From d4a64ba77f0bd3d19f330bd008b358f24f45bb3f Mon Sep 17 00:00:00 2001 From: Dan Date: Wed, 24 Mar 2021 15:40:03 +0100 Subject: [PATCH] typos and rewordings --- CONTRIBUTING.md | 18 +- docs/api/feature_elimination.md | 2 +- docs/api/imputation_selector.md | 2 +- docs/api/metric_volatility.md | 8 +- docs/api/model_interpret.md | 4 +- docs/api/sample_similarity.md | 8 +- docs/api/stat_tests.md | 2 +- docs/api/utils.md | 2 +- docs/howto/reproducibility.ipynb | 20 +- docs/tutorials/nb_binning.ipynb | 8 +- .../nb_distribution_statistics.ipynb | 14 +- docs/tutorials/nb_imputation_comparison.ipynb | 4128 ++++- docs/tutorials/nb_metric_volatility.ipynb | 16 +- docs/tutorials/nb_sample_similarity.ipynb | 18 +- docs/tutorials/nb_shap_dependence.ipynb | 7 - .../nb_shap_feature_elimination.ipynb | 24 +- .../tutorials/nb_shap_model_interpreter.ipynb | 14820 +++++++++++++++- .../feature_elimination.py | 16 +- probatus/interpret/inspector.py | 18 +- probatus/interpret/model_interpret.py | 8 +- probatus/interpret/shap_dependence.py | 2 +- probatus/metric_volatility/volatility.py | 8 +- .../sample_similarity/resemblance_model.py | 79 +- 23 files changed, 18994 insertions(+), 238 deletions(-) diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index 05078d5b..0abd1ad3 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -5,7 +5,7 @@ We're very much open to contributions but there are some things to keep in mind: - Discuss the feature and implementation you want to add on Github before you write a PR for it. On disagreements, maintainer(s) will have the final word. - Features need a somewhat general usecase. If the usecase is very niche it will be hard for us to consider maintaining it. -- If you’re going to add a feature consider if you could help out in the maintenance of it. +- If you’re going to add a feature, consider if you could help out in the maintenance of it. - When issues or pull requests are not going to be resolved or merged, they should be closed as soon as possible. This is kinder than deciding this after a long period. Our issue tracker should reflect work to be done. That said, there are many ways to contribute to probatus, including: @@ -49,20 +49,20 @@ pre-commit install ### Code structure -* Model validation modules assume that trained models passed for validation are developed in scikit-learn framework (have predict_proba and other standard functions), or follows scikit-learn API e.g. XGBoost. -* Every python file used for model validation, needs to be in `/probatus/` -* Class structure for a given module should have a base class, and specific functionality classes that inherit from base. If a given module implements only single way of computing the output, the base class is not required. -* Functions should not be as short a possible lines of code. If a lot of code is needed, try to put together snippets of code into +* Model validation modules assume that trained models passed for validation are developed in a scikit-learn framework (i.e. have predict_proba and other standard functions), or follow a scikit-learn API e.g. XGBoost. +* Every python file used for model validation needs to be in `/probatus/` +* Class structure for a given module should have a base class and specific functionality classes that inherit from base. If a given module implements only a single way of computing the output, the base class is not required. +* Functions should not be as short as possible in terms of lines of code. If a lot of code is needed, try to put together snippets of code into other functions. This make the code more readable, and easier to test. * Classes follow the probatus API structure: - * Each class implements fit(), compute() and fit_compute() methods. Fit is used to fit object with provided data (unless no fit is required), and compute calculates the output e.g. DataFrame with report for the user. Lastly, fit_compute applies one after the other. - * If applicable, plot() method presents user with the appropriate graphs. - * For compute(), and plot(), check if the object is fitted first. + * Each class implements `fit()`, `compute()` and `fit_compute()` methods. `fit()` is used to fit an object with provided data (unless no fit is required), and `compute()` calculates the output e.g. DataFrame with a report for the user. Lastly, `fit_compute()` applies one after the other. + * If applicable, the `plot()` method presents the user with the appropriate graphs. + * For `compute()` and `plot()`, check if the object is fitted first. ### Documentation -Documentation is a very crucial part of the project, because it ensures usability of the package. We develop the docs in the following