAC0/7xLKY98Tvg6KN9+x/iBImA47aJiADcC+CrFrvsAzBNCLHS6VwTJkwQ69at87J5oaOPV9VDALbcNglLm7fyYMoEztm3PVc00FphN1gzjBVuxrWlzVudJwmzZgELFkhlqq4O+NrXgDvvDOJfCQQielUIMaGkY0MQ+t8H8OP8x4fz71MAzgLw3wBGQAr+8UKIN+3OVYlC32pwZc9nJkysBtovjx+Kv7zxEU9CGe9JpYDLLgOWLOmlpdtOElIpYNQo4ODBwgH19cDmzRWj7Zcj9AM17xPRUQC+n/+4AsBlojDreIKINgLYCOBQyMnAlCDbFwWsTF28HsqEiTagspWJCYymJuCvf5V/DVp6Ty0Tq+OMCc60wmIVpO2XSqCaPhHNAfDT/McGIcR6k31+CeAGyOqdHxNCfGh1vkrU9AF3pi6GYZiKQ6+tu9XSx44FWlp6f9/QUDF1RmKj6QO4KP/3LTOBn+dRSKGfAPAFAPcF0K5IYTuLZRiGqXT02rpbLb1CBLtfBB2yNzb/9xWbfdYB0Gzb4/xtDsMwDBMpUing/vsLEU3ptPz8wQfhtqtCCEzoE9FQyLV6ANhstZ8QogvSsQ8AeCGbYRimmrBbk2fKJkhNf5Du/UcO+2rr+Ef61BaGYRgmipjkLUE6Dbz0UjjtqTCCXNM/RPf+oOVeEi1m7VDjBiK6BsA1ADB8+HBvWsYwDMNEA16T95XYpeEVQvxWCDFBCDFh8ODBYTeHYRiGYWJDkEJ/v+59X4d9tWTM+3xqC8MwDMNUHUEK/e26904qurZ9h09tYRiGYZiqI0ih346C5j7Saici6gPgmPzHNr8bxTAMwzDVQmBCP59uV/PQONNm1/EAkvn3r/naKIZhGIapIoJ25FuW/3siEZ1isc+l+b85AMv9bxLDMAzDVAdBC/0HARzIv/+xcSMRjUA+HA/AY3Z59xmGYRiGcUegQl8IsQ0FYX8REf2BiE4hosFE9AUAz0HG8+8D8F9Bto1hGIZhKp2gC+4AwE8AHA/gqwCm5V969gGYJoR4M+iGMQzDMEwlE3hyHiH5DwCXAFgFmZK3C8AWAHdDltxdGXS7GIZhGKbSIelUH0+I6CMA//T4tINQnFOACR++J9GE70v04HsSTby+L8cJIUpKSRtroe8HRLROCDEh7HYwBfieRBO+L9GD70k0idJ9iV3ufYZhGIZhSoOFPsMwDMNUCSz0e/PbsBvA9ILvSTTh+xI9+J5Ek8jcF17TZxiGYZgqgTV9hmEYhqkSKlLoE9FkIvozEW0jooNEtJmIfkNEx3tw7loiuoGI/kZEHUS0l4g2ENEPiOgwL9pfKRBRPRF9iYjuJqJX879Xhog+IqJniWgWEdWXcf55RCQUXr/28v+KM0Q0QvE32+d8Ntvr8HPiAiJ6R/G+aK+ZLs8/U/G8VVXvhCQfJ6KriOhOIvo7EXXlf4uDLs5zFBH9lIjeIKIDRLSdiF4goquJiDxq61giepCI3s3LtXYi+iMRfcrViYQQFfMCQAAWABAWr70ALizj/AMArLM5/9sARob9O0TlBWCPzW+lvf4XwAklnn+ewvkFgF+H/VtE5QVghOJvtq+Ma/Bz4v43e0fxvmiv8S7PP1PxvMvD/i0C/t3tnoeDiueYAGCbzXmeBNCnzHb+B4C0xflzAG5RPVelafpzIdP7AsDDAE4DMATAFyEfqkMBLCGiE0s8/xLI0r85SIEzEsCxAL4N4CBkeuEniKiuxPNXGodBZltcDOAyyN9nIIAGAHdBdtiTADxFRIeWcZ1389eyev1nGeeuZC6E9W92VBnn5efEPZ+AfR8+HMDW/L7/K4R4tYxr2V3nUpvjKp33AfwJwIuqBxDRIABPQMqZHQBmAPgY5Lh2Z363zwH4VamNIqJzANwDoBZAC4Dz8tc7C7JeDQGYR0RXKJ0w7JmWhzO2oyDz9gvIkrxk2D5St/2REs7/BRRmVnNMtk/Vbb8u7N8jCi8AvwZwlM32m3S/2fdKOP+8/LHvhP2/xuWFYs3mUz6cn58Tf+7bubrfbW4Jx8/Ujg/7f4nSC3Ki80UAR+u+08YVR00fwM/y+2YBTDTZ/hvd9lNKbOPf8+fYCqC/YVsd5ERAAHgPQF+n81WSpn8lZIU+APgvkf9FNIQQWwD8Lv/xEiIa4vL81+b/fgjgduNGIcTDkD++ft+qRgjxLSErK1rxc8jZMQB8PoAmMf7Dz4k/fCX/VwBYFGZDKgkhxF4hxONCiA/cHktEtQC+lv/4uBDiZZPdbgGQgfSf+0YJ1zgdcvkAAP6fEKJDv10Ikc5fA5DWtC84nbOShP5F+b9vCSHWW+zzaP5vAgo/jgYR9YM0qQDy5mYczj+GiEapnr9aEUJ0A3gr//GYMNvClA8/J/5ARH1RMLs/L4R4L8z2MD38G4Aj8u8fNdtBCPERgBfyHy8u4RoX6d6bXgPASgAHVK9RSUJ/bP7vKzb7rIM0swDAOBfn/gSAPgrn129zc/5qRls73lPuiYgoQURhlIuONR6urfNz4g8XoyBcHvTihHmv9VovzlXF6PuuSn8flvcBKOUa7wsh2s12yE+um03aZEpFCH0iGgrppAcAm632E0J0AUjlP452cQn9vpbnhywPbHYMYwIRjYX0tQAAM9OYKkOIqBXSuzVDRDuI6EkiupyIkmU3tHL5dT40rysfAtRMRE0lLH1p8HPiD5ppvxPAH8s9GRG9BulgmyaiPUT0FyL6JhH1cTqWKULruznYV3stp79r+9s9T/pr/AsR2cr1ihD6kGULNT5y2PfD/N8jfTj/h7r3bs5frczP/xUoL01lPaSWqQn4gZAes78H8FwJs+tqYQwKfjB9IKMqbgbwBhFdUML5+DnxGCIaDNmXAWCpEGKvB6cdC+kJDkhHtk9BRtOs4+UWV2j9fVd+qdKKcvq7dg1VudYHBQXYlEoR+ofo3jslVOjM/3UTIqZ6/k7d+3JC0CoeIpqDwvrvXUKIjSWcpgMyFOYCAKMA9IV8SL4IYG1+n3MAPM4afw85AKsAXAXgFAD9Ifv3OAB3QC5/DQDwJyJya3rn58R7LgOgLVmVY9rvBPAA5FLBiZAT5f4A/h1yTRgATgbwZyI6vIzrVBNaf1eVOYD7/u75NXj9kwmcvBb5k/zHjQC+W8p5hBC3m3zdBRkDvgIyP8AUAJ+ENJE+UMp1KgkhxLsoaI56mgE0E9FfADwGKRR+AakFMuExI//3AwBPl3oSIcQSyPwJeg4CeBbAs0T0c8h8Fifk/84r9VpMtKkUTX+/7n1fh321tK9u0oyqnl+fUrasNKaVChGNB/AIpCn+PQCThBCd9ke5RwiRhQwJ07xa1RJXVDlCiKWQia0A4FwichNVwc+Jh+STiJ2R//j7fJ/2i7mQzyPAz4oqWn9XlTmA+/7u+TUqRehv170f7LCvtn2H7V6lnV+/zc35q4L8IPYk5DriRwDO9zP8SAixA8BL+Y8Nfl2nAlmme9/g4jh+TrzlK7r3C/28UD7ee1X+4wn58EvGHq2/93eIGiqnv2vXUJVraVSJ0G9H4R8dabVT3jtV01zaXJxfv6/l+Q3b3Jy/4iGiYZDmycGQ4XmfE0K8EcClNQeX/gFcq1LQOx71d3EcPycekS/Sopn2NwohWgK4bKn3vVrR+m4SwHCb/crp79r+ds+TfvtbQoic3Y4VIfTz2fe0OMUzbXYdj4KH92suLvE65Fqx0/n129ycv6LJeyA/DflgdAK4SAgR1O9zdP5vR0DXqwSO1r3vcHEcPyfe8a+QKZMBn7V8HaXe92pF33dV+vt7QojtNvvZXWMYEX3MbIe8lWGcYX9LKkLo59FMkicS0SkW+2hZrXKQ+fmVEEIcgHR4AYAv2phytPO3CiGc4iqrgrwn8CrIeNMMgEuFEKsDuvZgSCc+oDApZJz5ou698u/Gz4mnaKb9HICH/L5Y3gqqOXi+mb+XjD0vAtidf29aqCgfLnxu/uMTJVxDv9RmVQzpQgDacozjNSpJ6D+IgtPWj40biWgEgGvyHx8TQnxo3MeBu/J/jwJwvcn5L0UhK+Bdxu3VSD596DLI3yUH4CtCiJX2Rymfe5BdMpF8trHfoeAA4/vAGQfyiazstk8FcEn+41+EECm7/U3g56RM8v16Sv7js1aZ2Fyc7zC7MLz8UsLPUVj65GdFgXwmvHvzH79IRGeY7DYPsiiOgKyU5/Yaf4fMJAsA3yOiI/Tb8+PcvPzH96GizIZd5cjLF4Dvo1CJ6g+QcciDIfPsb85/vxfAiSbHPqAda3P+p/L7dAP4PwCOg3xQroOccAgA/wBQF/ZvEfYLchnlcd39uBEyftTq1c/kHM/DoooegMmQvhw/BXA+5NLBEfm/lwF4VXft5wEkwv5NovCCdKB8GMB0yIRGR+afkX+FTJCUy/9m+wCcZnI8Pyf+36NLdX33K4rHWN4XSGfMnZBVL78Auf7bP39PLoK0zmjX+18Ah4b9GwT8e38CwETd6978b9Fl+H4igD6GYwflxyGRf7augJzwngCZ90L7XX9rce0Run3mWexzTv5ZEpDm+0/lr3sGgGd0x1+h9P+G/YN7fPMIwALdj2B87QVwocWxKoPZQMhZl9X53wYwMuzfIQovQ2dWeb1jco7nbbZNVjzvchjKUVbzC3Kt1uk32wrgXIvj+Tnx/x5pk+V9AA5RPMbyvkAKfZVnZS2AYWH//yH83s8r/j4CwAiT4ycA2GZzzJMwTBZ0x+rHyXk2bfwapGe+2flzAG5R/X8rKjmPkL/OfxDRcsgyhuMAHA45E1sF4GdCiE1lnH8nEX0SUmO5AjKzVRLSivAogF8Kb9JkMs6sAfAtyDX7UwEMgcwk1wV5v18BsFAIUXJCkwrlq5DVwc4EMBRSY6iB1ATXQ06S/qecfszPSekQ0ZEolJl+TAix325/Rd4G8HXIZ2Uc5LNyJKT2uA1ygvYHyMqIfuYCqEiEEOuI6FTIJGMXAxgG6bDcCuB+AA/kZVM517iXiF6FTJz0Kch7uBNyonaHEOJ51XNRmW1hGIZhGCYmVJIjH8MwDMMwNrDQZxiGYZgqgYU+wzAMw1QJLPQZhmEYpkpgoc8wDMMwVQILfYZhGIapEljoMwzDMEyVwEKfYRiGYaoEFvoMwzAMUyWw0GeYmEJEopxX/hyf0n03IuR/iWEYn2GhzzAMwzBVAgt9hokvh1m8LtTt802b/RiGqTIqqsoew1QTQoh9Zt8TUafuY5fVfvlzPA9ZkpphmCqANX2GYRiGqRJY6DNMFePkyKfbNpOIEkT0LSJaR0R7iGg7Ef2ZiM4yHPNZIlpJRB8QUScRtRDR1xXaUktE1xDR00S0jYjS+b/LiOhiD/9thqla2LzPMIwKtQBWArjA8P0FAD5NRBcLIVYR0f8B8CPDPqcB+C0RjRRCfN/s5PkJx3IAYwybhgD4AoAvENFCAF8VQnSX968wTPXCmj7DMCp8H8CnAPwXgBMADAJwEYD3AdRBCvUpkAL/fgDjARwJYAKAv+bPcRMRGYU6iOgIAM9BCvz3AVybv8ZAACcD+CmALICvAGjy5b9jmCqBNX2GYVQYAeBLQoiluu+WE9EBAM8CGA7g9wB+IYT4jm6fnUT0RQBbABwOKbgbDef+MYCRADYDOEsI8aFu2y7IycKbAO4F8B0i+rUQYqtn/xnDVBGs6TMMo8Jqg8DX+AuAj/LvDwK4xbiDEGIngKfzH8/QbyOiQwFcnf/YaBD4eu4DsAlymWGKq5YzDNMDa/oMw6jwlNmXQghBRFsADAbwsk144Kb836MN358FoF/+/Yv5SYAV6wEcD7lkwDBMCbDQZxhGhQ9stnW62Kfe8P1o3fuUYlsGK+7HMIwBNu8zDKNC1qN9jImAjiihLX1KOIZhGLCmzzBMuGjLAQJAjRAiF2ZjGKbSYU2fYZgw2Zz/S5Ae/AzD+AgLfYZhwmQ1gHT+/dQwG8Iw1QALfYZhQkMIsRvAgvzH7xPReLv9iWgIEQ3wv2UMU5mw0GcYJmy+D+AtAIdChu3dRkRnENEgIjqSiD5BRNOJ6A8A/gkZtscwTAmwIx/DMKEihOggok8D+COAMwHclH9ZkQmkYQxTgbDQZxgmdIQQW4nokwAuAXAZpPDX4vG3A3gdMvvfo0KIt8JpJcPEHxJChN0GhmEYhmECgNf0GYZhGKZKYKHPMAzDMFUCC32GYRiGqRJY6DMMwzBMlcBCn2EYhmGqBBb6DMMwDFMlsNBnGIZhmCqBhT7DMAzDVAks9BmGYRimSmChzzAMwzBVAgt9hmEYhqkSWOgzDMMwTJXw/wEQTar35upKPQAAAABJRU5ErkJggg==\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -1137,9 +787,9 @@ "Estimand name: backdoor1 (Default)\n", "Estimand expression:\n", " d \n", - "─────────(Expectation(Outcome|w4,w8,w1,w0,w6,w7,w3,w5,w9,w2))\n", + "─────────(Expectation(Outcome|w1,w7,w5,w0,w6,w9,w8,w4,w3,w2))\n", "d[Action] \n", - "Estimand assumption 1, Unconfoundedness: If U→{Action} and U→Outcome then P(Outcome|Action,w4,w8,w1,w0,w6,w7,w3,w5,w9,w2,U) = P(Outcome|Action,w4,w8,w1,w0,w6,w7,w3,w5,w9,w2)\n", + "Estimand assumption 1, Unconfoundedness: If U→{Action} and U→Outcome then P(Outcome|Action,w1,w7,w5,w0,w6,w9,w8,w4,w3,w2,U) = P(Outcome|Action,w1,w7,w5,w0,w6,w9,w8,w4,w3,w2)\n", "\n", "### Estimand : 2\n", "Estimand name: iv\n", @@ -1187,18 +837,18 @@ "Estimand type: nonparametric-ate\n", "\n", "## Realized estimand\n", - "b: Outcome~Action+w4+w8+w1+w0+w6+w7+w3+w5+w9+w2\n", + "b: Outcome~Action+w1+w7+w5+w0+w6+w9+w8+w4+w3+w2\n", "Target units: ate\n", "\n", "## Estimate\n", - "Mean value: 1.9992398846781603\n", + "Mean value: 1.998912214794943\n", "\n", - "Causal Estimate is 1.9992398846781603\n" + "Causal Estimate is 