way: +Documentation is a very crucial part of the project because it ensures usability of the package. We develop the docs in the following way: * We use [mkdocs](https://www.mkdocs.org/) with [mkdocs-material](https://squidfunk.github.io/mkdocs-material/) theme. The `docs/` folder contains all the relevant documentation. * We use `mkdocs serve` to view the documentation locally. Use it to test the documentation everytime you make any changes. diff --git a/docs/api/feature_elimination.md b/docs/api/feature_elimination.md index a92b3702..74fbe0d3 100644 --- a/docs/api/feature_elimination.md +++ b/docs/api/feature_elimination.md @@ -1,6 +1,6 @@ # Features Elimination -This module allows to apply features elimination. +This module allows us to apply features elimination. ::: probatus.feature_elimination.feature_elimination diff --git a/docs/api/imputation_selector.md b/docs/api/imputation_selector.md index 1e223f7f..d4fc675f 100644 --- a/docs/api/imputation_selector.md +++ b/docs/api/imputation_selector.md @@ -1,6 +1,6 @@ # Imputation Selector -This module allows to select imputation strategies. +This module allows us to select imputation strategies. ::: probatus.missing_values.imputation diff --git a/docs/api/metric_volatility.md b/docs/api/metric_volatility.md index 1d61098e..0bfa13ac 100644 --- a/docs/api/metric_volatility.md +++ b/docs/api/metric_volatility.md @@ -1,14 +1,14 @@ # Metric Volatility -The aim of this module is analysis of how well a model performs on a given dataset, and how stable the performance is. +The aim of this module is the analysis of how well a model performs on a given dataset, and how stable the performance is. The following features are implemented: -- **TrainTestVolatility** - Estimation of volatility of metrics. The estimation is done by splitting the data into train and test multiple times and training and scoring a model based on these metrics. +- **TrainTestVolatility** - Estimation of the volatility of metrics. The estimation is done by splitting the data into train and test multiple times and training and scoring a model based on these metrics. - - **SplitSeedVolatility** - Estimates volatility of metrics based on splitting the data into train and test sets multiple times randomly, each time with different seed. + - **SplitSeedVolatility** - Estimates the volatility of metrics based on splitting the data into train and test sets multiple times randomly, each time with a different seed. - - **BootstrappedVolatility** - Estimates volatility of metrics based on splitting the data into train and test with static seed, and bootstrapping train and test set. + - **BootstrappedVolatility** - Estimates the volatility of metrics based on splitting the data into train and test with static seed, and bootstrapping the train and test set. ::: probatus.metric_volatility.volatility \ No newline at end of file diff --git a/docs/api/model_interpret.md b/docs/api/model_interpret.md index c48263ab..fb2b4f0c 100644 --- a/docs/api/model_interpret.md +++ b/docs/api/model_interpret.md @@ -1,7 +1,7 @@ # Model Interpretation using SHAP -The aim of this module is providing tools for model interpretation using [SHAP](https://shap.readthedocs.io/en/latest/) library. -The class below is a convenience wrapper, that implements multiple plots for tree-based & linear models. +The aim of this module is to provide tools for model interpretation using the [SHAP](https://shap.readthedocs.io/en/latest/) library. +The class below is a convenience wrapper that implements multiple plots for tree-based & linear models. ::: probatus.interpret.model_interpret ::: probatus.interpret.shap_dependence diff --git a/docs/api/sample_similarity.md b/docs/api/sample_similarity.md index 87789296..77ccf433 100644 --- a/docs/api/sample_similarity.md +++ b/docs/api/sample_similarity.md @@ -2,17 +2,17 @@ The goal of sample similarity module is understanding how different two samples are from a multivariate perspective. -One of the ways to indicate that is Resemblance Model. Having two datasets say X1 and X2, one can analyse how easy is it to recognize which dataset a randomly selected row comes from. The Resemblance model assigns label 0 to X1 dataset, and label 1 to X2 and trains a binary classification model that to predict, which sample a given row comes from. -By looking at the test AUC, one can conclude that the samples have different distribution the AUC is significantly higher than 0.5. Further, by analysing feature importance one can understand, which of the features have predictive power. +One of the ways to indicate this is Resemblance Model. Having two datasets - say X1 and X2 - one can