1.998912214794943\n" ] }, { "data": { - "image/png": "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\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAoIAAAGTCAYAAAC8knO2AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAACbe0lEQVR4nOzdd1zT1/4/8NdJSCDsqeBArOKg4sS9BQcKyuyyVttbtba9rZ1f2/56W7t7a2+9djiubbV2WVlOXDhb6xaNexQ3yt5hJef3R0ia8ElCgISV9/Px4AF8zudzziEivDnjfRjnHIQQQgghxPaImrsDhBBCCCGkeVAgSAghhBBioygQJIQQQgixURQIEkIIIYTYKAoECSGEEEJsFAWChBBCCCE2yq65O0BIU/P29uYBAQHN3Q1CCCHELHfv3hVcq6ioQEFBAVQqlaCsffv28PT0RGFhITjnKCgoQFlZGTNUNwWCxOYEBATg+PHjzd0NQgghxCxLly5FYWEhAECpVGLv3r34/fffBfd5eHjg559/xpQpU/SeWblypdG6aWqYEEIIIaQFCw0NhUQiQUFBAdasWWMwCAwICMCLL76IKVOmAIA2CKwLBYKEEEIIIS1YcHAwnJ2dsWrVKty6dUuvjDGGcePG4YknntC77ubmZlbdFAgSQgghhLRQFRUVWLhwIV588UWUlZXplTk7O+OJJ57AuHHjIBKJ9II/zShiXWiNICGEEEJIC3Tt2jU8/PDDOHHihKCsR48emDFjBpycnAAAEokEoaGh2vLg4GAAQFpamsk2GOfcgl0mpOULCQnhtFmEEEJIS7Z+/XrMnTsXxcXFetft7Ozw0UcfYdKkSdi7dy8KCwvh5uaG0NBQbfBXG2PsBOc8xFAZjQgSQgghhLQQCoUCCxcuxKpVqwRlXbp0wa+//ophw4YBAPr169fo9igQJIQQQghpAc6fP4+HH34YZ8+eFZRFR0fj22+/xe3bt7WpYeoaCTQHbRYhhBBCCGlGnHOsWbMGgwcPFgSBUqkUX331FRITE3H79m1s3rxZmxqmsLAQmzdvhlwub3DbFAgSQgghhDST4uJiPPHEE3jyyScFu4K7d++Ow4cP47nnngNjDGlpaaiqqtK7p6qqqs4NIabQ1DAhhBBCiBXI5XKkpaUZncZNT0/Hww8/jMuXLwuenTlzJp5//nns378fmzZtgpubm9Ek0eYmjzaEAkFCCCGEEAuTy+XYvHmzdgRPM4178+ZNXL58Gbt378bOnTtRXV2t95xMJsPXX3+NQYMGYcuWLXrPG2Nu8mhDKBAkxEyccxQXF6OoqAhlZWVQKpXN3SVCCBEQiUSwt7eHk5MTXF1dYW9v39xdsknGpnEPHjyITZs24cKFC4JnHnzwQfz2228ICgrC0qVLBc8bUjt/YH1RIEiIGTjnyMrKQmlpKTw9PeHr6wuxWAzGWHN3jRBCtDjnUKlUKC8vR0lJCW7cuAFfX1+4uro2d9danbqmdetiaATv9u3bSEhIQEFBgaBs7ty5WLp0KRwdHY0+r6GZJrbErmEKBAkxQ3FxMUpLS9GlSxeIxeLm7g4hhBjEGINYLIaTkxOcnJzg5uaGW7duQSKRQCaTNXf3Wg1j07oA9IIuU8Gi7po+lUqFP//8E2lpaVCpVHptSaVSREZGomPHjli1apW2DmNrAt3c3LBw4UKLfa0UCBJihqKiInh6elIQSAhpVRwcHODh4YH8/HwKBOvB1O5cTaBnKFhMSkpCUlISGGPQnNxWWlqKlJQUXLlyRdCOn58f4uLi4OXlpa1DE3CGhobq1Q80fhrYEAoECTFDWVkZfH19m7sbhBBSb25ubsjIyGjubrQq5uzONRQsamiCwOvXryMxMVFwTBwADB06FBMnToSdnX4opgk4NaN+jZmeNgcFgoSYQalU0mggIaRVsrOzo81t9WRqWlbD1Bo+lUqFAwcOYP/+/dqgUMPBwQEzZsxA7969jT6vqTs4ONjigV9tlFCaEDPRxhBCSGtEP7vqLzQ0FBKJRHC9sLAQS5cuhVwuN5qypbi4GD/88AP27dsnCAI7deqEBQsWoHfv3nBzczM6Xd+YdDD1RSOChBBCCLFJhjZ7AH9P++qu9dPQrOPr168fTp8+rTc9fPXqVSQlJQlOCAGARYsW4b333tMLMGuvMwSssw7QFAoECSGEEGJzDG322LhxozYFDwBBEKhRVVWF48ePQyaTwc7ODiUlJdi7dy9+//13wb1OTk5YsmQJnnnmGUGZZtrX2usATaFAkBBCCCE2x9Bmj/qupVQoFCgtLcXOnTtx+vRpQXlAQADi4+MxcuRIg883NlehJVAgSAghhJA2zVDA1ZjzeTUuXryIlJQUlJeX611njGHs2LEYM2YMRCKRXtoZ3T6Zk6vQ2mizCCGkRbh+/To++OADjBkzBp06dYK9vT1cXFzQvXt3PPzww1i3bp3BdTe2as2aNWCMtfmNAOPGjQNjDHPmzGm2Puzbt0/7Wl+/fr3Z+kEaRhNwaQI/Tb6/xqiurkZqaip+/fVXQRDo4uKC2bNnY9y4cRCJRNo2azOVq7ApUSBICGlWVVVVeOWVV9CzZ0+8/fbbOHjwIO7cuYPKykqUlJTg2rVr+O233/DEE0+gS5cu+O6775q7y6SNePfdd8EYQ0BAQHN3hViRqXx/DZGbm4tvv/0WR44cEZT17t0bzzzzjOB7ytAuYHNyFTYFCgQJIc2mrKwMU6ZMwX/+8x9UVlaie/fu+OKLL5Ceno6srCzcuXMHhw8fxuLFixEQEICcnBwsW7asubtNCGlFLBlYyeVyrFy5EpmZmXrX7ezssGTJEvzyyy9wd3fXKzO2C9hYipimTB0D0BpB0sQYY5MBzAYwHIAvgDIAdwAcBpDEOd9u5DkXAC8BiAPQFYASwGUAPwP4mnNuuT/3SJN57rnnsGfPHgDA/Pnz8eWXXwpyd3Xo0AFDhw7FokWLsGTJkkZP6ZDWZd++fc3dBYwbN87o7lHSssnlcoMpYOqrsrIS27dvx8mTJwVl7u7u2L59O4YOHQoAEIlESE1NhUKhAADBySEaTXWEXF0oECRNgjHmBOAHADG1ihwAeAIIBtADgCAQZIx1BbAbwAO1igbXvD3OGJvIOc+3dL+J9aSlpWHNmjUAgMjISKxYscLk/VKpFG+++SaioqKs3zlCSKunWRvY2CAwKysLGzZsQHZ2tqCsd+/emDVrljYI1KiurtZ+rFAoDG4CaQmpYwCaGiZNgDEmBbAF6iCwEsASAEMA+ADoACAMwHIAuUae3Qx1EKgA8E8AnaAeFXwXgArAIADrrfxlEAtbsmQJAPVfz/WZ7g0KChJcO3PmjHajiY+PDyQSCTw8PDB06FB88MEHKCgoMFrfnDlzwBjDuHHjjN5z/fp17WYBQyNUlZWV+PLLLzFmzBh4eXlBIpHA29sbvXv3RlxcHFavXm1wo0tj+t1QW7du1X4tcrnc5L2pqanae8+cOaO9zjnHkSNH8Oabb2Lo0KHw8PCARCKBl5cXxowZg2XLlmlHQwyp/ZofOHAAsbGx6NixI+zs7PSCfVObRRraD83mj8WLFwMAbty4of06NW+67Zm7WSQxMRGRkZHw9fWFvb092rVrh8mTJ+PHH380GozU/t6qqKjAp59+ir59+8LJyQnu7u4IDQ1Famqq0XaJYY1dG8g5x8mTJ7Fq1SpBECgWizF16lTMnDkTERERdbZrbBNIcHAwFi5ciHfeeQcLFy5s8iAQoBFB0jQWARgH9TTwZM557YybmQCMbZOaB+DBmo9nc8436JQtZowpAHwKYCJjbBrnfKvluk2spbi4GLt27QIATJgwoVGL9U+fPo3+/fsLrhcUFODo0aM4evQovv32W+zcuROBgYENbseY4uJiTJgwAcePH9e7npubi9zcXFy8eBGJiYkICQnR62dz9Xvy5Mnw9vZGTk4Ofv75Z3z88cdG7/3pp58AAH369EHfvn211zdt2mRwZDYvLw8HDx7EwYMH8f3332Pnzp3w8fEx2Z8vv/wSCxcu1CbwrQ9L9qMxysvL8eijjyIlJUXvenZ2Nnbu3ImdO3dizZo1SE5OhouLi9F6iouLMWbMGBw9elTv+p49e7Bnzx6sXLkS8+bNs8aX0CY1Zm1gRUUFtmzZYvCPJS8vL8TFxaFDhw6YPn26IHhrKZtAzEUjgsSqGGNeAN6o+fR9A0FgXRbUvD9VKwjU+AJAdq17SQt3+PBhbeLW0aNHN6ouxhjCwsLw1Vdf4ffff8e1a9eQk5MDuVyOZcuWwd/fH9evX8cjjzxilXVen3zyCY4fPw6xWIy3334b6enpyM7ORlZWFk6ePImvv/4aY8aMEaR5aa5+29nZ4aGHHgIA/Prrr0brLisr0wY2M2fOFNQRFRWF1atX49ChQ7h+/Tqys7ORnp6ODz74AF5eXkhPT8f8+fNN9uXixYt46aWXMGbMGKSlpSE7OxvXrl3DCy+8YPbX0pB+jB49GsXFxXjjDfWPJn9/fxQXF+u9rVy50qw+AOq1rprX6tFHH8WxY8eQk5OD9PR0PP300wDUo0SzZs0yWc+LL76Iixcv4osvvsBff/2FnJwcbNu2DQ88oF4V89JLLyErK8vsftk6Y+f41iUzMxMrV640GAT27dsX8+bNg7+/P6Kjow2O4LWUTSDmohFBYm2zoF4HWAnA9CKwWhhjDwDQzAMmGLqHc17FGNsI4GkAoYwxR855kyeba+u53CwdQGVkZGg/7tWrV6Pq6tu3r3Z0UZeXlxf69OmD+Ph4BAUF4eTJk0hLS0NYWFij2qtNM2X3wgsv4L333tMr8/HxwYABA/Dss8+2qH7PnDkT33zzDa5fv45Dhw4ZPPVg48aNKC0tBWMMjz32mF7ZtGnTMG3aNMEz3t7e6NevH6Kjo9GvXz+kpKTgypUrRkc079+/jwkTJmDHjh3aBfXe3t7awKcuDe2HWCyGs7MzpFIpAPX/X2dnZ7ParO3EiRPalEZz587FqlWrtGVeXl743//+B09PT/z73//Gxo0bsX37dkyZMsVgXbdu3cLBgwcxbNgw7bXw8HCkpKSgb9++KCsrQ0JCgsHvJ9J4nHMcO3YMO3bsEJwwIpFIMHXqVPTv3187XW9sGrelbAIxF40IEmvT/MQ7xjkv0FxkjIlZ3dHTQJ2PhQmbhGUO+DtwJC1Yfv7f+3qs/Veyr6+vNoiyRqJWzS+MDh06WLRea/Z7xIgR6Nq1KwDg559/NniPZlp41KhR8Pf3r1f9QUFBGDBgADjn2l3hxvz73/82uquyserTj4bSBIGOjo747LPPDN7z3nvvoV27dgCAb7/91mhdjzzyiF4QqBEcHKxdRnDs2LFG9th2mFqnauje9evXY9u2bYIgsF27dpg7dy4GDBgAe3v7OtfyBQcHIzIyUvuzzc3NDZGRkc2y/s8cNCJIrG1QzfvzNRs/XoU6fUw3AGCM/QUgGcBnnPOcWs/21Pn4LxNtZOh83BPAcWM3krZJpVJh/fr1WL9+PU6ePIns7GxBtn8AuHz5ssXb7t+/P86cOYPPPvsMQUFBmDJlivY0gbo0Z78fe+wxfPjhh/jtt9/w3//+Vy8Yy83Nxc6dOwEIp4U1KisrsWbNGiQlJeHMmTPIzc1FZWVlvfru4+ODQYMGGS03hyX60Ri//65e7RIWFmb0jxp7e3tERkbi22+/xcGDB43WNXnyZKNlgYGBSE9Px/379xvXYRvi5uZm1rq8W7duISEhweC9gwYNwpQpU7RprQx9bxkSHBzcYgO/2igQJFbDGJMB8K75tBLAQah3C+sKBPA6gFmMsXDOue6p3d46Hwv37f9Nd9GMVwO7S5qQh4eH9uPGLqAuLi5GREQEDhw4UOe91lis/e677yIlJQVZWVmYNm0avLy8MHbsWIwePRphYWHo06ePweeau98zZ87Ehx9+iJycHOzcuRNTp07Vlv3222+oqqqCRCJBfHy84NnMzExMnDgR586dq7MdU33XjEo2lKX60Rg3btwAYHg3u64HH1Tvebt//z4qKyu109K6/Pz8jD7v6OgIoH6jXLYuNDTUZN5RlUqFP//8E2lpaYLNSlKpFNOnTzf4/1cul7eaIM8cNDVMrEn3z+O5UAeBiQD6ArAH4A/gPahTwPgB2FiTOFrDSedj4TDJ33R/Mhpc6MMYm8cYO84YO24oF1Rjcc7b9Jul6QYAFy9ebFRdL730Eg4cOADGGJ5++mls374dGRkZyMvL0y7816xx083tZSldu3bFiRMnMHPmTMhkMuTm5iIpKQkvvfSSdkpPM7rWkvrdu3dvDBgwAIBweljzeXh4ODw9PQXPzpo1C+fOnYNEIsErr7yCvXv34ubNm8jPz9f2XbPu0FTfG7qY39L9aIySkhIAqHONoe5u4eLiYoP3iMXiOtujxNbmMxWslZaW4ueff8auXbsEQaCfnx/mz59v9I+4zZs315l6qTWhEUFiTbp/aEgBbAIQz//+SXYLwDuMsSKocwt2gTpdzOeW7gjnfBWAVQAQEhJCP0mb2bBhwyAWi6FUKk1OldWltLQUP/74IwDgjTfewIcffmj0PmPM2ehTVxDRvXt3/PjjjygvL8fRo0dx6NAh7Nq1C/v27cPp06cxZcoUbNq0SZtvzBL9toSZM2fi1KlT2LhxI8rKyuDo6IgbN27gjz/+0JbXdvXqVe2axa+//hpz5841WLcmQLKWltIPZ2dnFBYW1tmObrmpFDKk/uRyud5JHjKZDOHh4QgODjY4PZyRkYGkpCSDAfmwYcMQFhZmct2qJidgWxkVpECQWFPt32LvcsN/zn4JdYoZLwAz8HcgqPu8AwBjP2l1hxWs+1OfWISLiwvCwsKwY8cO7NmzBzdu3ECXLl3qXc+lS5dQUVEBANqUKIacPXvWaJmDgwMA01Nutc8VNVXXmDFjMGbMGCxatAgXLlzAuHHjkJWVhffff18bCFqi35bw6KOP4vXXX0dJSQk2bdqERx55BL/88gs453BxcUFkZKTgGd3E0sb6XlVVZbU1eS2tHwEBATh9+jQuXLhg8j7N9LWvr6/BaWGiJpfL6zxpQ/cemUyGiooKvVE9hUKBlJQU3Lx5Uy+Ru0qlwoEDB7B//37ByKqDgwOioqLMzmLQUnMCNgRNDRNrKoZ6bSCgDtDSDd3EOa/E3zt/dRfa6G4eMZUNVrdMcDoJaZleffVVAOofzv/85z/Nfu78+fPajzXBFADBTj+No0eP4tq1a0br8/X1BaAeYTI27WYozYs5evfujUcffRSA/hS4JfptCR06dNCe7qHZJax5HxMTY3Dq1py+JycnW30tmyX6odkAYOx5c4waNQoAsHv3bhQVFRm8p6KiQnvEmOZ+IqQ5Ek4TZBUWFupNw8rlcvz73/9GUlKS9h6FQmEwGblKpcLx48e1KVyKiorwww8/YN++fYL/5507d8YzzzxTr1RWLTUnYENQIEishnOuAqD5c7zAyGighiafiKvOtUs6H5taVa5bdsnoXaRFCQsL0ybY3bx5M5555hmTx0FVVlbio48+whNPPKG9pnsiieYXra6ysrI6c64NHjwYgPo0iu3bBUddIyMjw+QReHWtcdQEc15ef+9jskS/LUUz/btjxw7s27dPOwppbLew7vpOQ33PysrC66+/boWeWr4fmn+TnJycBq8hfOqppwCop/EXLVpk8J53331Xu9v3H//4R4PasQWmjmbTBIkN+QPjypUrWLFihcHjAUeNGoU5c+bA3d29XnVa45Si5kKBILE2TSoXjzryBmp+S+qOt5/U+Vj/RG99mrJyAOdN3EdamOXLl2Ps2LEAgJUrVyIoKAj//e9/cebMGeTk5CAzMxNHjhzB+++/j549e+Ktt97S+4Xt5+enHWH56KOP8NFHH+Hq1avIzs7Gtm3bMGrUKJw6dQo9e/Y02D4ATJw4UZsDcM6cOdiwYQNycnJw584drFmzBqNGjdIL4moLCgrCxIkTsWLFCpw6dUp7qsiRI0fwj3/8A1u2bAGgzhFnyX5bSmxsLOzt7VFVVaU9X9fX1xcTJkwweH9ISIh2Gv+FF17A8uXLcePGDdy7dw/r16/H8OHDkZ2d3aCp/vqwRD8GDlSnKi0vL8d7772He/fuobq6GtXV1WYfeTdw4EBtMLh8+XI8/vjjOHHiBPLy8nDmzBnMnz8fn3zyCQBg+vTpRpNJE9NHszXk3GClUoldu3bhp59+Epz17eTkhMcffxxhYWEmN+loRo1ru3LlSr360pLRGkFibRsBzIF6B3AIAEE2VMaYPf4O5tI11znnfzHGzkM9XRwHQHAoKmPMDup1hQCQ1hynipCGc3Jyws6dO/Haa69h+fLluHr1KhYuXGj0/vbt2+OVV17Ru7Z8+XKMHj0aBQUFeOutt/DWW29pyxhjWLJkCc6cOYNLlwwPFkskEvzvf/9DVFQUsrKyBOvNevToge+//97g6RuAehfn7t27sXv3bqP9njx5Mt5++22L9ttS3NzcEBERgcTERG0qlEceecToL0c7OzusXr0aERERKCoqEoxcSqVS/PDDD9rAzFos0Y8hQ4Zg2LBhOHz4MN5//328//772rLZs2djzZo1ZvXl66+/Rl5eHlJSUvDTTz9pp9d1hYaGajcIEcOM5f2TyWT1XpNXUFCAhIQE3L59W1DWtWtXxMTEmNy0o1mfaCz9DK0RJMR82/B3wuf3GWOGvudeA6BJLPdrrbLlNe8HMsZiDTy7EEC7WveSVkQqleK///0vLl26hMWLF2PUqFHw8/ODVCqFs7Mzunfvjocffhg//vgjrl+/LjivtU+fPjh+/DhmzZoFX19fSCQS+Pn5ISoqCnv27MHLL79cZx+mTp2KAwcOIDIyEp6enrC3t0dgYCDefPNNHD9+3OSpISdOnMAnn3yCyZMnIzAwUHt0WceOHREZGYn169cjNTVVsN7OEv22lNrTwLWPlKstLCwMf/75J6Kjo+Hp6QmpVIrOnTvj8ccfx+HDh/Hwww9bs7sW7ce2bdvw8ssvo1evXtqNQ/Xl4OCA5ORkJCQkYNq0aWjfvj0kEgl8fHwwceJErFu3Drt27aLdwnUIDQ01OAJX3+ngCxcuYMWKFYIgkDGG8ePHY9asWUb/LSQSCWJiYrSnh7S2c4MbglFOImJtjLEZUJ8ewgCkQp078BKA9lCfEfxyTdkJAMM551U6z0qhniJ+EOp8ga8CSAEggfqEkn8BEAPYxTmfZE5/QkJC+PHj9Tt85MKFC+jdu3e9niGEkJaitfwMq50Kpj6qqqqwa9cuHD16VFDm4uKC2NhYvfW5tRnapaxZm1j73OCWfGScIYyxE5zzEENlNDVMrI5zvpExthDqtDDhNW+1pQOYoRsE1jxbyRiLBLAbwAMAvq5503UCQNMMQRBCCLGa4OBgpKWl1TsQzM3NxYYNG3Dv3j1BWWBgIKKiouDk5GTgyb8ZWpaiCfbqSmnTmlEgSJoE53wZY+x3AC8CGAfAF0AZADnU08Gra9LIGHo2gzHWH8BLUK8VfACAEuodyT8D+Kp2AEkIIaTlMpUvsL7r786cOYMtW7YIzgEWiUQICwvDsGHD6jz/29RUb2s6N7ghKBAkTYZzfhLq6dyGPFsM9ZTyexbtFCGEEKsxFPAB0JtuLSwsRFJSEk6dOqU99tAclZWVSE1NxalTpwRl7u7uiIuLQ6dOncyqS9MvW0SBICGEEEIsrvb6Ok2CaDs7O4OpYDIyMpCRkSG4bkhWVhY2bNgAQ2fHBwUFITIy0uyzrM05ZrIto0CQEEIIIRZnLEF0ffMB6uKc49SpU9i2bZsgCbhYLMaUKVMQEhJSr+COc47k5GQAaNNTwMZQIEgIIYQQi7N0rr3y8nJs2bLF4BncXl5eiI+P1x4ZaYixPIWAOhjUnFBja8Eg5REkhBBCiMVZMtfe3bt3sWrVKoNBYL9+/TBv3rw6g8CFCxea7JPmODtbQ4EgIYQQQizOEufxcs5x5MgRfPvtt8jLy9Mrk0gkiIqKQnR0NOzt7c3qi7Gk1Rpt6cQQc9HUMCGEEEIaTXeHsEwmE6RzqS+FQoGNGzfi4sWLgrJ27dohPj4ePj4+ZtWlORtYM+2bnJwMQwdqtKUTQ8xFgSAhhBBCGqX2DuGGnAyi69atW0hISDA4QhcSEoLJkyebHNmrTbceTTBo6MQQW0wjQ4EgIYQQQkwylQAaMLxDuCFUKhUOHTqEtLQ0wYidvb09IiMj0adPH4PPSiQSMMYMjkTWTiVjCyeGmIsCQUIIIYQYZSwfIIAGnwZiSElJCZKTk3Ht2jVBWYcOHRAXFwdPT0+Dz8pkMoSHhyM1NdXs9tr6iSHmokCQEEIIIUYZyweYlpaG4OBgyOXyRreRkZGBxMRElJSUCMqGDRuGsLAw2NkJQxZNAKgJ6JKSkgzW39ip6raMAkFCCCGEGGVstK+wsFA7WthQKpUK+/fvx/79+wVlMpkMUVFR6Nmzp8FnZTIZXn/9db1rxnIF2uImEHNR+hhCCCGEGGUsiHJzc2vU2sCioiKsXbvWYBDYuXNnPPPMM0aDQMDwKJ+h9DC2ugnEXDQiSAghhBCt2htDAgMDcfr0acEO28DAQBw/frxBbVy5cgXJyckoKysTlI0ePRrjxo2DWCw2WYehY+RoE0j9USBICCGEEADqIDAlJQUqlQqAevr3+PHj6Nq1K/Ly8gTBYX0plUqkpaXh0KFDgjInJyfExMSgW7duZtVlKA8gQJtA6osCQUIIIYQAAFJTU7VBoK6MjAzExMRoA6ylS5fWe0o4Pz8fiYmJuH37tqCsa9euiImJgYuLi9n10bo/y6A1goSQFuHUqVN45ZVX0K9fP3h7e8Pe3h6dOnVCaGgovvjiC+Tn59dZB2MMjDGsWbPG+h0mBq1Zs0b772Atc+bMAWMM48aNs1obtsrU7trk5GQsXrwYS5curXe6mPPnz2PlypWCIJAxhgkTJmDWrFn1CgJp3Z/lUCBICGlWZWVlmDNnDgYNGoT//Oc/OHPmDHJzc1FZWYk7d+5gz549ePnllxEYGIgffvihubtL2rh3330XjDEEBAQ0d1daHM1UbH2CwKqqKmzduhW//fYbysvL9cpcXV0xZ84cjBkzBiKR+eGIm5sbIiMjafrXQmhqmBDSbAoLCzFlyhQcPnwYADB06FC88MILGDZsGNzc3HDnzh1s3boVS5YsQW5uLmbPno3MzEz83//9XzP3nJC2SSqVNvqMYI2cnBwkJCTg3r17grIePXogKioKjo6O9arTzc0NCxcutEj/iBoFgoSQZvPMM89og8CXXnoJS5Ys0RsZ8PLyQt++ffHkk08iLCwM586dwxtvvIH+/ftj8uTJzdVtQtosYxsw6uvMmTPYsmWLIKgUiUSYOHEihg0bVu/lAzQdbB0UCBJCmsWOHTvw66+/AgAiIiLwn//8x+i9vr6+SE1NRVBQEEpKSjBv3jxcu3bN4EkDhJCGkcvljT4vuLKyEtu2bUN6erqgzN3dHfHx8ejYsaPZ9clkMigUCkoDY0W0RpAQ0iy++OILAOrF4kuXLq3z/s6dO+OVV14BANy8eRMJCQkm76+qqsLnn3+OAQMGwMXFBe7u7hg/frzRI6g00tPT8dRTTyEwMBCOjo6QyWTw9/fHsGHDsGjRIpN5027duoXXXnsNffv2hZubG2QyGQIDA/Hcc88hIyPD6HMBAQFgjOHdd98FoN5wMXbsWHh7e2tfn9deew2MMbRv3x5KpdLk1/B///d/YIyhXbt2qK6uFpTn5eXh3XffxeDBg+Hh4QEHBwcEBARgzpw5OHPmjMm6ASAxMRHjxo2Du7s7XFxcMGDAAHz++eeNDiI0Kisr8dlnn6Fv375wdHSEt7c3QkNDkZycXOeznHMcOXIEb775JoYOHQoPDw9IJBJ4eXlhzJgxWLZsmcENEfv27QNjDIsXLwYA3LhxQ7vpRfM2Z84cvWfOnDmDDz74AGPGjIGPjw8kEgk8PDwwdOhQfPDBBygoKLDEy9Fk0tLSGvX8/fv3sWrVKoNBYFBQEJ555hlBEFjXzt/XX38d77zzDhYuXEhBoJXQn9OEkCZXXFyM3bt3AwDGjh1rdt6wp556SvuLOjk5GY888ojB+yorKxEWFoYDBw7oXd+3bx/27duHf/7zn1i2bJnguXXr1uHJJ58UBFq3bt3CrVu3cOTIEVy8eBEpKSmCZ3/88UfMnTtXsCD+6tWruHr1Kr7//nv8/PPPiIqKMvr1cc7x0EMPYcOGDYKymTNnYsmSJcjKysLu3buNTo1zzvHLL78AAB5++GHBqOmuXbvw0EMPCYKUGzduYO3atVi3bh2++uorLFiwwGD9zz77LJYvX653LT09Henp6di8eTMee+wxo1+fOYqKijBp0iQcOXJEe02hUGDPnj3Ys2cPFi1aZPL5TZs2GXyN8/LycPDgQRw8eBDff/89du7cCR8fnwb38/Tp0+jfv7/gekFBAY4ePYqjR4/i22+/xc6dOxEYGNjgdqxNN3l0Q3HOcfLkSaSmpgr+8BCLxQgPD8egQYMEU8EikcjkekRKD9M0aESQENLkDh8+rA22xo4da/Zz/v7+8Pf3BwD88ccfRu/7+OOP8fvvv+OVV17B+fPnkZOTg71792LkyJEAgC+//BIrV67UeyY/Px8LFiyAUqlESEgIUlJSkJGRgfz8fFy/fh3btm3DCy+8AC8vL0F7GzduxKxZs1BeXo4pU6Zg+/btyMzMRHZ2Nnbt2oXRo0dDoVDgkUceMZmE99tvv8WGDRvwzDPP4OTJk8jJycGpU6cwbtw49O/fH0FBQQCAn3/+2WgdBw4cwK1btwCog0ddx48fR0REBAoKCjBs2DAkJSXh9u3byM3NxR9//IEZM2ZApVLhueeew/bt2wV1f/3119ogcPTo0di3bx9ycnJw7tw5vPTSSzhw4AA+/vhjo30zxz/+8Q9tEDh37lycPn0aOTk5+OOPPzBlyhR88sknBo8k07Czs0NUVBRWr16NQ4cO4fr168jOzkZ6ejo++OADeHl5IT09HfPnz9d7bvTo0SguLsYbb7wBQP29VlxcrPem+z3DGENYWBi++uor/P7777h27RpycnIgl8uxbNky+Pv74/r163jkkUcstu7O0jTnBDcmCCwvL0diYiI2b94sCAK9vb0xd+5chISEGF0PaCxdDa0HbDqspX6DEmItISEhvL7HIl24cAG9e/e2Uo9sz6pVq7S/iH/99Vc8/PDDZj8bHh6uDVKqqqr0Rrx0f9l8+umnggPpKyoqMHr0aBw7dgyenp64c+cOHBwcAKhHkmbMmAGxWIysrCx4enqa1R+FQoEuXbogOzsbzz//PL788kvBPdXV1Zg4cSL27duH8PBwbNu2Ta88ICAAN27cAAD861//0o561vbRRx/hrbfegouLC7KysrR91zV//nysWrUK3bp1w9WrV7XXOefo27cvzp49ixkzZiAxMdHgEV5PPPEE1q1bhwcffBByuVz7mioUCnTo0EEbRO7fvx9SqVTv2Q8//BD/7//9P7026+Pw4cMYPnw4APXmodrrRpVKJaZNm4YdO3YAUP8RsW/fvnq1cf78efTr1w9KpRKXLl0SjNa9++67WLx4Mbp06YLr16/Xq25d9+7dQ1BQEPLz87Fr1y6EhYU1uC5LMPQzrCH5AHXdvXsXGzZsMJjjs1+/fpg6dSrs7e3rXS+tB7Q8xtgJznmIoTIaESSkFZHL5Vi6dKk2qatcLm/uLjWI7i8Od3f3ej2re39eXp7Be3TXE+qyt7fHp59+qn128+bN2jLNCKWTk1O9+vTLL78gOzsbnp6e+OyzzwzeY2dnh/feew8AsH37dqPJsb28vPDWW28Zbeuxxx4DYwzFxcV6fdeoqqrSrp2sPUW7d+9enD17FmKxGMuXLzd6juuHH34IADh37pze6OWmTZu008mffvqpIAgE1GsT67MRoLa1a9cCAFxcXPD+++8LysVisXZtaUMFBQVhwIAB4Jxjz549jarLFF9fX23w19i1d9bS0CCQc47Dhw9j9erVgu9liUSCqKgoREdHNygIBEDrAZsYBYKEtBK1p3EKCwuxefPmVhsMWoKxEaeIiAijgc64cePg4eEBQH96uW/fvmCMoaioCHPnzkVmZqZZfdD8kh89ejSqq6tRUlJi8K1Xr17aPp84ccJgXRMmTDAYYGkEBARgxIgRAICffvpJUJ6amqoNjmtPC2v6GRwcDBcXF6P99PDw0K6d0x0515wN6+npidGjRxvsn52dHSIiIoz2vy6aNsLCwuDk5GTwnt69e6Nnz54m66msrMSqVaswZcoUdOjQAfb29nqbPo4dOwYAuHz5coP7CgAqlQq//PILoqKi4O/vD5lMpteOZq1nY9uxloaswSsrK8Ovv/6K7du3C46ia9++PebNm2dw7aQ1+0QahzaLENJKpKWlCXZlVlVVIS0trdX99awJxADUe2el7v3Gpm81QZchjDH06NEDR44c0U7HAkC3bt3w3HPP4auvvsJ3332HNWvWYMCAARg9ejTGjh2LsLAwODs7C+q7dOkSAPU6QXOPyMrOzjZ4vWvXrnU+O3PmTPzxxx9ITU1FQUGB3uilJjgcNGiQIFjS9DM9Pb1B/dRMk/bs2dNk/jdTr31dNG3UVUevXr20X09tmZmZmDhxIs6dO1dne42ZFi0uLkZERIRgQ5Kl27GEsrIyFBUVYfHixXrTroGBgSZ3wdem2a1fVFQkKAsJCcHkyZMhkUjMqkskEoExprcxi9YFNg8aESSklTD2y6S5f8k0hO7xXRcvXqzXsxcuXAAAdOzY0egvHWOjSRqagK6kpETv+rJly7BixQr07t0bKpUKJ06cwNKlSxEdHY127drhn//8J4qLi/WeacjrX1FRYfC6TCar89mHHnoIEokElZWVeil0dKeLa48GWqKfpaWlAMx/bRvCEm3MmjUL586dg0QiwSuvvIK9e/fi5s2byM/P12760GwaMpRax1yazTGMMTz99NPYvn07MjIykJeXp21HMz3fmHYaq6ysDIWFhdrRu8LCQiQlJWHx4sVmB4EqlUq727p2EGhvb4/4+HhERESYHQS6ubkhKioKM2bM0I4A0rFxzYdGBAlpJdzc3Az+Mm+NUynDhg2DWCyGUqk0uQO0tlu3bmlH8TS/zA3RBBTGaALA2gEFYwzz58/H/Pnzcf36dRw6dAgHDhzA5s2bcffuXXz11Vc4duwYDh06pD0BRVPHE088oV3jZk1eXl6YPHkytmzZgp9//hlPP/00AHU6HYVCAZFIZDCtjqafY8aMqddrrqEJzsx9bRvCyckJRUVFDW7j6tWr2inwr7/+GnPnzrV4HwH1a/Djjz8CAN544w3tukpD9zW34uLiRu1aLikpQXJyMq5duyYo69ChA+Li4szeWAUAMTExesEeBX7Nj0YECWklQkNDBX9xt9apFFdXV0yYMAEAsH//fpPJlnWtWbNG+3F0dLTR+0yNMnLOtWu2unTpYvS+gIAAPPbYY1ixYgVu3LiBl156CQBw5MgR7a5VAHjggQcAAH/99ZdZX4MlaEb89u/fj7t37wL4O6XMhAkT4OfnJ3imsf3UjOJeunTJZGBR3xFeQ23UVYexct1k2A899JDBe6qqqhq9Zu/SpUva0VJj7QDA2bNnG9WOJdSVfNyUv/76CytWrDAYBA4fPhxPPfVUvYLA+h4pR5oGBYKEtBLBwcGIjIxsM1MpmsBKpVJpPzblzp072l25/v7+iIuLM3rvli1bjP4C3L9/v3ano6lRRV12dnb417/+pf1cNxCZOHEiAHXqk5s3b5pVX2NNnz4dzs7O2s0KmiTTgOFpYd1+3r59G3/++We929RsUtEkZjakuroaW7ZsqXfdtdvYvXu30dG0CxcuGF0fqDuVbezfXzNyaozmjy1TAZQ57Rw9etRgAGUtZWVluH//Pu7evYv79++jrKyswSebKJVK7NmzBz/88INg9FQmk+HRRx/F5MmT633EI+e8xe6gtmUUCBLSigQHB2PhwoVt4sil8PBwxMfHA1BvtHjttdcEuxA1srKyMHXqVBQXF4MxhlWrVpn8JXTr1i18/vnngusVFRXa3IKenp6IjIzUlmVkZJg85UD3l7puUulZs2bB29sb1dXVePLJJ00GGQCMBjH14ejoqB0R/fnnn7F+/XoolUo4ODggJibG4DOTJk3Cgw8+CACYN2+e0dQ7xvo5ffp07caU//u//zP4Wv373//GnTt36vvlaM2ePRuAejrz7bffFpQrlUq8/PLLRp/X3WxjKL1OVlaWILdkbZp/25ycHKNr+3TXuBpqp6ysDM8++6zJdixJsw5QE5QqlUoUFBSgrKys3nUVFRXhhx9+MLgJxt/fH88880ydu7ZNaY1rmts6CgQJIc1m1apVGDx4MABgyZIlGDlyJH755RdkZGQgNzcXcrkcn3zyCYKCgrTTfh9//LHR49U0AgIC8MYbb+C1117DxYsXkZubi/379yMsLEybOuSjjz7SS8i8du1adOnSBS+99BK2bduGGzduoKCgAH/99RfWrl2rDbBcXFwQHh6ufc7JyQnfffcdGGPYs2cPBg4ciG+//RZXr15FQUEBMjMzcejQISxZsgRDhgxBbGysRV47zcjfyZMntUFvZGQkXF1dDd4vEomwdu1aODg44OzZs+jXrx+WLVuGCxcuID8/H/fv38exY8fw9ddfY8KECQgJ0c89K5PJ8MEHHwBQj35OnDgRBw4cQG5uLi5cuICXX34Z/+///T+9IKm+hg0bph3p/eKLLzB//nzI5XLk5eXhzz//RGRkJLZv3260jZCQEO10/wsvvIDly5fjxo0buHfvHtavX4/hw4cjOzvb5JKAgQMHAlCfmPHee+/h3r17qK6uRnV1tfYPFT8/P4waNQqA+vvoo48+wtWrV5GdnY1t27Zh1KhROHXqVKMCJl2GRvt0NXYdoMbly5e1r1ltY8aMwezZsxu9Jrk1rmlu6+hkEWJz6GSRlqWkpAQLFizATz/9ZPKXmaenJ/7zn/9oR40M0axBWrFiBX766SejU5iGTgDRnChhiqOjI9avX28wV15SUhLmzJkj2FVc24ABA3Dy5Em9a5qTRd555x28++67Jp/XUCqV6NixI+7fv6+9lpKSghkzZph87uDBg3jooYdw7949k/d5eHgYHDU0dNawxujRo/H4449rT41pyO8XQ2cN63r99ddx//59rF271uDJIrt370ZERITBndlSqRQ//PADli9fjv3792P27Nl66041hg8fjsOHDwuu695/9uxZjB492uD0K2MMS5YswZkzZ4z201ya0b7ar6Wjo6N2hFazTtSUGzduYOfOnQbLqqursWfPHm0eR11OTk6IjY3VrjFtDIlE0qqXs7RmdLIIIaTFcnZ2xrp163D8+HHtdLeHhwekUin8/Pwwfvx4fP7557h69arJIFCXvb090tLS8Mknn6Bv375wcnKCq6srxo0bh8TERIPHwC1cuBC//fYb5s2bh4EDB8LPzw92dnZwcXFB//79taOLxhImx8TEICMjAx988AFGjhwJLy8v2NnZwdnZGb1798bMmTOxbt06s/LOmUMsFusdzefh4aE3UmnM6NGjcfXqVSxduhQTJkxAu3btYGdnB0dHRwQGBiI2NhYrVqzQO55O1zfffIMNGzZg7NixcHV1hZOTE/r27YtPPvkEaWlpJhNim8PV1RUHDhzAp59+iuDgYDg4OMDT0xPjxo3Db7/9pj0ZxpiwsDD8+eefiI6OhqenJ6RSKTp37ozHH38chw8fNus4w23btuHll19Gr169DB7jBwB9+vTB8ePHMWvWLPj6+kIikcDPzw9RUVHYs2ePySns+jA22qdZA6j7h0BD5Ofn4/vvvzcYBD7wwANYsGCBRYLA1r6muS2jEUFic2hEkBDSWpgz2mcOQyOC58+fx8aNGwWjp4wxTJgwASNHjtSmSWoIOjO45TA1Ikh5BAkhhJAWSpNv05Kqqqqwc+dO7XpZXa6uroiLi4O/v3+D6w8JCcG0adMa00XShCgQJIQQQlooFxeXBqeBMSQnJwcbNmwwOKXco0cPREVFwdHRsUF1M8YwaNAgCgJbGQoECSGEkBbK0dERlZWVDUoFU9vp06exZcsWwZnlIpEIEydOxLBhw+qd9FkmkyE8PJymf1sxCgQJIYSQFkwqlUKhUDQ4RQznHAqFAsnJyYIyDw8PxMXFoWPHjvWuVyaT1ZmXkbR8FAgSQgghLVRDE0NrVFVVIT8/32AC8AcffBCRkZFGd0abIpFIzNqlTlo+CgQJIYSQZlRWVobi4mIolUqIxWK4uLjA0dEROTk5Jk+7qUtpaSmKiooEI4l2dnaYMmUKBg0a1ODzfykVTNtBgSAhhBDSTGonjFYqlSgsLERZWVmDg0CVSoXCwkKDxx16e3sjPj4e7du3b1S/KQhsOygQJIQQQpqQ7gigIZzzBgeBlZWVyM/PN1h3//79MXXq1EYn/aZj4toWCgQJIYSQJmLsyDhLKCkpQVFRkeA6YwyOjo6IiopqdBsSiQShoaGNroe0HBQIEmImznmD19MQQghg/Mi4xlCpVCgoKEB5ebmgTCKRwN3dvVFrDTUYY7Q2sA2iQJAQM2iy+9vZ0X8ZQkjDWfqUEFNTwZoztlUqFaqrqxvVjkQioSCwjaLfaoSYwdHRESUlJXB3d2/urhBCWjFLHRnHOUdJSQmKi4sFZYwxuLu7QyaTAQDKy8uRmZnZ4LbozOC2jQJBQszg6uqKnJwcuLi4QCwWN3d3CCGtQF2bQhpKqVSioKAAFRUVgjKJRAIPDw/t7EV1dTVyc3Nx69aterURExNDgZ+NoECQEDO4uLhAoVDgxo0b8PT0hLOzM8RiMa0ZJIQY1NhE0MZUVFQgPz8fKpVKUObs7AxnZ2cA6inj8vJy5OXl4dSpUygpKTG7DZlMRkGgDaFAkBAzMMbQrl07FBcXo6ioCFlZWRb/K58Q0rJpgiuVSgWRSAQHBweDqVgsdTZwbeXl5QZHATW7gjWniKhUKpSUlCAzMxNZWVkG8wkaIxKJ6MQQG0OBICFmYozB1dUVrq6uzd0VQkgTk8vl2Lp1K6qqqrTXdDdQyOVypKWlobCwEIwxi+4MLiwsRGJiIm7evCko69KlC2JiYiyS208mkyE8PJxGA20MBYKEEEJIHdLS0vSCQEB9jm9qaioAYPPmzdpySwaBly5dQkpKisFRvbFjx2LMmDGNWrcsEokQFRVFwZ8No0CQEEIIAfRG9WrvlC0sLDT4jEKhQGpqqiBIbKzq6mqkpaXhzz//FJQ5OzsjJiYGDzzwQKPboSCQUCBICCHE5snlcr1RvcLCQmzevBmA+lxdNzc3k8GgJeXl5SEhIQF3794VlHXr1g3R0dHaTSGNlZaWBoDODrZloubuACGEENLcjE39agKlwMDAJunHuXPnsHLlSkEQyBhDaGgoZs6cabEgEPg74JXL5Rark7QuNCJICCHE5hkb7SssLIRcLsfp06et2n5VVRV27NiB48ePC8pcXV0RFxcHf3//etfLGIODg4PJUUtNwEujgraJAkFCCCE2z9jUr5ubm8HRQkvKzs5GQkIC7t+/Lyjr2bMnZsyYAUdHxwbVPWjQIEybNk1v/aMhxq6Tto+mhgkhhNi80NBQSCQSwfXKykqrBknp6elYtWqVIAgUi8WYMmUKHnnkkQYHgRKJBNOmTQOgXgO4cOFCo2lmLJF+hrRONCJICCHEpmlGywyN+ll6I4hGRUUFtm3bZnDK2cPDA/Hx8ejQoUOD6xeLxYiMjBRcDw0N1dsUA6gDxtDQ0Aa3RVo3CgRtGGPMHsBkACEAfABIOef/0CmXAHABoOSc07wBIaTNqb1buCncu3cPCQkJyMnJEZT16dMHERERcHBwaHD9phJDa64ZS5NDbA8FgjaKMTYXwAcAvDWXAHAA/9C5rQOAqwBUjLEAznlm0/aSEEKsy9rr/3RxznH8+HFs375dcESlnZ0dwsPDMXDgwAafYe7m5oaFCxfWeV9wcDAFfkSLAkEbxBj7EMAiqIO/UgAXAQyqfR/n/AZjLBXANADxAJY1ZT8JIcQSGpIo2tLKy8uxadMmnD9/XlDm7e2N+Ph4tG/fvsH10/QuaSgKBG0MY2wUgDegHv37AMBHUH8fFBl5JAVABIBQUCBICGllGpMo2lLu3LmDhIQE5OfnC8oGDBiA8PBwSKXSBtdPZwSTxqBA0Pb8s+b9t5zzfwEAY8zJxP2apFZ9rNorQgixAlOJooODgw1unrAUzjkOHz6MXbt2QaVS6ZVJpVJERESgb9++9aozJCQEV65cofV9xGIoELQ9I6EeDfzGzPs16e19rdMdQgixnrry5t28eRPV1dUWb7esrAwpKSm4fPmyoMzX1xdxcXHw9vY28KRxUqlUmw6GEEuhQND2aH7yZJh5v2ZFM+WcJIS0OsamfhljWLx4sVXavHHjBhITE1FUJFxxM2TIEEycONFgzsK6REREWKJ7hOihQND2FAPwBOAKwJyFMZ1r3udarUeEEGIlnp6eBgNBzrnF21KpVPj999+xd+9eQf329vaYMWMGgoKCGlw/TQETa6BA0PZcgHp6eBSAX8y4X/Mn6Emr9YgQQqxALpcjI8PcyY/GKS4uRnJyMv766y9BWceOHREXFwcPD48G1x8SEtKY7hFiFAWCtmcj1EHgm4yxZM55ubEbGWNdALwM9ZrCpCbqHyGEWERaWlqTtHPt2jUkJSWhtLRUUDZixAiEhoZCLBY3uP6QkBBaG0ishgJB27McwEIAQQD2MMaeB3BJ94aaE0eiAHwO9TTyJQA/NmkvCSGkkaydFkapVGLfvn04ePCgoMzR0RHR0dEIDAxsVBsymYyCQGJVFAjaGM55GWMsEsAeAMMAHAOQpylnjGUA8AMggTrhdA6AaM655bfVEUKIlcjlcqvWX1hYiMTERNy8eVNQ1qVLF8TGxsLV1dWqfSDEEigQtEGc83TGWH+oU8iEA/CqKWIAuujcmgpgAedc+JOOEEJaKLlcjpSUFKvVf+nSJaSkpEChUAjKxo4di7Fjx0IkskyiBUNtEGJJFAjaqJrgLoIx9gCACQB6AnADUALgLwC7OecXm7GLhBBilKFj4wBor1lDdXU1du/ejcOHDwvKnJ2dERsbi65du1q0TTc3N4vWR0htFAjaOM75X1AHfoQQ0uIYC/hqHxu3ceNGKJVKU1U1Sl5eHhISEnD37l1BWbdu3RAdHQ1nZ+cG1W0s1yGdH0yaAgWCpMkxxroDkANwqLn0JOd8jYn7XQC8BCAOQFeok1xfBvAzgK8555Y/G4oQ0uyMnRNsZ2cnOBLOmkHg2bNnsXnzZlRUVOhdZ4whNDQUI0aMaPBUsJubGxYuXAjAcNBLuQOJtVEgSJrDcvwdBJrEGOsKYDeAB2oVDa55e5wxNpFzLjzNnRDSqhk7J9ga5wIbUlVVhe3bt+PEiROCMjc3N8TGxsLf379RbeiO+AUHB1PgR5ocBYI2ijHmCfUI2xCozxGWQb1ZxBjOOW/0HAVj7HEAYVAfcWdyMQ1jTApgM9RBoALA6wCSod7RPBvAvwAMArAewKTG9o0Q0rJYO/2LKdnZ2diwYQOysrIEZb169cL06dPh6OjY6HYo8CPNjQJBG8QYexrAfwA4aS6Z8Vijz2NijHlAnZtQCfVUb0odj8wD8GDNx7M55xt0yhYzxhQAPgUwkTE2jXO+tbF9JIS0HMbWzllbeno6tm7dKhh5FIvFmDRpEoYMGQLGzPmxaRptBCEtAQWCNoYxFg1gVc2nHOqj465BPeJmbZ8BaAdgKYDTZty/oOb9qVpBoMYXAF4F4FNzLwWChLQBumvlmlJFRQW2bt2KM2fOCMo8PT0RFxeHDh06WKQt2ghCWgoKBG3PazXvzwGI4ZxfaYpGGWOjADwFIBPAO1CfWGLq/gegPv0EABIM3cM5r2KMbQTwNIBQxpgj57zMcr0mhDQluVyO1NTUZsmdd+/ePWzYsAG5ubmCsj59+iAiIgIODmYtbTYoJCQEV65coY0gpMWhQND29IV6JHBBEwaBEgAroZ6CfplzXlSzRtGUgTofHzFx3xGoA0EHqAPH443pKyGkedTeIdxUOOc4fvw4tm/fLth5bGdnh6lTp2LAgAGNmgpmjNExcaTFokDQ9iig3hhytgnbfB3qIG035/xXM5/pqfOxqTyHGbWeoUCQkFbI0A5ha1MoFNi8eTPOnz8vKPPx8UF8fDzatWvX6HYGDRrU6DoIsRYKBG3POQCjod4pXGDtxhhj3QD8PwCVAJ6vx6PeOh9nm7hPd0ufl9G7CCEtllwub/L1gLdv30ZCQgIKCgoEZQMGDEB4eDikUmmj2mCMYdCgQTQaSFo0CgRtzzcAxkCdfuWNJmhPkzPwI875pXo856TzcbmJ+3QXEzUsrT8hpNlopoSbikqlwuHDh7F7926oVCq9MqlUioiICPTt27dRbUgkEkRGRtIaQNIqUCBoYzjnvzHGJgJ4jTF2i3P+jbXaYow9BmAigOsAPrBWO2b2ZR7U6WganQCWEGK+uk7LaMop4dLSUqSkpODKFeHyaF9fX8THx8PLq/ETCxQEktaEAkEbxDmfyxiTA/iCMfYygDQA96DO72fquffMbaMmZ+B/aj59gXNe322ApTofOwAoMXKfTOdjY/eAc74KNWlzQkJCGp0TkRBSN2NHxAF/J1JuqinhGzduICEhAcXFxYKyIUOGYNKkSbCza/yvRMYYBYGkVaFA0AYxxvoBeBjqEzq6Qr3r1hxmB4JQn/rRHsAmznlD5n1ydD72gfEgz0fnY2HeB0JIszF2RFxaWpo2WLJ20miVSoWDBw9i37594Fz/b0AHBwfMmDEDvXv3tlh7tdsgpKWjQNDGMMZ6A9gHwFXncjYsn1Bac3zcdMZYXT8Zv2eMfa95jnN+HYDuesKu0N8dbKgd1HqGENLMjAV4hYWFWLx4MRhjVg2ciouLkZSUhIwM4Y+PTp06IS4uDu7u7hZtk04LIa0NBYK25z0AbgAKAbwMYAPn3OiUajM6qfPxUAB7jNw3tOZ9OQBhDghCSLOpa7TPmkHgtWvXkJSUhNLSUkHZyJEjMWHCBIjF4gbXL5FIAEBvxJNOCyGtEQWCtmcU/k4obW5Ov4Z4CcC7Jso74O8j4d4BsKnm47sAwDn/izF2Hur8g3EAPq5dAWPMDsCMmk/T6FQRQlqWwMBAHD/etKk9lUol9u7di99//11Q5ujoiOjoaAQGBja6naqqKsTExJjcCENIa0CBoO1xqXm/y5qNcM6NTeUCABhjBTqf3uScpxu4bTmALwEMZIzFcs4Ta5UvhPrsYs29hJAWQi6X4/Rpc44Ut5yCggIkJibi1q1bgrKAgADExMTA1dXVwJP15+bmhuDgYAr8SKtHgaDt+QvAg9DfbdtSrQLwDNT9XccYaw8gBepNLrOh3pACALs451sN1kAIaXJyuRzJyclNunHi4sWLSElJQXm5ftpRxhjGjh2LMWPGQCQSWaQtmgImbQkFgrZnPYD3AUQAWNHMfTGJc17JGIsEsBvAAwC+rnnTdQLqHdCEkBZAkzKmqYLA6upq7Nq1C0eOCI8kd3Z2RmxsLLp27WrgyYahKWDS1lAgaHuWAIgF8CFj7Cjn/GRdDzQnznkGY6w/1GsO46AOCJUALgP4GcBXnPOmPaCUEGJUUyaIzsvLw4YNG5CZmSko6969O6Kjo+Hk5GTgSeNkMhkefPBBwdpGOi2EtFUUCNqeoVBv4lgK4BBj7Geo1wuak1D6gKU6UZMihpl5bzHUu53rk8eQENLEmvLM4LNnz2LTpk2orKzUuy4SiRAaGorhw4c3aCpYoVDA398f/v7+tBGE2ARGyS9tC2NMBfWuYUAdiJn7DcA5523iD4eQkBDe1DsZCWlrah8dFxgYiNOnT1t9NLCqqgrbt2/HiRMnBGVubm6Ii4tD586dG9WGSCRCVFQUBX6kzWCMneCchxgqaxO/2Em9MSMfE0JInQwdHdcUf1xlZWUhISEBWVlZgrJevXphxowZkMkavw9OpVLpnX5CSFtGgaDtsdyqaUKITZHL5UhNTYVCYemDiEzjnCM9PR3btm0TjDiKxWJMnjwZgwcPBmOW+7u2qaa4CWluFAjaGM75jebuAyGk9ZHL5UhJSYFKpWrSdisqKrB161acOXNGUObp6Yn4+Hj4+flZvF06Ko7YCgoECSGE1CktLa3Jg8DMzEwkJCQgNzdXUBYcHIyIiAjY29s3uH6pVIqqqipBqhvNhhNCbAEFggSA9rg2FwDFnPPq5u4PIaRlacqpUs45jh07hh07dkCp1E9mYGdnh2nTpqF///6NngqurKyEWCyGSCTSTjnLZDKEh4fT+kBiMygQtGGMsUkA/gFgNID2OtfvAfgdwGrOuVWPoiOEtA4ymaxJ1gYqFAps2rQJFy5cEJT5+PggPj4e7dq1M/Bk3RhjgtE/pVIJZ2dnLFy4sEF1EtLaUSBogxhjTgDWAZihuVTrFj+okzfHMcY2ApjFOS9twi4SQloQuVwuyNdnDbdv30ZCQgIKCgoEZQMHDsSUKVMglUrrXW9ISAimTZuGxYsXGyynjSHEllEgaJtSAEyAOgDMArAZwDkAxQCcoT7bNxLqUcIZAJIATG6OjhJCml9aWppgitaSVCoV/vzzT4PrEKVSaaNO9NAEgYB6A4ihoI82hhBbRoGgjWGMzQQQCnUi6U8BvMM5F/ypzxiTAlgM4P8AhDHGHuWc/9KknSWENClDSaKvXLli1RGz0tJSpKSk4MqVK4IyPz8/xMXFwcvLq151xsTEGAwcQ0ND9fIfAuqj42hjCLFlFAjanllQB4GrOedvGLupJjh8gzHmCWAugNkAKBAkpI1qjiTR169fR2JiIoqLiwVlQ4cOxcSJE2FnV79fUyEhIUZHDzXX6eg4Qv5GgaDtGVDz/isz7/8K6kCwv1V6QwhpEVJTU61+PJyGSqXCgQMHsH//fsHmDQcHB8yYMQO9e/eud71du3bVTgMbExwcTIEfITooELQ9msUwt828X3Ofu+W7QghpCeRyeZOdFlJcXIzExERcv35dUNapUyfExcXB3d293vWKRCIMGDCg7hsJIXooELQ9+QDaQX3UXL4Z92uOpMuzWo8IIc0qNTW1Sdq5evUqkpKSUFZWJigbNWoUxo8fD7FY3KC66XxgQhqGAkHbcwzANAAvQb1esC4v1by3/onyhJAm1xSjgUqlEnv37sXvv/8uKHN0dERMTAy6d+/e6HYoDQwh9UeBoO1ZCyACwGOMsRIAr3POBSu1GWPOAD4BMBPqzSXfN2kvCSFNIi0tzar1FxQUICEhAbdvC1ejBAQEIDY2Fi4uLhZpi9LAEFJ/FAjaGM55Yk2S6BkA5gF4nDG2G8B5ACVQ5xHsDSAMgFPNYymc8+Tm6C8hxDrkcjlSU1OtOhp48eJFpKSkoLy8XO86Ywxjx47FmDFjIBKJ6l2vTCZDdXU1pYEhxAIoELRNjwD4BsCTUAd702vedGlOG/kOwHNN1zVCiLXJ5XKkpKQIkjdbSnV1NXbt2oUjR44IylxcXBAbG4uAgIAG1S2RSBAeHg6A0sAQYgkUCNogznkFgH8wxr4G8BSAUQC6QD0aWALgOtRnDX/HOT/VXP0khFhG7UTRCoXCakFgbm4uEhISkJmZKSjr3r07oqOj4eTkZODJutUO+CjwI6TxKBC0YZzzkwBONnc/CCHWI5fLsXHjRu0RcdbcUKFJSl37XGKRSITQ0FAMHz68QVPBxk4KIYQ0HgWChBDSRsnlciQnJwuSNltaZWUltm/fjpMnhX9Xuru7Iy4uDp06dWpQ3aZOCiGENB4FgoQQ0gZpRuesHQRmZWVhw4YNyM7OFpT17t0b06dPh0wmM6sukUgEe3t7KBQKWvdHSBOhQNDGMMYmA9gGIANAD8650YVCjDExgMtQrx8cwzk/1DS9JIQ0lrWPjOOc49SpU9i2bRuqq6v1ysRiMSZPnozBgweDMWakBn2MMURFRVHgR0gTo0DQ9jwC9Y7g700FgQDAOVcyxlYD+LDmOQoECWkFrJ0kuqKiAlu2bIFcLheUeXl5IS4uDn5+fvWqk3NOQSAhzaD+q3ZJazcQ6gTRu8y8X3PfaOt0hxBiadY8Mi4zMxMrV640GAT27dsX8+bNq3cQqLF06VKD9RJCrIdGBG2PZsX2NTPvz6h539EKfSGEWIBmPaC1p4KPHTuGHTt2aHcga0gkEkydOhX9+/c3eyrYkMLCQmzevBkApYYhpKlQIGh7NKu2zR0N1txnmTOgCCEWJZfLkZSUZNU2FAoFNm7ciIsXLwrK2rVrh7i4OLRr184ibVVVVSEtLY0CQUKaCAWCtuce1Js/HgSwz4z7H6x5L9wSSAhpdtY+K/jWrVtISEgwmH9w0KBBmDJlCiQSiUXbtGauQ0KIPgoEbc8hAAEAnoV5geACqNcUHrZelwghDWWtoEmlUuHPP/9EWlqa4BQSqVSK6dOno0+fPlZp283NzSr1EkKEKBC0PWsAPAYgljH2L875e8ZuZIy9BSAe6kBwTZP0jhBitq1bt1ql3tLSUiQnJ+Pq1auCMj8/P8TFxcHLy8sqbUskEoSGhlqlbkKIEAWCNoZzvpsxlgwgGsA7jLFYAD8BOA2gGOq1gH0BzASgWaSziXO+rTn6Swj5m1wuR2pqqlVTw2RkZCApKQnFxcWCsmHDhiEsLAx2do371SGRSAxubKEk0oQ0PQoEbdMsABsAhEMd7H1s4B7N1r9tUI8gEkKaQVMEf4B6KvjAgQPYv3+/4DQSBwcHREVFoVevXo1uRxPspaWlobCwkII/QpoZBYI2iHNeBmAaY+xxAM8DCIH+LmIlgGMAvuKc/9wMXSSEQB0Ebty4UZCuxdKKioqQlJSE69evC8o6d+6M2NhYuLu716tOqVSKyspKwfXAwEAEBwdT4EdIC0GBoA3jnP8I4EfGmAvUG0hcoJ4evs45F84LEUKaVFpamtWDwCtXriA5ORllZWWCslGjRmH8+PEQi8X1qlMmkxkNBK9cudLgvhJCLI8CQYKaoI/S+RPSwlgzjYpSqcSePXvwxx9/CMqcnJwQHR2N7t27N6ju8PBwo7kNKTUMIS0LBYI2hjG2B+pdwFM55xVm3C8CsBsA55zTVj5CmoBcLrdqfsCCggIkJCTg9u3bgrKuXbsiJiYGLi4NzyEfHBysXQNYG6WGIaRloUDQ9oyDOhA0d66H6TxDCLEwTdCn2TgRGBiI06dPW+24uAsXLmDjxo0oLy/Xu84Yw7hx4zB69GiIRA0/hl4kEkEulyM0NFRw7B2lhiGk5aFAkNSl4QeHEkJMqn1GcGFhIY4fP26VtqqqqrBr1y4cPXpUUObi4oLY2FgEBAQ0uh2VSoW0tDQsXLgQAGh3MCEtHAWCpC6+Ne+FK8kJIQ0ml8uRnJwsSNViDbm5udiwYQPu3bsnKAsMDERUVBScnJws1p5mSph2BxPS8lEgaLtM/vZhjDEAfgDeqbkkPGKAENIgmpHApggCz5w5gy1btgh28IpEIoSFhWHYsGGNmgo2hNYBEtJ6UCDYxjHGDOWeYABK1LGeWTgAw1sACSEm1V4DqEmmbK01gBqVlZVITU3FqVOnBGXu7u6Ii4tDp06dLN4urQMkpHWhQLDtMxbtmRsFqqA+heRTy3SHENtROyF0YWFhkySIzsrKwoYNG5CdnS0oCwoKQmRkJGQyWYPrl8lkUCgU2s0tV65coXWAhLRSFAi2fU/W+vx7qEf4FgAwlT6mCkAugHTO+X0r9Y2QNi01NVUQ9CmVSjDGrDItzDnHqVOnsG3bNlRXV+uVicViTJkyBSEhIajHbIBBr7/+eqOeJ4S0HBQItnGc87W6nzPGvq/58Meao+YIIVZi7HxgzjlEIhFUKpXF2iovL8eWLVtw9uxZQZmXlxfi4+Ph6+tr4Mn6ofV/hLQtFAjanq6A9rxhQkgzsbe3Nxoo1tfdu3eRkJCAvLw8QVm/fv0wdepU2Nvbm11fSEgI/P39KQ8gITaAAkEbwzm/0dx9IKQtMbQZBIDJk0EkEolFgkDOOY4ePYqdO3cKpqAlEgmmTZuG/v37m12fSCRCVFSU3ho/ygNISNtGgSAhhDSQoYTQGzduBOfc5LSvJXYMKxQKbNy4ERcvXhSUtWvXDvHx8fDx8alXnfb29nqBHuUBJKTto0DQxtScNdwQdNYwIbUYSgNj7R3BAHDr1i0kJCQYPMs3JCQEkydPhkQiqXe9lpqqJoS0HhQI2p5xUO8armvboGZLI6v1OSGkhqFAzJpUKhUOHTqEtLQ0wa5je3t7REZGok+fPg2unzaCEGJ7KBC0PT/AdFBnB/WJIkMBOAH4C8DBJugXIa2Om5tbkwWDJSUlSE5OxrVr1wRlHTp0QFxcHDw9PRtcv1gspo0ghNggCgRtDOd8jjn3McZcAHwA4DkAyznnn1uzX4S0RqGhoYKdtdaQkZGBxMRElJSUCMqGDRuGsLAw2Nk1/Me5VCpFREQErQckxAZRIEgM4pwXA3iRMeYJ4FPG2AnO+b5m7hYhLUpwcDBOnTqFjIwMq9SvUqmwf/9+7N+/X1Amk8kQFRWFnj17Nrh+TWLrxpwyQghp3SgQJHVZCmAmgFcB7GvWnhDSAl2/ft0q9RYVFSExMRE3bggzPnXu3BlxcXGNXtOnWWdYWFiIzZs3AwCNChJiYygQJHXRLEga0qy9IKQF0M0ZaK1j4gDgypUrSE5ORlmZMO/76NGjMW7cOIjFYou2WVVVhbS0NAoECbExFAiSumjOpHJu1l4Q0sxq5wy0RhCoVCqRlpaGQ4cOCcqcnJwQExODbt26WbxdjabeBU0IaX4UCJK6vFnz/q9m7QUhzcxQzkBLys/PR2JiIm7fvi0o69q1K2JiYuDi4mK19gFKH0OILaJA0MYwxsaYcZsMQCCAx6BOI8MB/GLNfhHSUhg6Mi44ONiqo2Xnz5/Hpk2bUF5ernedMYbx48dj1KhREIlEFmuva9euuH37Np0jTAgBs9YaF9IyMcZUMD85tCaZ9AEAkznnFdbpVdMKCQnhx48fb+5ukBao9vSvtVVVVWHnzp04duyYoMzV1RWxsbHo0qWLRdsMCQnBtGnTjAa8hJC2pybzR4ihMhoRtE11nSoCAFUA0gGsBbCSc279c7MIaWbWnv7VlZOTg4SEBNy7d09Q1qNHD0RFRcHR0dFi7THGEB0drQ326BxhQghAgaAt6lpHOQdQDiCPc17dBP0hpMVoqs0SZ86cwZYtW1BZWal3XSQSYeLEiRg2bBgYM+fvNfNIJBJERkZS4EcIEaBA0MZwzoVJyQghAGDVlDAAUFlZiW3btiE9PV1Q5u7ujvj4eHTs2NEibWm+Fpr2JYSYQoGgDWOMiQB4Qj1VnEfTv8TWWTMIvH//PjZs2ICcnBxBWVBQEKZPnw4HBweLtEUjgIQQc1EgaENqAr84ANEARgLoWKv8DoA/ACQDSOCcq5q8k4Q0Id0NEzKZzCojgpxznDx5Eqmpqaiu1l9tIRaLER4ejkGDBllsKphGAAkh9UGBoI1gjE0EsAqAv+aSgds6AXio5u0Txtg8zvnuJuoiIU2q9g5hhUJh8TbKy8uxZcsWnD17VlDm7e2NuLg4+Pr6Gniy4RYuXGjR+gghbRsFgjaAMfYagI+hDv40AeBNAOcBFNR87g6gNwBNrooAANsZY4s450uaqq+ENJXU1FSr7hC+e/cuNmzYgPz8fEFZv379MHXqVNjb21u0TUoITQipLwoE2zjG2CwAn9Z8WgLgcwDfcc5vGbm/E4AnAbwKwAXAp4yxe5zzH5uiv4Q0BblcbpURQEA9FXzkyBHs3LkTKpX+6gqJRIJp06ahf//+Fm+XEkITQhqCAsE2jDHmDuA/NZ+eBTCVcy48v0pHTfn7jLHvAGwDEAzgC8bYZs45HURKWj25XI6kpCSr1F1WVoaNGzfi0qVLgrL27dsjLi4OPj4+FmuPdgYTQhqLAsG2bQ4ALwC5AMI451nmPsg5v1OzrvAc1DuL5wD4rxX6SEiTkMvlSE1NtdpI4M2bN5GQkICioiJBWUhICCZPngyJRGLRNjnneOeddyxaJyHEtlAg2LZFQJ0gekl9gkANznkWY+xzAB8BiAQFgqSVsubRcSqVCn/88Qf27Nkj2HFsb2+P6dOn48EHH7R4uwCtCSSENB4Fgm1bz5r3WxtRxxaoA8Feje8OIdZn6Axdax0dV1JSguTkZFy7dk1Q1qFDB8TFxcHT09Pi7QK0JpAQYhkUCLZtmsVIdxpRx92a916N7AshVld75K+wsNBqI4F//fUXkpKSUFJSIigbPnw4QkNDYWdnuR+xYrEYUqkUCoWC1gQSQiyGAsG2rRjq9X1uAIQ5LMzjWvO+1CI9IsSKDI38WToIVCqV2L9/Pw4cOCAok8lkiIqKQs+ePQ082fh2pVIpXn/9dYvXTQixXRQItm13oA4EhwK43sA6htW8N7nbmJDmojsVbG1FRUVITEzEjRvCI7v9/f0RGxvb4HV7MplMO9pn7Gtpiq+REGJbKBBs23YD6AvgBQDr6/swU5959QLUG07SLNs1Qhqm9rFw5eXlVj0jWOPy5ctITk42uOt4zJgxGDt2LMRicYPqlslkeiN9S5cuNRj00eYQQoiliZq7A8SqfoQ6iBvGGPusAc9/AvWIIAewzpIdI6QhNGsANUGSQqGwehBYXV2NnTt34ueffxYEgU5OTnjiiScwYcKEBgeBgPB4u9DQUEGqGdocQgixBhoRbMM45+mMsXUAngDwMmMsEMBCzvl1U88xxroA+ALADKiDwJ845+lW7i4hdbLW7l9j8vPzkZCQgDt3hPutHnjgAcTExMDZ2bnR7chkMr3PNZtAau9+ps0hhBBLo0Cw7VsAdRqZoVDnAoxgjO0HsB/ABeifNdwLwNiaNxHU5xIframjQRhjMgBTAEwGMBhANwBONe2eAZAI4HvOucksv4yx9gBeATAdgD+AMqiTXa8BsIY3xdwgaVZyubxJ18idP38eGzduREVFhd51xhgmTJiAkSNHQiSyzKRKdXW14FpwcDAFfoQQq2P0+7PtY4y5AlgJ4OGaS3X9o7Oa9xsAzGvM0XKMsSKozyw25SKA6ZzzK0bqCIE6F2I7I89vBxDFOa8wUq4nJCSEHz9+3JxbSQthzYTQtVVVVWHnzp04duyYoMzV1RVxcXHw9/e3eLsxMTEU+BFCrIIxdoJzHmKojNYI2gDOeRHn/FEA4QB2AFBBHewZelMB2AX1ucQPW+B8YRcAFQB+AfAI1COCngD6A1gOdVDaC8BOxphgjo0x5g1gE9RBYC6AxwH41Tzzdc1tUwB82ch+khZq69atSEpKapIgMCcnB6tXrzYYBPbo0QPPPPNMg4NANzc3k5s90tJoPxYhpOnR1LAN4ZzvALCjJuAaBqArAI+a4nyoU8wc5pwXW7DZrwG8zzm/X+t6PoBnGWM3oN6UEgDgWQD/rnXfIqgDPxWACM754Zrr9wA8zxgTQT11/Q/G2Jecc7kF+06a2Q8//ICMjIwmaev06dPYsmWLIOAUiUSYOHEihg0bBvVG+obRbPRISkoyWE6pYQghzYECQRvEOS+BOrVMU7T1fB23fA7gNahPLgmHTiDIGJMAeLrm0406QaCud2rukQCYD6Cu9kgrIZfLmyQIrKysxLZt25Ceni4o8/DwQFxcHDp27NioNiQSiXbaNzU11WAKGkoNQwhpDjQ1TJoV57wagGZtYIdaxaOhPhUFABKMPJ8N9cYXQL2RhLQRTTFVev/+faxatcpgEPjggw9i/vz5jQ4CAf3TTcLDwyk1DCGkxaARQdIStK95X1Tr+kCdj4+YeP4IgDAAnRlj3pzzHEt2jjSdpjolhHOOEydOYPv27YIdu3Z2dpgyZQoGDRrUqKlgXbqjfZQahhDSklAgSJoVY2wA1GsVAaD21K/mwFYVAOGZXn/TnT/sCYACwVaoqXYGl5eXY/PmzTh37pygzNvbG/Hx8Wjfvr2BJ+umGenT/RoMjfZRahhCSEtBgSBpbpoTTziAVbXKvGve59dMIRuTpfOxl6U6RppWUySLvnPnDhISEpCfny8o69+/P6ZOnQqpVNrg+iMjIwHQaB8hpPWgQJA0G8bYawA0QyXLDez4dap5X15HVbor7w0e88AYmwdgHgCr5IAjjWfN6WDOOQ4fPoxdu3ZBpVLplUkkEkRERKBfv36NaiMkJEQb8FHgRwhpLSgQJM2CMTYZwMc1n8oBvGrN9jjnq1Az4hgSEkJZ1FuIplgTWFZWho0bN+LSpUuCsvbt2yM+Ph7e3t4GnjSfTCbDtGnTGlUHIYQ0BwoESZNjjA2C+tQSMYBbAKYZOWKutOa9Qx1V6h7UWtL4HhJL0w34NNOlN2/ehLVPeLl58yYSEhJQVFR7HxIwePBgTJo0SbCDtyEMpYMhhJDWgAJB0qQYYz0ApEJ94kg2gEmc81tGbtds+nBnjNmZWCfoo/NxrmV6Siyl9iaQwsJCpKSkCKZoLUmlUuGPP/7Anj17UPsYTXt7e8yYMQNBQUEWa49yABJCWisKBEmTYYx1hvr4Oh+oU8VM4ZxfNPGIZi5PDMAfwF9G7uuq87Fw/o80K0ObQKwZBJaUlCApKQl//SX8dunYsSPi4uLg4eFh4EnzSCSSOncFE0JIa0GBIGkSjDEfqINAf6g3d0Ryzk/W8Zhu+VAYDwSH1ry/RTkEW56mPDrtr7/+QmJiIkpLSwVlI0aMwIQJE2Bn1/Afe1KpFBEREbQrmBDSZlAgSKyOMeYKYAfUOf6qAMRxzg+Y8ehBAIVQny4SB+AXA3V7Axhb8+kmi3SYNErt9YC1R9CsQalUYv/+/ThwQPhtJZPJEB0djR49ejSqDZFIhIiICMoBSAhpUygQJFbFGHMAsBnAAKgTQ8/inG8z51nOeRVjbDWAVwDMYIwN4ZwfrXXbuwCkUOchXGmxjpMGMbQe0NoKCwuRmJiImzdvCsq6dOmCmJiYBq/hY4yBc04jf4SQNosCQWI1jDExgPUAxtRcehXAVsaYwVx/AFSc87Ja1z4B8BgAv5pnXwSQBsAVwD8BPFdz32oDeQhJE0tNTbX66J+uy5cvIzk52eCu3bFjx2LMmDEQi8UNrp9zjnfeeacxXSSEkBaNAkFiTZ0BTNf5/D81b8bcABCge4FznsMYmw5gK4B2AH4y8Nx2qINC0ozkcnmTpVGprq5GWloa/vzzT0GZs7MzYmJi8MADDzS6HdoNTAhp6ygQJC0e5/w4Y6wv1COK06EOMBUAzgH4HsAaXjtHCGlyqampTdJOXl4eEhIScPfuXUFZt27dEB0dDWdnY4PO5qPdwIQQW0CBILEazvl1AMxCdd0H8FrNG2kh5HI5UlNTm2wk8Ny5c9i0aRMqKir0rjPGMGHCBIwcORIikajR7dCaQEKIraBAkBDSIHK5HBs3boRSqbR6W1VVVdixY4fBk0hcXV0RFxdn0TOkFy5caLG6CCGkJaNAkBDSIGlpaU0SBGZnZyMhIQH3798XlPXs2RMzZsyAo6OjxdqjdYGEEFtCgSAhpE6GzgpuitQw6enp2Lp1q2AnslgsxsSJEzF06FAwZpHVBwBoXSAhxPZQIEgIMcnYWcHWVFFRgW3btuH06dOCMg8PD8THx6NDhw4WaYtyBRJCbBkFgoQQk5r6rOB79+4hISEBOTnC0wL79OmDiIgIODg4WKQtiUSCyMhICv4IITaLAkFCiElNdVYw5xzHjx/H9u3bBWsP7ezsEB4ejoEDBzZ4Klgz4kfnBBNCyN8oECSEmOTm5mb1YLC8vBybNm3C+fPnBWXe3t6Ij49H+/btG9VGYWEhnRNMCCG1UCBICDEpMDDQYNoWS7lz5w4SEhKQn58vKBswYADCw8MhlUob3Q7tBiaEECEKBAkhBlk7WTTnHIcPH8auXbsEaw6lUikiIiLQt29fi7RFu4EJIcQwCgQJIYL0MJ6ensjIyLBae2VlZUhJScHly5cFZb6+voiLi4O3t7dF2qK1gIQQYhwFgoTYMEOjfoWFhVZdE3jjxg0kJiaiqKhIUDZkyBBMnDgREonEYu3RKSGEEGIcBYKE2Kja+QGtTaVS4ffff8fevXvBOdcrs7e3x4wZMxAUFGTRNmUymUXrI4SQtoYCQUJslKH8gNZSXFyM5ORk/PXXX4Kyjh07Ii4uDh4eHhZtUywWIzw83KJ1EkJIW0OBICE2qqnyA167dg1JSUkoLS0VlI0YMQKhoaEQi8UWbVMmkyE8PJzWBRJCSB0oECTERuhuCGmKKVOlUol9+/bh4MGDgjJHR0dER0cjMDDQom3SxhBCCKkfCgQJsQG11wNaKyWMRmFhIRITE3Hz5k1BWZcuXRAbGwtXV9cG1S0Wi8E510s5Q0fFEUJIw1AgSIgNaMr1gJcuXUJKSorBYHPs2LEYO3YsRCJRg+tXKpUQi8WQyWRQKBQ0CkgIIY1AgSAhbZxcLm+S9YDV1dXYvXs3Dh8+LChzdnZGbGwsunbtapG2lEolpFIpXn/9dYvURwghtooCQULaiNpJoTUnaWzcuNHqbefl5SEhIQF3794VlHXr1g3R0dFwdna2aJtNtdmFEELaMgoECWkDaq8BLCwsxObNm8EYg1KptGrbZ8+exebNm1FRUaF3nTGG0NBQjBgxolFTwcbQ2cGEENJ4FAgS0gYYWgNo7TWBVVVV2L59O06cOCEoc3NzQ2xsLPz9/RvdjkwmQ3V1td7XQ2cHE0KIZVAgSEgb0NTTpNnZ2diwYQOysrIEZb169cL06dPh6Ohodn1ubm4IDAzE6dOnBQGfJil07Wlv2hxCCCGNR4EgIW2ASCTSS6diTenp6di6datgxFEsFmPSpEkYMmQIGGNm1RUTE6MX0Pn7+xsN+CjwI4QQy6NAkJBWSndzSFOoqKjA1q1bcebMGUGZp6cn4uLi0KFDh3rVWTu4Cw4OpoCPEEKaEAWChLQCtXcEG5pGtaZ79+5hw4YNyM3NFZT16dMHERERcHBwaJK+EEIIsRwKBAlp4QztCD5+/HiTtM05x/Hjx7F9+3bB7mM7OztMnToVAwYMMHsqWFdTHHNHCCHENAoECWnhUlNTm2zkT5dCocDmzZtx/vx5QZmPjw/i4+PRrl27BtUtFou1m0AIIYQ0HwoECWnB5HK51c8FNuT27dtISEhAQUGBoGzAgAEIDw+HVCptUN0ymQzh4eG0FpAQQloACgQJacHS0tKatD2VSoXDhw9j9+7dgl3IUqkUERER6Nu3r9n1aaZ/6UxgQghpmSgQJKSFkMvlSE1N1Y4AymSyJh0NLC0tRUpKCq5cuSIo8/X1RXx8PLy8vOpVJ50FTAghLRsFgoS0AHK5HCkpKXqjcE0ZBN64cQMJCQkoLi4WlA0ZMgSTJk2CnV39flzQEXCEENLyUSBISAuQlpbWZAmhdalUKhw8eBD79u0D51yvzMHBATNmzEDv3r3rXS8dAUcIIa0DBYKEtABNfUQcABQXFyMpKQkZGRmCsk6dOiEuLg7u7u71rpcxhsjISFoLSAghrQAFgoS0AG5ubk0aDF67dg1JSUkoLS0VlI0cORITJkyAWCw2+jxjTDCCqBEdHU1BICGEtBKi5u4AIQQIDQ2FSGT9/45KpRK7d+/GunXrBEGgo6MjZs6ciYkTJ5oMAgF1sCeRSATXQ0JCKAgkhJBWhEYECbGy2sfDGUuhYm9vb9UNIgUFBUhMTMStW7cEZQEBAYiJiYGrq2ud9chkMm3/zfm6CCGEtFzM2PQOIW1VSEgIb6oj2mofD6dLM70qk8lQUVFh1c0iFy9eREpKCsrLywV9GDt2LMaMGWP2iGRMTAwFfIQQ0oowxk5wzkMMldGIICFWlJaWZvR4OM0fYdYcBayursauXbtw5MgRQZmzszNiY2PRtWtXs+ujqV9CCGlbKBAkxIqaYzewRl5eHjZs2IDMzExBWffu3REdHQ0nJyeTdWhGLWnqlxBC2iYKBAmxoqbeDaxx9uxZbNq0CZWVlXrXRSIRQkNDMXz48DqngiUSCaWBIYSQNo4CQUKsKDQ0FElJSU3WXlVVFbZv344TJ04Iytzc3BAXF4fOnTvXWY9MJkN4eDgFgYQQ0sZRIEiIlZnKuWdJWVlZSEhIQFZWlqCsV69emDFjBmQymck6aAqYEEJsCwWChFiJZsewtYNAzjnS09Oxbds2wcYUsViMyZMnY/DgwWCMATAcmIrFYsyYMYMCQEIIsTEUCBJiJaZ2DFtKRUUFtm7dijNnzgjKPD09ER8fDz8/P73rmpQ1UqmUcgASQoiNo0CQECux9iaRzMxMJCQkIDc3V1AWHByMiIgI2NvbG3xWoVDg9ddft2r/CCGEtHwUCBJiJdbaMcw5x7Fjx7Bjxw4olUq9Mjs7O0ybNg39+/fXTgUb6xshhBBCgSAhVhIYGAhLn2CiUCiwadMmXLhwQVDm4+OD+Ph4tGvXzmQdEokEoaGhFu0XIYSQ1okCQULqwdxzgwHgypUrFm379u3bSEhIQEFBgaBs4MCBiI6OBgC9UUKJRIJ+/frhypUrtB6QEEKIAAWChJip9rnBhYWF2Lx5s7a8doBoqWlhlUqFP//8E2lpaYLziJ2cnLB69Wo88sgj2j6aG6gSQgghrCnymxHSkoSEhPCGTNkuXbrUYHAnk8lQXV2tt0NYIpHAzs6u0ecIl5aWIiUlxeDoYu/evbFp0yZ07969UW0QQghp2xhjJzjnIYbKBCOCixcvdgUQCiBAJBI5MVMrzglphXr06IGDBw/W+7mioiKDGzDKy8sBQK+suroa1dXVJjds1CU/Px/nz59Hhw4d0KFDB72ygIAAzJ49G5mZmQbPEiaWJZFI4O7ujq5duxrdiU0IIa2RdkRw8eLFrjKZbAWAsM6dOyv9/PwcpFKp6cNICWmFFAqFq7Ozc72fq6iosEJvhDjnUCqVqK6uFpQxxmBnZwepVAo7O1rZ0VQqKyuRk5ODO3fuICAgAJMmTYJUKm3ubhFCiFlMjQgyzjkWL17s6uDgsKVfv35B48aNq3JwcKD5YtJmZWdn+/n4+Jh1b1VVFSoqKkyeDtLQI+QMPadSqVBeXi5ICwOoT/9wcHCASCQCYwwNCWZJ41RUVGD//v3Izc1FbGwsBYOEkFbBVCAoAgCJRPJaUFBQn8mTJ1dSEEiIWlVVFcrLy40GeYwxODg4NHiqsHYgV11djbKyMoNBoFQqhUwmg0ikHqSntb3Nw97eHhMnToSrqyuOHTvW3N0hhJBGEy1evJjZ2dk9MmTIkGpaDkjI38yZCpZIJJBIJBCLxfWqu/a5vxUVFVAoFIIAjzEGmUwGe3t7vfWG9H+1+TDGMGjQIFy6dIkCckJIqycC0Esmk7m0a9dOuCCJEBtW1y953XJDo3imaEYRNTuLKysrBfeIxWI4OTkZXAtIGxaaV/v27VFdXW0wpyMhhLQmdgDaubm5KWl3MLFFumsAGWOwt7eHRCIBUPfav4b+l9GMIpaXlyM/P99gGy4uLnB2dgZjzGQfSfNgjMHV1RWlpaXw8PBo7u4QQkiD2QGQ0u5DYos0awA1OOfazyUSCezt7fXKa2voqJy9vT0KCwtRWloqKBOJRPDw8NCrWxM4kpbFzs7O4M5uQghpTSgCJDbL2BrAiooKbfClVCr1EkVrNDQ445wjJyfHYJ329vZwd3ev93pDQgghpKEoECQ2y9i0r+51BwcHiMXiOqdmzUkhYyoVjaurK5ycnGgTCCGEkCbV5hNGX7t2TcwY83vsscfcm7svTcUWv+aGMBZ01b4ukUjg7OyM559/Hg888IDBjR3Gpok1AeLly5fh5eWFF198Ua9cLBbD29tbux6wJThx4gQYY1i9enVzd4UQQoiVmR0IMsb8dN/s7e39vLy82vft29f7iSeecNu4caN9Y9fLHD58WMIY8xs0aJC3ofK1a9fKNO1fuXJFMH9WVlYGBwcHP5lM5mtqbVdb0BqDvZbWZ2PBm6Hrx44dw7p167Bo0SI4OTkJyiUSCRwcHPTSwmhyDJaXlxtcS+bg4AAfH58Wl5R40KBBiIqKwttvv42SkpLm7g4AIDc3F6tXr0Z0dDS6d+8OmUwGNzc3jBo1Ct9++y1UKlWD6r19+zaeeuopdOjQAfb29ggICMDChQuRn59v4a+AEEJapnpPDS9atKgEUKfLKCgoYBcuXJCsX7/ecd26dY79+/ev+umnn/KDgoLql0ujxuDBg6vc3d356dOnJQUFBczd3V1vDi0tLU2qGWHZtWuXfWBgYJlu+b59+6QVFRUYN25cpYODQ0O60CZ07txZKZfLs93d3Rv229FGaKZ3zdmR+9Zbb8HV1RULFiwwWZ/mWc45FAoFCgsLDU4Fu7m5wdHRscWMAtb2xhtvYOjQoVi2bBnefPPN5u4ONmzYgAULFsDPzw/jx4+Hv78/7t+/j6SkJDz99NNITU3Fhg0b6vV6Xrt2DSNGjEBWVhZmzJiBXr164ejRo/jvf/+L7du3448//oCXl5cVvypCCGl+9Q4EP/744+La1+7evSt69tln3TZu3OgwadIkr2PHjuX4+fnVOwgRi8UYOXJkxdatWx3S0tKksbGxeqv59+/fbz9y5MjK8+fP2+3du1f67LPP6gWCaWlp9gAwfvx44dydDZFKpejTpw9tZzSDOZs+Ll++jN27d+Ppp5+GTCars06VSoXCwkIoFApBmUgkgo+PT4vfBTxkyBD06tULK1euxKJFi7QnmjSXHj16YNOmTZg2bZpeXz766CMMGTIEiYmJSEpKQmxsrNl1Pvvss8jKysKyZcvwz3/+U3v95ZdfxhdffIG33noLK1assOjXQQghLY1Ffrp36NBBlZiYmD9y5MjKO3fuiN977z3BIag//vijw/Dhw71cXV19ZTKZX1BQkM+7777rXHsKd8KECRXA30GdxrVr18Q3b94Ujx8/vmLEiBGVBw8eFMzf7d+/3x4AJk6caHA76LVr18QxMTEeXl5e7R0cHPz69+/vnZSUpFfP2bNn7RhjfqNGjTI6FBAUFOQjkUj8bt++bdbr9/vvv0umT5/u0a5du/ZSqdSvY8eO7Z988km3W7duCZ5PSEiwHzNmjFf79u3b29vb+/n6+rYfMWKE1xdffOGoueeNN95w6d69ezsA+OWXX2S6U/YrV66UGZuC1b1+6dIl8fTp0z08PT19nZ2dfcePH++Znp5uBwD37t0TzZkzx619+/btHRwc/AYMGOC9Y8cOwfzlypUrZZGRkR5dunRpJ5PJ/FxcXHyHDh3q/e233wqipbr63JDXylx79+6Vjhw50svFxcXXzs7Ob+DAgfjggw/qnE6sqqpCSUkJVqxYAc650SBj06ZNCA0NhZ+fH+zt7dGhQwdMnToVa9asEdxrbMTxt99+w5gxY+Dm5gaZTIbg4GB8/PHHgp3N169fB2MMc+bMwcWLFxEVFQVPT084OTlh1KhR2Llzp9Gv58iRI4iLi4Ovry+kUik6d+6M+fPn4+7duwbvf+SRR3Dz5k3s2rXLxKvUNCZMmIDIyEhBQOrr64tnnnkGALBv3z6z67t27Rp27tyJgIAAPPfcc3plixcvhpOTE9atW2cwxQ8hhLQlFvszXywW46233ioGgISEBJnuL9lXXnnFZdasWR6XL1+2i4+PV8ydO7eUc47Fixe7hIaGeun+sps4cWIlAOzbt08vQNuxY4c9AISFhVWMGzeu4v79+6IzZ85oRzQLCgpYenq6xN3dnQ8ZMkSQm+PmzZvioUOHet+8eVP88MMPK6KjoxUXL16UxMfHe+oGOX369KkeNWpU5R9//CE9f/68YB3i/v37JRcuXLALDw8v79SpU52jnsuXL5eNGzfOe/fu3fajR4+uWLBgQWn//v0rf/jhB8fBgwf7ZGRkaNv48ssvHePj4z0vXbpkFx4eXv7888+XTJo0qby8vJz98MMP2kBw/PjxFc8880wpADz44IPVixYtKtG8DRw4UJiXpJYbN26IR4wY4Z2VlSV67LHHysaPH1+xf/9++9DQUK8LFy6Ihw0b5n3ixAlpTEyMIioqSnHu3DnJjBkzvHT7CgALFy50v3XrlnjkyJGVCxYsKImNjVXcunVL/PTTT7u//vrrLrr3mtPn+rxW5nr11VddQkNDvQoLC0VPPfVU2ezZs8uKi4vx9ttv41//+hdKSkoMpnLRPWd43759EIvF6Nevn+DeVatWYcaMGTh//jymTJmCefPmYcKECSgvL8dvv/0G4O/kw5qPa3vzzTfx8MMP48KFC3jsscfw/PPPg3OON998E5MnTza4OSUjIwPDhw9HXl4e5s+fj/j4eJw4cQLh4eFYv3694P7vvvsOI0eORGpqKsaPH4+FCxciJCQEq1evRkhICG7evCl4ZuTIkQDQIgJBUzSBdX3yoe7duxcAMGnSJEFw6eLigpEjR6KsrAyHDx+2XEcJIaQlevfddyf+9NNPdzjnd029AeAAuKl7FArFXTs7Ow6AX758+T7n/O7+/fuzAfCOHTtW3759+57m3srKyruTJ08uB8D/9a9/FenW0759eyVjjGdmZmrvj4uLK3NyclJVVlbeTU9PzwLAlyxZUqApT0hIyAXAp02bptCt6+rVq/c1fX/jjTf02tm0aVMuAB4WFlaue/2nn37KA8Cff/75ktpf46OPPloGgG/ZsiW3rtfs3Llz9yUSCQ8ICKi+cePGPd2ybdu25YhEIj516lRtf/v27VsplUr53bt379Wu6/79+/cMfV2PPvpoWe17jZXpvhZvv/223mvx1ltvFQHg7u7uqieffLK0urpaW/a///0vHwBfsGCB3utx8eLF+7XbLi8vvzt69OgKOzs7XvtrNtXn+r5W5rx9/PHHhQD4iy++WKxUKrXXd+7cyWu+Vl5QUMCLiop4ZWUl11VcXMyLiop4ZmYmF4vFPCgoiBcVFfHi4mK9+wYOHMilUim/ePEiv3Pnjt6bXC7n9+/f55WVlTwjI4MD4LNnz9Z7/tChQxwA79y5M8/MzNRer6qq4hERERwA//DDD7XXNfUA4K+++qpeXceOHeN2dnbc3d2dFxYWaq9funSJSyQS3q1bN3779m29Z3bv3s1FIhGPioritRUUFHAAfPDgwYIyY7744gv+zjvvmP2WnJxsdt2GVFVV8T59+nAAfPv27WY/9+qrr/KanyEGy5977jkOgH/zzTdG69iwYQPPyMiob5cJIaTJATjOOYehN4sGgpzzuz4+PkoA/Pfff8/mnN994oknSgHwZcuWFdS+99y5c/dFIhH39/ev1r3+0EMPlQHg69aty9Nca9++vVI3YPP29lbqBn0LFiwoAcA///xzvXY0wUfnzp2rq6qqBP3t1KlTtaenp1L3WmVl5d327dsrPT09lQqFQns9Nzc308HBQRUQEFCtG1gYe9P0KSkpyWDQOHXqVIVYLOYFBQWZvCYQlMlkqpycnMy66m5MIGjotfjrr7/uA+AymUyl6Y/mraqq6q6dnR0fOXJkRV394pzf/eWXX/IA8FWrVuWb2+f6vlZ1vd28efOeo6OjavDgwZW6QS3n/G5mZibv0aMHB8DPnj1rMMArKiriRUVF/MSJExwAHz9+vPaargEDBnCZTMbPnTsnCATz8/O5SqXinHOjgeDTTz/NAfCVK1fy2i5dusRFIhHv2rWr9pqmHjc3N0FfOOd89uzZHABfs2aN9trChQt5zR8vgvs55zwqKoqLxWKD9Tk4OPD27dsbfM6QLl26aANVc95qvx719corr3AAfOrUqfV6bu7cuRwA/9///mew/M033+QA+EcffWS0DgoECSGthalA0OIJpXnNDknNFFh6eroEMLxuLygoSNmhQwflzZs3xfn5+czDw4MD6mnE3377TbZnzx77xx9/vFwul9vdv39f9Pzzz2vrGDVqVOW+ffvslUolxGKxdn3gpEmTDG4U6dOnT5WhqaOOHTsqjx07prf+TSKRYM6cOWWffvqp86+//iqbM2eOAgC+++47x/LycvbUU0+VmrN4/ujRo1JAPc2t+VhXTk6OWKlU4sKFC3bDhg2reuSRRxRvvvmma1BQULvY2FjFuHHjKseMGVPp6+tr0d2/wcHBgteiU6dOSgDo1q2b0s3NTW+bq52dHXx8fFR3797Vm5rNyMgQf/TRR8779u2T3r59W1xeXq4373nnzh2zp3Lr+1rVVd+6detkZWVlbOHChSWGTurQpIDRlPFaO3s1u9Pz8vIAQHuerOb7mnOO0tJSREZG4tSpUxg/fjymT5+O4cOHY8iQIejWrZtZG0tOnjwJQL0GrrYePXqgU6dOyMjIQGFhIdzc3LRlAwcOhIuLi+CZcePGYe3atTh16hRmz54NAPjzzz8BAPv378exY8cEz2RlZUGpVOLy5csYNGiQXpmnpyfu379f59ehcf36dbPvbaxly5bh888/R69evbBu3boma5cQQtoSiwaCCoUCBQUFIgBo3769CgCKiopEANCxY0eDwUz79u1Vt2/fFufn54s8PDyUADB58mS9dYI7d+6UAkBYWJg2yBs7dmxFSkqKw7FjxyQBAQHKCxcu2Pn5+amM7ZatnYpGQywWG9w08Oyzz5YuWbLEedWqVY66gaBUKsXcuXOF20ENyMvLEwHAsmXLhInndJSUlDAAeOONN0q9vb1VK1eudFq5cqXT8uXLnRhjGDFiROVnn31WNHz48DoDIHPUDvSAv9dZubq6Gvx3srOz47q58C5fviwePny4d0FBgWjYsGGVEyZMqHBzc+NisRg3btwQ//LLL7KKigqzc3nU97Wqy+bNmx1EIhGmT59ucOPQ3bt34eTkBD8/PwDCtXua/H+aYE6zqcne3l6TOgkVFRWYP38+PD09sXbtWnz33XdYvXo1GGMYO3YsPvvsM4SEhJjsZ2FhIQBo+1Gbn58fbt68iYKCAr1AsH379gbv9/X11asXUOfgA4DPPvvMZF8M5QxUKBRmBbRN7auvvsKLL76IoKAgpKWlwdPTs17Pa15L3ddJl+a6u7t7o/pJCCEtnUUDwb1790qrq6vh4+Oj6tatmxLQBhbiu3fvinr27CnIL3j//n0RAOjmvOvatasyICBAmZGRIb5+/bpo79699m5ubnzw4MHaQEgTFO7evVsaEBCg5Jxj7Nixhg+PbQB/f3/VlClTyrdu3epw9uxZu9zcXHbhwgW76OjocnNH6DRfe35+/j1jgWhtc+fOVcydO1eRl5fHDhw4IE1OTnb48ccfHadNm+Z1/vz5LEuPDjbUZ5995pyXlydasWJFwfz58/UC47Vr18p++eWXekUPDXmtjKmqqsKpU6eknp6eKmdnZ0Fd6enpuH//PqZMmaLdKFA7ibQmMG7Xrh0AIC8vDw4ODlCpVMjNzdX74yE+Ph7x8fGoqqrC2bNnkZKSgu+++w6TJ0/GxYsX4ePjY7SvmoDk3r176Natm6A8MzNT7z4NY6N09+7dE9yvG/RoNq2YQ6VSoaCgAF27djX7maVLl6KgoMDs+/v374+oqCiz79e08dJLL6FPnz5IS0vT/hvVR8+ePQGoUwMZcuXKFQDqUVlCCGnLLBYIKpVKfPTRRy4AEB8frw0M+vXrV3XmzBlJWlqafc+ePfXy/l28eFF89+5dsb+/v9LT01PvF/a4ceMq1qxZ47h9+3b7P/74w37EiBEVulN8ffr0qW7Xrp1q37599v7+/krg79QzlvLss8+Wbt261eGbb75x1Ix0zp8/3+x8EoMHD646ffq0ZO/evdLo6Oh69c3T05NHRUVVREVFVahUKvz444+Oe/fulT766KPlACAWizmgft2bw19//SUGgEceeURwhMv+/fsNHpVhqs+Nea1qO336tF1FRQUcHByYZumALk1uuCeffNJkEmmJRILu3bvDx8cHV69eRXl5OYqLBWk0wRiDh4cHHBwc0KVLF0ybNg0qlQrfffcdDhw4YDK33YABA3Dy5Ens27dPEAhevXoVt2/fRteuXQUjUydPnkRxcbFgeliTQmXAgAHaa8OGDcOJEydw8OBBTJs2zWhfart06RI45+jfv7/ZzyxduhQ3btww+/7Zs2fXKxD89NNPsWjRIvTv3x+7du2Ct7fBQ4jqNH78eADAzp07oVKp9HYOFxcX448//oCjoyOGDRvWoPoJIaS1sEj6mMzMTFFsbKzHH3/8Ie3UqZPynXfe0c4x/eMf/ygDgE8++cT53r172vaqq6vx8ssvu6pUKsyePbusdp2aoG7ZsmXOBQUFbNy4cYK1f6NHj674888/pZqcg5MnT7ZoIDhlypTKbt26KX/55RfH5ORkh+7du1drpq3N8eKLL5ZKJBK8+uqrboZS0VRUVGDPnj3aoGnHjh1SQ9PU2dnZYgBwdHTUBsuenp6cMYZbt27VO6WKJXTp0kUJqEdkda9v3rzZXjfVjS5Tfa7va2XKsWPHJABQWFjINm/erDfU95///Mdx48aNmDJlCh566CE4OzubTO7MGMPo0aORk5ODM2fOCMqPHj0Kb29v1D7JJisrCwDg6GjwpdB66qmnAAAffPABsrOztdeVSiVeffVVqFQq/OMf/xA8V1hYiPfee0/v2vHjx/HTTz/Bzc0N0dHR2uvPP/88JBIJXnrpJYMjYJWVlTh48KDguiZ1iiZoMsf169cNLkY29mYo16Ix77//PhYtWoRBgwYhLS3NrCDw2rVruHjxoiDtT7du3TBp0iRcv34dX3/9tV7ZO++8g9LSUsyaNcvgcYKEENKW1HtE8I033nABtNNG7MKFC5IjR45IKysrMXDgwKqffvopv127dtpoZuzYsVUvvvhiyX//+1/n4OBgn6ioqHJHR0e+a9cu+wsXLtgNHTq08q233hIsTpo4cWIlYwwXLlywq/lcEOSNGzeuMjExUXbz5k1x9+7dq/39/S06bSoSifD000+XvvHGG64A8NRTTwkCVlP69OlTvXz58oIFCxa49+vXr11oaGhF9+7dq6uqqnDr1i3x4cOH7b28vJRXrlzJBoCHHnrI08nJiYeEhFR26dJFyTnHoUOHpKdOnZL069evKjw8XPsauLq68kGDBlUdPnxYGh8f7x4YGFgtFosRHR1dbmgNoKU9//zzpT/99JPjzJkzPdevX6/w8/NTnT9/3i4tLc0+KiqqPDk5WXDGn6k+Dxw4sF6vlSknT56UAsCUKVMqHn30UY/4+HiFt7e36tChQ9IjR45I+/fvj59//tmsr7O8vBxhYWFISkrC/v379aZJnZ2dMWfOHDg7O2PYsGEICAgA5xwHDx7EsWPHMGjQIISFhZmsf8SIEXj99dfx73//G3369EFcXBycnJyQmpqKs2fPYtSoUXjttdcEz40ZMwarV6/GkSNHMHLkSGRmZmL9+vVQqVRYuXKl3hRwr1698N133+Gpp57Cgw8+iClTpqBHjx6oqqrCzZs3cfDgQfj4+ODixYt6bezcuRNisRgzZsww67WyprVr1+Jf//oXxGIxRo8ejWXLlgnuCQgIwJw5c/SuhYaG4saNG8jIyEBAQIBe2TfffIMRI0bghRdeQFpaGnr37o0jR45g79696NGjBz788EMrfkWEENJC1Dd9jOZNKpVyT09PZd++fSufeOKJ0o0bN+bWTtOh+7Z27dq8oUOHVjg5Oans7e15z549q95+++2isrIyo88EBQVVAeCenp5KQ+laLly4oM2L9/TTT5caqsNUyhLO+d0RI0ZUwERanJycnEyRSMQdHBx4VlaWWalLar+dPHky69FHHy3r1KlTtVQq5e7u7qpevXpVzZkzpzQ1NTVHc9/SpUsLpk6dqvD39692cHBQubu7q/r06VP13nvvFRpKm3LhwoX7kydPLvfw8FAxxjgAvmLFivy60scYey0A8BEjRhhMEdO5c+fqzp0766X52bdvX/aoUaMq3NzcVE5OTqqhQ4dWrF+/Pm/Hjh05APiiRYuKze1zfV8rU2+DBg2qlEqlvLS09O5LL71U7Ovrq5RKpbx79+5Vb7/9dtHNmzd5XVQqFS8sLOR37tzhGRkZ3Nvbmw8YMIDfuXOHZ2ZmcoVCwTnnfPny5TwqKop37dqVy2Qy7uHhwfv3788//fRTvXQsxtLHaPzyyy985MiR3NnZmdvb2/OgoCD+wQcfaNsxVM/58+f59OnTubu7O5fJZHzEiBEmc+mdOXOGz549m/v7+3OpVMo9PDz4gw8+yOfNm8fT0tL07i0oKOAODg58xowZdb5WTeGdd96pMxXN2LFjBc9p0tkYS/Ny8+ZNPmfOHO7r68slEgn39/fnL774Is/Ly6uzT5Q+hhDSWsBE+hj27rvvTgwMDFzz2GOPmX9auw3ZuXOndPLkyV4PPfSQYv369QXN3R9bplAoHEpLSx2rq6slKpVKJBaLlfb29hXOzs4ldnZ2SkC95MDFxcW3R48e1adPn84xVE92drafqQ0c1dXVyM/P15tO/PLLL/HJJ58gLS0NY8eOFaw7bCrXr19H165dMXv27HpNq9bXl19+iRdeeAEHDx7EqFGjrNZOa5aQkICQkBDBSCMhhLQ0jLETnHODaSya9yT5VuCzzz5zBoB//vOfdOhoM+GcIz8/3y0/P9+jsrLSXqVSiQBAqVSKy8rKHLOzs30UCoU9AMjlcrvy8nLWv3//BqXaUSgUyM7OFqwpmzt3Ljp37owvvvii2YLApqJQKPDxxx8jNjaWgkBCCGnjLJ5Qui04deqU3caNGx1Onjwp2b17t/3kyZMrRo0aZZEcfqT+iouLnRUKhSMAODg4lDs7OxeLxWJVZWWlpKioyE2pVIoLCgo87Ozsso8ePSoBgPoGgpxzFBUVobRUGO+LRCJ07NgRP/74I/bu3YvS0tI2vYng+vXrmDdvnmC9HSGEkLaHAkEDjh49Klm8eLGLi4sLnzFjRvmqVasMZ50lVqdUKkWlpaXOAGBvb1/h4eGRr0n+LJPJKiQSSW52drYP55wVFxe7njhxQgkAISEhZgeChqaCNezt7eHu7g6xWIwxY8ZgzJgxlvnCWrDevXvj3Xffbe5uEEIIaQIUCBowf/58Re0kyaR5lJWVyTjnDABcXFyKap8AYmdnp5TJZGVlZWVO5eXlDsuXL7+/atWqonrUj8LCQsERcwDg6uoKJycnwakjzUmzM5kQQgixBBGASkMjIYS0BOXl5Q4AIBaLlVKp1ODxgTKZTKFzv72he2rTnJpRUFAgCKzEYjG8vb3h7OzcooJA0rJUV1fD0PnlhBDSmtgBuFdYWGjHOVfSLz3S0lRXV0sAQCKRGE3kLZVKtX/JVFVVSQGYHM3lnCMnJwe6ZydrODg4wN3dXe+kCUJq45yjsLAQzs7Ozd0VQghpFBGAyxUVFfn37t2jP21Ji1JdXS3STAtr0sMYwhiDSCRS1TxT55be6upqg0Ggm5sbPDw8KAgkdcrMzNSuHyWEkNZM9M477/DKysqf/vzzT7Gh480IaS6aNDEAIBKJTB6qrCnXfcZcdnZ28PHxaXHrAUnLpFKpcOzYMfTs2bO5u0IIIY0mAoDq6urPr1y5cmLLli3S4uJiGg4hLYJmNBBAnQFazSkl4JzX6/tXJpPB29vb5HnDhGiUlJQgNTUVVVVVCAkxmJuVEEJaFaZZKL948WJHBweHLwDMaN++Pff19bWTSqV2NEJCmotSqbRTKBTOACCVShVSqVRw3rRGWVmZs0qlsmOMqZycnAS7hisrK+2rq6ulAKBSqcR1jTCSto9zzjR/QNRxHxQKBc/MzOS5ubns4sWL+cnJybeqqqpo+3bb4Q3A4ElExKa05e+DLpxzg0dqsdo7JhcvXmwPYAwAfwCO1u8bIYbdvXu3Q3p6+iIA6NixY0K/fv0OGLs3LS3ttYqKis729va3Q0ND/22q3rS0tNdCQ0M/s3R/SetSz++DUgA3ARx45513jG5cIq0TY+y4seO3iO2w1e8DQSBISEvBGOsI4HbNp+9xzt8xce9NAJ0B7OacT6yjXpv8z