analyse how easy it is to recognize which dataset a randomly selected row comes from. The Resemblance model assigns label 0 to the dataset X1, and label 1 to X2 and trains a binary classification model to predict which sample a given row comes from. +By looking at the test AUC, one can conclude that the samples have a different distribution if the AUC is significantly higher than 0.5. Furthermore, by analysing feature importance one can understand which of the features have predictive power. The following features are implemented: -- **SHAPImportanceResemblance (Recommended)** - The class applies SHAP library, in order to interpret the tree based resemblance model model. +- **SHAPImportanceResemblance (Recommended)** - The class applies SHAP library, in order to interpret the tree based resemblance model. -- **PermutationImportanceResemblance** - The class applies permutation feature importance, in order to understand, which features does the current model rely the most on. The higher the importance of the feature, the more a given feature possibly differs in X2 compared to X1. The importance indicates how much the test AUC drops if a given feature is permuted. +- **PermutationImportanceResemblance** - The class applies permutation feature importance in order to understand which features the current model relies on the most. The higher the importance of the feature, the more a given feature possibly differs in X2 compared to X1. The importance indicates how much the test AUC drops if a given feature is permuted. ::: probatus.sample_similarity.resemblance_model diff --git a/docs/api/stat_tests.md b/docs/api/stat_tests.md index 09727577..8fce6fa3 100644 --- a/docs/api/stat_tests.md +++ b/docs/api/stat_tests.md @@ -1,5 +1,5 @@ # Statistical Tests -This module allows to apply different statistical tests. +This module allows us to apply different statistical tests. ::: probatus.stat_tests.distribution_statistics diff --git a/docs/api/utils.md b/docs/api/utils.md index 0768e5f6..51190fc5 100644 --- a/docs/api/utils.md +++ b/docs/api/utils.md @@ -1,6 +1,6 @@ # Utility Functions -This module contains various smaller functionalities, that can be used across the `probatus` package. +This module contains various smaller functionalities that can be used across the `probatus` package. ::: probatus.utils.scoring diff --git a/docs/howto/reproducibility.ipynb b/docs/howto/reproducibility.ipynb index 471f4930..14794ce4 100644 --- a/docs/howto/reproducibility.ipynb +++ b/docs/howto/reproducibility.ipynb @@ -18,7 +18,7 @@ "- Inputs of `probatus` modules,\n", "- The `random_state` of `probatus` modules.\n", "\n", - "Below sections cover how to ensure reproducibility of the results, by controling these aspects\n", + "The below sections cover how to ensure reproducibility of the results by controling these aspects.\n", "\n", "## Inputs of probatus modules\n", "\n", @@ -26,9 +26,9 @@ "\n", "### Static dataset\n", "\n", - "When using `probatus`, one of the most crucial aspects is the provided dataset. Therefore, the first thing to do, is to ensure that the passed dataset does not change along the way. \n", + "When using `probatus`, one of the most crucial aspects is the provided dataset. Therefore, the first thing to do is to ensure that the passed dataset does not change along the way. \n", "\n", - "Below is a code snipped of random data preparation. In sklearn, you can ensure this by setting `random_state` parameter. Possibly in your projects, you will use a different dataset, however, always make sure that the input data is static." + "Below is a code snipped of random data preparation. In sklearn, you can ensure this by setting the `random_state` parameter. You will probably use a different dataset in your projects, but always make sure that the input data is static." ] }, { @@ -50,9 +50,9 @@ "\n", "Whenever you split the data in any way, you need to make sure that the splits are always the same. \n", "\n", - "If you use `train_test_split` functionality from sklearn, this can be enforced by setting the `random_state` parameter. \n", + "If you use the `train_test_split` functionality from sklearn, this can be enforced by setting the `random_state` parameter. \n", "\n", - "Another crucial aspect, is how you use the `cv` parameter, which defines the folds settings that you will use in the experiments. If the `cv` is set to integer, you don't need to worry about it, the `random_state` of `probatus` will take care of it. However, if you want to pass a custom cv generator object, you have to set the `random_state` there as well.