0700fcB0aDvBQLY7vcBrQckLdldACU1H3c1dhNjzB5Ah5pPL1m7U4QQQkhbQYEgabG4erj6VM2nQ03cOgiAJm3MSTOqXtWYfpE2g74PiAZ9LxDARr8PaGqYtGiMsdcAaNb89eWcyw3c8x8ALwFQAfDjnGc1YRcJIYSQVotGBElL9wOAspqPP6xdyBgLADCv5tMkCgIJIYQQ81EgSFo0zvl9/B0ARjLGfmWMBTPGfBhjEQD2AHCCei3hW8bqYYxFMca2M8buM8bKGWN/Mca+YYx1s/5XQayNqfVmjM1mjH3NGDvGGKtgjHHGWHk96mnPGPs3Y+wiY6yMMZbDGNvPGHuSUS6tFo8xJmOMRTPGVjDGTjDGChhjVYyxbMZYGmPsWcaYzIx66PuglWKMdWaMPc8YW8sYO8kYu1Pzs6CYMSZnjC1jjPUysy4Xxti/GGNnap4vYIwdZYwtZIy1meSzNDVMWryaH7yrATxl5JYSAA9zzrdZ8lnSetSMDGcYKa7gnDuYUUcIgK0A2hm5ZTuAKM650XyWpHkxxooAuNRx20UA0znnV4zUQd8HrRhj7HkAX9ZxWxWAlznnX5mopyuA3QAeMHLLCQATOef5DepoC0IjgqTF42r/ABADYAeAbAAVUP/iXwGgv4lA7g38HQT+BqAf1D/gZwC4DsAZwHrGWA+rfQGkqd0GkAzgoLkPMMa8AWyC+nsjF8DjAPwA9ALwdc1tU1D3LxjSvFyg/tnwC4BHAHQD4AmgP4DlADjU/6Y7GWPOtR+m74M2QQF1IP8KgAlQ/9t5A3gQwNMArgKQAPiSMTbNUAWMMSmAzVAHgQoA/wTQCersFe9CvR59EID1Vvw6mg7nnN7orU2+AWgP9YgfB7AFNSPgOuVddco3NHd/6a1R/9YuUAf3vjrX3q35ty034/klNfcqAQwzUP6NTnlwc3+99Gb03/ErAO1NlP9fzb8jB/A6fR/Y3hsAdwB3av4d9xm553md75N4A+X/v717j5mjqsM4/n1aqnjjakUIxYLYKsilgDVGLUht5WLRIJFE5aI1XGK9xKDRYhCNisQLSNSqoCLeBZGC4oWLUIyhgoKAFyDSYhOoLVgslda+2J9/nLO+0+3MvPuW7vvu7jyfZLIzc87Mnp33tPvbM2fO+WAh/djx/kxPdXGLoA2yk0n9BwHOjvwvuCUilgEX583jJVXdCrIeFxGPR8TiiFg52mNzX5935s3FEXFrSbaPkm4nTQBO3/qSWjdFxIJI/YqrfI7U0gdwdDHB9aAZIuIx4Md589CKbGfm1zsi4vKS9AtId6aKefuWA0EbZPPy6/0R8ceKPFfk1wnA67tfJOtBrwZ2zOtXlGWIiNXAzXnzuLEolG17EfEk0OobuEdbsutBcwzl1y36eUraB9gvb1bVgyFgcd6cLamvp+N1IGiDbEZ+XVqT53bSbR6AQ7pbHOtRxb97XV1ppU3JfcmsP+2WX9e27Xc9aABJ2zMcxN9WkmW09WB7hgPHvuRA0AZSnqe41Rn8gap8kZ78ezhvTu92uawntf7um4AHa/IVn0p2XelDkmYwPF1l+61f14MBJWmCpN0lHQcsAfYFNpJu9bcr/k0rvzsYoHqw3XgXwKxLir/UV1fmSlaRngjbtXvFsR7Wqitr8q3DKsXByl1X+tNn8muw5XRirgcDRtIvgNeVJN0LnB4RvytJ6/S7Y2DqgVsEbVA9q7A+0oDC6/PrFsNJWCO06kqn9QRcV/pOnq5ydt5cFFtOV+l60AyrgQtJ4wCW6fS7Y2DqgQNBMzMbaJJeB5yXN+8GzhrH4tjYeSNpaKkdSWMCngw8QhpTcmkeNLrxHAjaoPp3YX2kWSVaU06t61JZrLe16kqn9QRcV/qGpEOBy4GJwArSuG/rS7K6HgyYiNgQEesiYm1ELIuIb5OGjFlKesBjsaT2OKjT746BqQcOBG1QPVJYnzxC3lb6o7W5bFC16spOkur6TRfrketKH8gzBv2c1Cq0GpgbESsqsrseNED+EfChvHkAafaRok6/OwamHjgQtEH1EMO/0iqb/yU9neHxxO7tdqGsJ7X+7hOBvWryFeuR60qPkzQFuI70hb0WOCoi/lpziOtBcxSHhZnRllb8m9bdOh6YeuBA0AZSnkXkjrz58pqsh5L+4wf4Q1cLZb2q+HevqyuttBUR8UhNPhtnkiaTgsC9SJ3650XESP++XQ+ao9jiG21po60HG4A/b4tCjRcHgjbIrsmv0yQdUJHnhPy6iTQfsTXPLcC/8voJZRnywMGH582rx6JQtnUk7QD8kjS22xBwQkQs6eBQ14PmmFVY32yswIh4gOHArqoebEea2xzghoh4YpuXcAw5ELRBdhnQ+gf6yfZESVOB0/LmlRGxqj2PDb48XdQlefMNkmaWZDsXeBqp9eCrY1Q0G6U8a8Q1pNt9m4CTIuLaTo51PRgMkl48QvrOwPl583Hg+pJsi/LrIZLeVJL+PuB5bXn7ltIdNLPBJGkhw0HgD/P6SlKz/kWkfh7rgEMj4r5xKaRtE5L2A3Yo7HonMJ80g8DhbdnvyLPKtI59LnAXsDups/h7gRvy+d6dF4CLI+I0rOdImghcyfD0Ye8HLq45ZFN7S47rQf+T9CTpx8BPSGMF/oP0o2AP0oMhZwFTcvYFEfGlknM8jXSLeH9S14KzgKuAScApwDmkLkXXRcTcLn6cMeFA0AaaJJF+5b+jIss64MROWw2sd0m6iS0Dvip7R8TytuMPA37G8C/9dr8A3lgMIK135Bb+ZSPlK3gwIqaWnMf1oI9J6iSo2QAsjIgLas6zN6m1cJ+KLL8H5kTEmtGXsrf41rANtEjmA8eT+g2tBv5D+sL4CnCwg0ADiIjbgQOBzwL3kVoC/knqO/YO4Bh/+Q8+14O+N4t0C/9G4G+kH/sbSf/3LwE+BkyvCwIBImIZcDBpPuK7SeMLrgVuJ7U2v2IQgkBwi6CZmZlZY7lF0MzMzKyhHAiamZmZNZQDQTMzM7OGciBoZmZm1lAOBM3MzMwayoGgmZmZWUM5EDQzMzNrKAeCZmZmZg3lQNDMzMysobYb7wKYmQ26Duc/rRQR2lZlMTMrcougmZk9JZJOlRRPNeAdFJKmtq6HpCPGuzxmdRwImpl133MqlmMKec6oyWdm1hW+NWxm1mURsa5sv6T1hc3/VOUzM+sWtwiamZmZNZQDQTOzHiXp3NzPbHnePkjSZZL+LmmjpDvb8kvSiZKulvRQzvOopBsknSKp8v98SQdK+oikJZJWSxqStEbS0rx/p5JjpuZ+gd8s7Iu25aZC2hGF/VMl7SzpfEn3SXpC0gpJ35A0pXDMREmn5XI8Jmlt/jyHd3D9dsnX8Lb8WTZIWi7pUkkH1hx3Uy7jpXl7jqSf5+uyQdJfJJ0j6Rklxy4HlhV2/br9moxUbrOx5FvDZmZ9QNKbgO8CT69I3xW4EpjVlrQLcGRe3irp+PZb0JIOAu4sOe1OwMy8zJc0NyLufwofo2gv4FvA1MK+PYG3A3MkvRJYBVwBHNt27JHALElviIhry04uaQ7wo/wZil4AnAKcJGlBRCyqK6SkDwGfAopPbr8Y+BgwW9LsiHiy7hxmvcwtgmZmvW9n4FLgr8BxwG6kQOrDAJImAdeSgsA1wAeA/UhB4LScbz0wB7i45PwBXA8sAF4FvBB4LnAA8B7g76SA7QeSigHRg6SHWc4o7Gt/0OXois/0LWAScBKwR14WAEOkgPB84DxgNrAQeFEu0zzgIVJDxtfyZ9+MpMOAn5KCwFuB4/M5dwVeCSwmff99SdJRFeUDOJwUBH6fFAzvCuwPfC+nzwJObztmv5yn5ZiSa2LWOyLCixcvXryMwwIcQQrCAji1JP3cQvqfgGdXnOesnOdRYFpFntcWzjVzlOV8PvDPfOxrS9JPbZ17FJ93DTC1JM/Hc/p/8zJvhM9yVFuagLtz2lXAxIqyXJbz3AOoLe2mwvm/WHKsgNty+u9K0qcWjj9ivOuZFy91i1sEzcz6wzlR/VTxu/PrpyPivrIMEXE9cGPefMto3jgiVpJaDCG10G0LF0XE8pL9P8yvE4BbIuKakjw3AI/k9Zltaa8BXkoKIs+MiP9WvP/Z+XV/4KCKPP8mtUZuJiIC+E7ePLisVdKsX7iPoJlZ7wvgl2UJkl5Euk0McLOkZ9ec5y5S/7rDSs4zATgxL4cAk4HtS84xrfNi1/pVxf4HCuvXlWWIiJD0AOlW8fPbkluB6t3A4zXXYw2wmvQ5D6O8j+StEbG24vhWX8lJpFv3qyrymfU0B4JmZr1vdU1r4PTC+tIOzze5uCHpOaQ+de0PmpTZscP3GMnKsp0Rsb7QDbE0T9Yag7H9yd3W9TgYeLzDskyu2P9wzTFPFNa3eHrYrF/41rCZWe9bX5O2NYFZ+5PHF5CCwAAuAY4C9iY9bNJ6wKH1gMS2akCoumU72jzt8zBvi+sxmvcvK4NZ33CLoJlZfyu2FO4TEcsqc5aQ9CzgbXnzvIg4uyZfP2hdjyURMeJYg2ZN5xZBM7P+VuxTt89WHD+d4RaxH9Xke+lWnHs8tK7H1lwLs8ZxIGhm1t/uYbgv3Zu34vjibdGJZRkkzSSNLVhlqJC39BxjqPWAyZ6SXjFOZRgqrI/39TCr5UDQzKyP5aFMLsib8yUdU5df0g6Sdi/sWl5Yn1eS/5nAl0coxqOF9d0rc42NX5HGXIQ04PQudZklTa9L30prSP0tYfyvh1ktB4JmZv3vQuC3pNanqyUtkvRqSc/L8/lOk3SCpK8DK0izawAQEQ8Dv8mbCyUtlLSvpMk5qPwNMAO4t+b97wQ25fVzJU2RNEnSdmPdQhgRm0hTyG0g3c7+o6T3SHpJvha7SXqZpHdJuhG4vQtleII0CwzAAkkHSHpGvh7um289xRXSzKzPRcRGSceSBjk+ljTl2xk1h2xs2z4TuIU0Jdsn8/L/05NmLjmQzYeqKb7/SkmXk8YgnJ+XlptJM4qMmYj4vaS5pD6PewJfqMm+pkvFuAhYBLycNH5jkZ8ytp7hFkEzswEQEY9FxOuBucB3gWWkYWeGSH0IbwY+AcyIiKvbjr2HNKjyt3PeIdIYelcBR0bE5zsowqmkKfHuYvMx9sZFRNwC7Au8jzSjyirgSVLZ7gd+TAqW9+3S+38FOJkUYD/GcIupWU9R6l5iZmZmZk3jFkEzMzOzhnIgaGZmZtZQDgTNzMzMGsqBoJmZmVlDORA0MzMzaygHgmZmZmYN5UDQzMzMrKEcCJqZmZk1lANBMzMzs4ZyIGhmZmbWUA4EzczMzBrKgaCZmZlZQ/0PaYFQ91GsCwwAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] @@ -1244,11 +894,11 @@ "Estimand type: nonparametric-ate\n", "\n", "## Realized estimand\n", - "b: Outcome~Action+w4+w8+w1+w0+w6+w7+w3+w5+w9+w2 | \n", + "b: Outcome~Action+w1+w7+w5+w0+w6+w9+w8+w4+w3+w2 | \n", "Target units: ate\n", "\n", "## Estimate\n", - "Mean value: 2.209422956336211\n", + "Mean value: 1.55024679537048\n", "\n" ] }, @@ -1304,8 +954,8 @@ "output_type": "stream", "text": [ "Refute: Add a Random Common Cause\n", - "Estimated effect:2.209422956336211\n", - "New effect:2.1905981166894186\n", + "Estimated effect:1.55024679537048\n", + "New effect:1.5912134348688016\n", "\n" ] }, @@ -1467,19 +1117,11 @@ "output_type": "stream", "text": [ "Refute: Use a Placebo Treatment\n", - "Estimated effect:2.209422956336211\n", - "New effect:0.0\n", - "p value:nan\n", + "Estimated effect:1.55024679537048\n", + "New effect:0.00019073641916635391\n", + "p value:0.37178746512481153\n", "\n" ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/mnt/c/Users/amshar.FAREAST/code/dowhy/dowhy/causal_refuter.py:210: RuntimeWarning: invalid value encountered in double_scalars\n", - " z_score = (estimate.value - mean_refute_values)/ std_dev_refute_values\n" - ] } ], "source": [ @@ -1494,9 +1136,7 @@ "metadata": {}, "source": [ "## Case-studies using DoWhy+EconML\n", - "However, as the data becomes high-dimensional, simple regression estimators will not estimate the correct causal effect. This is because they are optimized for minimizing the predictive loss, not to estimate the change due to any particular input variable. More advanced supervised machine learning models also do not work and often are worse than simple regression, because they include additional regularization techniques that help in minimizing predictive error, but can have unwanted effects on estimating the causal effect. \n", - "\n", - "Therefore, we need methods targeted to estimate the causal effect. \n", + "In practice, as the data becomes high-dimensional, simple estimators will not estimate the correct causal effect. More advanced supervised machine learning models also do not work and often are worse than simple regression, because they include additional regularization techniques that help in minimizing predictive error, but can have unwanted effects on estimating the causal effect. Therefore, we need methods targeted to estimate the causal effect. At the same time, we also need suitable refutation methods that can check the robustness of the estimate. \n", "\n", "\n", "Here is an example of using DoWhy+EconML for a high-dimensional dataset.\n",