\n", + "Another crucial aspect is how you use the `cv` parameter, which defines the folds settings that you will use in the experiments. If the `cv` is set to an integer, you don't need to worry about it - the `random_state` of `probatus` will take care of it. However, if you want to pass a custom cv generator object, you have to set the `random_state` there as well.\n", "\n", "Below are some examples of static splits:" ] @@ -80,7 +80,7 @@ "source": [ "### Static classifier\n", "\n", - "Most of `probatus` modules work with the provided classifiers. Whenever, one needs to provide a not fitted classifier, it is enough to set the `random_state`. However, if the classifier needs to be fitted beforehand, you have to make sure that the model training is reproducible as well." + "Most of `probatus` modules work with the provided classifiers. Whenever one needs to provide a not-fitted classifier, it is enough to set the `random_state`. However, if the classifier needs to be fitted beforehand, you have to make sure that the model training is reproducible as well." ] }, { @@ -100,7 +100,7 @@ "source": [ "### Static search CV for hyperparameter tuning\n", "\n", - "Some of the modules e.g. `ShapRFECV`, allow you to perform optimization of the model. Whenever, you use such functionality, make sure that the these classes have set `random_state`. This way, in every round of optimization, you will explore the same set of parameters permutations. In case the search space is also generated based on randomness, make sure that the `random_state` is set to it as well." + "Some of the modules e.g. `ShapRFECV`, allow you to perform optimization of the model. Whenever, you use such functionality, make sure that these classes have set the `random_state`. This way, in every round of optimization, you will explore the same set of parameter permutations. In case the search space is also generated based on randomness, make sure that the `random_state` is set to it as well." ] }, { @@ -129,7 +129,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Before running `probatus` modules think about the inputs, and consider if there is any other type of randomness involved. If there is, one option to possibly solve the issue is setting the random seed at the beginning of the code" + "Before running `probatus` modules think about the inputs, and consider if there is any other type of randomness involved. If there is, one option to possibly solve the issue is setting the random seed at the beginning of the code." ] }, { @@ -149,7 +149,7 @@ "source": [ "## Reproducibility in probatus\n", "\n", - "Most of the modules in `probatus` allow you to set the `random_state`. This setting essentially, makes sure that any code that the functions operate on, has a static flow. So as long as set it and you ensure all other inputs do not cause an additional fluctuations between runs, you can make sure that your results are reproducible" + "Most of the modules in `probatus` allow you to set the `random_state`. This setting essentially makes sure that any code that the functions operate on has a static flow. As long as it is seet and you ensure all other inputs do not cause additional fluctuations between runs, you can make sure that your results are reproducible." ] }, { @@ -299,4 +299,4 @@ }, "nbformat": 4, "nbformat_minor": 4 -} +} \ No newline at end of file diff --git a/docs/tutorials/nb_binning.ipynb b/docs/tutorials/nb_binning.ipynb index 58162ed4..f64cd072 100644 --- a/docs/tutorials/nb_binning.ipynb +++ b/docs/tutorials/nb_binning.ipynb @@ -99,7 +99,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The `SimpleBucketer` object creates binning of the values in `x` into equally sized bins. The attributes `counts`, the number of elements per bin, and `boundaries`, the actual boundaries that resulted from the binning strategy are assigned to the object instance. In this example we choose to get 4 bins:" + "The `SimpleBucketer` object creates binning of the values of `x` into equally sized bins. The attributes `counts`, the number of elements per bin, and `boundaries`, the actual boundaries that resulted from the binning strategy, are assigned to the object instance. In this example we choose to get 4 bins:" ] }, { @@ -589,9 +589,9 @@ "metadata": {}, "source": [ "Comparing the `TreeBucketer` and the `QuantileBucketer` (the dots compare the average distribution of class 1 in the bin):
\n", - "Each buckets obtained by the `TreeBucketer` follow the probability distribution (ie the entries in the bucket have the same probability of being class 1).
\n", + "Each buckets obtained by the `TreeBucketer` follow the probability distribution (i.e. the entries in the bucket have the same probability of being class 1).
\n", "On the contrary, the `QuantileBucketer` splits the values below 4 in 6 buckets, which all have the same probability of being class 1.
\n", - "Note also that the tree is grown with the maximum depth of 4, which potentially let's it grow up to 16 buckets ($2^4$).
\n", + "Note also that the tree is grown with the maximum depth of 4, which potentially lets it grow up to 16 buckets ($2^4$).
\n", "\n", "The learned tree is visualized below, whreere the splitting according to the step function is visualized clearly.\n", "\n" @@ -643,4 +643,4 @@ }, "nbformat": 4, "nbformat_minor": 4 -} \ No newline at end of file +} diff --git a/docs/tutorials/nb_distribution_statistics.ipynb b/docs/tutorials/nb_distribution_statistics.ipynb index dc657b02..848c1a83 100644 --- a/docs/tutorials/nb_distribution_statistics.ipynb +++ b/docs/tutorials/nb_distribution_statistics.ipynb @@ -193,9 +193,9 @@ "metadata": {}, "source": [ "## PSI - Population Stability Index\n", - "The population stability index ([Karakoulas, 2004](https://cms.rmau.org/uploadedFiles/Credit_Risk/Library/RMA_Journal/Other_Topics_(1998_to_present)/Empirical%20Validation%20of%20Retail%20Credit-Scoring%20Models.pdf)) has long been used to evaluate distribution similarity in banking industry, while developing credit decision models.\n", + "The population stability index ([Karakoulas, 2004](https://cms.rmau.org/uploadedFiles/Credit_Risk/Library/RMA_Journal/Other_Topics_(1998_to_present)/Empirical%20Validation%20of%20Retail%20Credit-Scoring%20Models.pdf)) has long been used to evaluate distribution similarity in the banking industry, while developing credit decision models.\n", "\n", - "In probatus we have implemented the PSI according to [Yurdakul 2018](https://scholarworks.wmich.edu/cgi/viewcontent.cgi?article=4249&context=dissertations), which derives a p-value, based on the hard to interpret PSI statistic. Using the p-value is a more reliable choice, because the banking industry-standard PSI critical values of 0.1 and 0.25 are unreliable heuristics because there is a strong dependency on sample sizes and number of bins. Aside from these heuristics, the PSI value is not easily interpretable in the context of common statistical frameworks (like a p-value or confidence levels)." + "In `probatus` we have implemented the PSI according to [Yurdakul 2018](https://scholarworks.wmich.edu/cgi/viewcontent.cgi?article=4249&context=dissertations), which derives a p-value, based on the hard to interpret PSI statistic. Using the p-value is a more reliable choice, because the banking industry-standard PSI critical values of 0.1 and 0.25 are unreliable heuristics as there is a strong dependency on sample sizes and number of bins. Aside from these heuristics, the PSI value is not easily interpretable in the context of common statistical frameworks (like a p-value or confidence levels)." ] }, { @@ -226,8 +226,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Bssed on the above test, the distribution between the two samples significantly differ.\n", - "Not only the PSI statistic is above the commonly used critical value, but also the p-value shows a very high confidence." + "Based on the above test, the distribution between the two samples significantly differ.\n", + "Not only is the PSI statistic above the commonly used critical value, but also the p-value shows a very high confidence." ] }, { @@ -241,7 +241,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Using `DistributionStatistics` class one can apply the above test, without the need to manually perform the binning. We initialize a `DistributionStatistics` instance with the desired test, binning_strategy (or choose `\"default\"` to choose the test's most appropriate binning strategy) and the number of bins. Then we start the test with the unbinned values as input." + "Using the `DistributionStatistics` class one can apply the above test without the need to manually perform the binning. We initialize a `DistributionStatistics` instance with the desired test, binning_strategy (or choose `\"default\"` to choose the test's most appropriate binning strategy) and the number of bins. Then we start the test with the unbinned values as input." ] }, { @@ -280,7 +280,7 @@ "\n", "The main advantage of this method is its sensitivity to differences in both location and shape of the empirical cumulative distribution functions of the two samples.\n", "\n", - "The main disadwantages are that: it works for continuous distributions (unless modified, e.g. see ([Jeng 2006](https://bmcmedresmethodol.biomedcentral.com/track/pdf/10.1186/1471-2288-6-45)), in large samples, small and unimportant differences can be statistically significant ([Taplin & Hunt 2019](https://www.mdpi.com/2227-9091/7/2/53/pdf)), and finally in small samples, large and important differences can be statistically insignificant ([Taplin & Hunt 2019](https://www.mdpi.com/2227-9091/7/2/53/pdf))." + "The main disadvantages are that: it works for continuous distributions (unless modified, e.g. see ([Jeng 2006](https://bmcmedresmethodol.biomedcentral.com/track/pdf/10.1186/1471-2288-6-45))); in large samples, small and unimportant differences can be statistically significant ([Taplin & Hunt 2019](https://www.mdpi.com/2227-9091/7/2/53/pdf)); and finally in small samples, large and important differences can be statistically insignificant ([Taplin & Hunt 2019](https://www.mdpi.com/2227-9091/7/2/53/pdf))." ] }, { @@ -507,4 +507,4 @@ }, "nbformat": 4, "nbformat_minor": 4 -} \ No newline at end of file +} diff --git a/docs/tutorials/nb_imputation_comparison.ipynb b/docs/tutorials/nb_imputation_comparison.ipynb index 24b56de5..a094c6ef 100644 --- a/docs/tutorials/nb_imputation_comparison.ipynb +++ b/docs/tutorials/nb_imputation_comparison.ipynb @@ -15,8 +15,8 @@ "source": [ "This notebook explains how the `ImputationSelector` class works in `probatus`. With `ImputationSelector` you can compare multiple imputation strategies\n", "and choose a strategy which works the best for a given model and a dataset.\n", - "Currently `ImputationSelector` supports any [scikit-learn](https://scikit-learn.org/stable/) compatible imputation strategy. For categorical variables the missing values are replaced by a `missing` token and `OneHotEncoder` is applied. The user supplied imputation strategies are applied to numerical columns only. \n", - "Support for user supplied imputation strategies for categorical columns can be added in the future releases.\n", + "Currently `ImputationSelector` supports any [scikit-learn](https://scikit-learn.org/stable/) compatible imputation strategy. For categorical variables the missing values are replaced by a `missing` token and `OneHotEncoder` is applied. The user-supplied imputation strategies are applied to numerical columns only. \n", + "Support for user-supplied imputation strategies for categorical columns can be added in the future releases.\n", "\n", "Let us look at an example and start by importing all the required classes and methods.\n", "\n" @@ -73,8 +73,8 @@ "metadata": {}, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "Shape of X,y : (2000, 20),(2000,)\n" ] @@ -93,8 +93,76 @@ "metadata": {}, "outputs": [ { - "output_type": "execute_result", "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
f_0f_1f_2f_3f_4f_5f_6f_7f_8f_9f_10f_11f_12f_13f_14f_15f_16f_17f_18f_19
Missing %0.2260.19850.1860.2130.2010.20450.1960.19950.20950.1950.20150.2010.18350.1760.1990.1930.19650.2120.2020.21
\n", + "
" + ], "text/plain": [ " f_0 f_1 f_2 f_3 f_4 f_5 f_6 f_7 f_8 \\\n", "Missing % 0.226 0.1985 0.186 0.213 0.201 0.2045 0.196 0.1995 0.2095 \n", @@ -104,11 +172,11 @@ "\n", " f_18 f_19 \n", "Missing % 0.202 0.21 " - ], - "text/html": "
\n\n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n
f_0f_1f_2f_3f_4f_5f_6f_7f_8f_9f_10f_11f_12f_13f_14f_15f_16f_17f_18f_19
Missing %0.2260.19850.1860.2130.2010.20450.1960.19950.20950.1950.20150.2010.18350.1760.1990.1930.19650.2120.2020.21
\n
" + ] }, + "execution_count": 4, "metadata": {}, - "execution_count": 4 + "output_type": "execute_result" } ], "source": [ @@ -147,15 +215,1404 @@ "metadata": {}, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-03-15T09:32:06.660519\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "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\n" + "image/png": "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\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-03-15T09:32:06.660519\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, "metadata": { "needs_background": "light" - } + }, + "output_type": "display_data" } ], "source": [ @@ -175,7 +1632,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "However if the model does not support missing values by default (f.e. `LogisticRegression`), results for only the imputation strategies are calculated. \n" + "However if the model does not support missing values by default (e.g. `LogisticRegression`), results for only the imputation strategies are calculated. \n" ] }, { @@ -184,15 +1641,1322 @@ "metadata": {}, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-03-15T09:32:25.260269\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "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\n" + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAA3pklEQVR4nO3deXxU5dn/8c+XHQGDgKW4AIK44IZKEUE0UkQURBGsYqX61CpaN1q1UJ/2J9alWpeKVq0+VbG1Vm0U61JFUaMIWgUVENx3FKgIBFAkINfvj3MSh0AgQJKZJN/365UXM+ecmXOdO8NcuZdz34oIzMzMck29bAdgZma2Pk5QZmaWk5ygzMwsJzlBmZlZTnKCMjOznOQEZWZmOckJyuocST+W9GS249gYSSFp58187eOSTq7smOoqScslddrIMX0kvV1dMdUFTlC2WSR9JKlfFs57iqQXNuH4jukXfYOSbRHx94joX8lx/Tj9ElsuaYWkNRnPl1fmuSoiIo6IiLuq+7wbImkXSf+UtFBSkaSZkn4pqX62Y9uYiGgeER9s5JjJEbFrdcVUFzhBmVWCNOk1j4jmwBHA5yXP0211mqTOwH+AT4G9IiIPOA7oDrTIZmwbkvmHjVU/JyjbYmmtZoqkP0paIukDSb3S7Z9K+m9mc5Ok8ZL+LOkpScskPSepQ7pvnRqPpEJJP5O0O/Bn4MC0ZrIk3T9Q0muSlqbnG5sR3vPpv0vS1xxYthaWxvpK+lf9K5J6lTn3pen1LZP0pKQ2m1g+20l6QNIXkj6UdG7GvvqSLpL0fvr+0yXtmPHyfpLeTcv1JknKKPMXJF0jaXH6vkeULbOMc1yT1lw+kHRWZhmXrQ1LGivp7oznPSVNTWOYISk/Y98p6XsuS2P4cTnFcAkwNSJ+GRHzACLi7Yg4MSKWpO81WNLs9DyF6e+75DwfSbowrXV9Jel2SW2VNGUukzRJ0jbpsSWfodMlfS5pnqQLMt6rh6QX0/PMk/QnSY0y9kdaRu8C72Zs2zl9fKSkOel5Pyt5b0n5kuZmvM/u6XUsSa9rcMa+8env87H0ff6jJIlbpojwj382+Qf4COiXPj4FWA38D1AfuAz4BLgJaAz0B5YBzdPjx6fPD073jwNeSPd1BAJokHGuQuBnGed6oUws+cBeJH9w7Q0sAI7ZwPudknG+VsBiYATQABiePm+dce73gV2ApunzKzdSNvnA3PRxPWA68P+ARkAn4APg8HT/hcAsYFdAwD4Z5w7gUaAl0B74AhiQcQ2rgNPSMj8T+BzQesrsDOAtYMf0ep/NLJPM32X6fCxwd/p4e+BL4Mj0Wg5Ln28LNAOWArumx7YD9iinTOYD/7OBMtsF+Cp9/4bAr4D3gEYZMb4EtE1j+i/wKrAv0AR4Bri4zO/8H2mMe6VlV/J53R/omf6+OwJvAqMyYgngqbSsmmZs2zl9PA/okz7eBthvPb/3hmn8F6W/974kn/mSshqflmOPNI6/A/dm+/91rv24BmWV5cOIuDMivgXuI/ky/F1ErIyIJ4FiILPD/7GIeD4iVgL/S1Ir2nHdt924iCiMiFkRsSYiZpJ8MR1SwZcPBN6NiL9FxOqI+AfJl/lRGcfcGRHvRMQK4H6g2yaE9wNg24j4XUQUR9KP8X/ACen+nwG/iaQ2ERExIyK+zHj9lRGxJCI+IUksmef+OCL+Ly3zu0gSRNv1xPAj4PqI+DQiFgG/34T4TwL+HRH/Tsv3KWAaScICWAPsKalpRMyLiNnlvE9rki/28hxP8pl4KiJWAdeQ/EHQK+OYGyNiQUR8BkwG/hMRr0XEN8AEkmSV6ZKI+CoiZgF3kvzxQURMj4iX0t/3R8CtrPt5+X1ELEp/52WtArpK2joiFkfEq+s5pifQnOT3VxwRz5D8sTE845gJEfFyRKwmSVDdNlA+dZITlFWWBRmPVwBERNltmX0xn5Y8iIjlwCJgu805saQDJD2bNqEVkdQYKtoMtx3wcZltH5P8lV5ifsbjr1n7OjamA7Bd2syzJG2WvIjvEsmOJDW08mzo3KX7IuLr9OH6YtuOjPJm3evdkA7AcWXiPwhoFxFfkSSWM4B5aXPVbuW8z5ckCbQ8a/0eImJNGnPm76Hs52lDny9Y95q3g9LBGo9Kmi9pKXAF635ePqV8Q0kS9MdKmqcPLOd6Pk2vIzOGyvpc1QlOUJYtpbUlSc1JmlM+J2nmAdgq49jvZzxe3/T79wAPAztG0vn+Z5LmsvKOz/Q5yZdwpvbAZxt5XUV9SlK7bJnx0yIijszYX9V9D/PIKG+S68v0FeWX96fA38rE3ywirgSIiIkRcRhJ8nmLpHa4PpNIvtjLs9bvIe1r25Et+z2UvebP08e3kMTaJSK2JvmDQWVeW+7nJiJeiYijge8BD5HUqsv6HNhRUuZ3bGV+ruoEJyjLliMlHZR2Tl8KvJQ2QX1B8p/4pLRz/6es/QW+ANghs1ObZBTYooj4RlIP4MSMfV+QNEOVdw/Lv4FdJJ0oqYGk44GuJM0xleFlYJmk0ZKapte0p6QfpPv/AlwqqYsSe0tqXUnnLnE/cK6kHdKBBGPK7H8dOEFSQ0ndgWEZ++4GjpJ0eBp7k3QwwA7pIIWjJTUDVgLLScp6fS4Gekm6WtL3ASTtLOluSS3TGAdK+qGkhsD56XtO3YLr/q2krSTtQdI/el+6vQVJ39nytMZ3ZkXfUFIjJbcU5KVNkUtZ/zX/h6RW9Ku0XPNJmo3v3eyrqYOcoCxb7iH50lpE0ml9Usa+00gGD3wJ7MHaX1LPALOB+ZIWptt+DvxO0jKSwQilf9GmTV+XA1PSJqqemUGk/T2DSL4QvyTpnB8UEQupBGn/0CCS/oUPgYUkSSkvPeS6NN4nSb7sbifpe6lM/wdMBGaQDCx4sMz+35L8EbCYZLTdPRnxfwocTVLL+IKkRnUhyXdHPeCXJLWFRST9OOv9so+I94EDSQYlzE6bYh8g6c9aFhFvk3wGbiQpo6OAoyKieAuu+zmSgQpPA9ekfaEAF5D8EbOMpGzuW//LyzUC+ChtHjwDWGfkYhr3USS3HCwEbgZ+EhFvbcZ11FklI37Mqo2k8SSjnX6T7VjqIkkdSZJlw7SDvlap7ddXl7gGZWZmOckJyszMcpKb+MzMLCe5BmVmZjnJEyFWgjZt2kTHjh2zHYaZWY00ffr0hRGxbdntTlCVoGPHjkybNi3bYZiZ1UiS1ju7iZv4zMwsJzlBmZlZTnKCMjOznOQ+KLM6ZtWqVcydO5dvvvkm26FYHdOkSRN22GEHGjZsWKHjnaDM6pi5c+fSokULOnbsSDJpuFnViwi+/PJL5s6dy0477VSh17iJz6yO+eabb2jdurWTk1UrSbRu3XqTau5OUGZ1kJOTZcOmfu6coMzMLCe5D8qsjus45rFKfb+PrhxYoePmz5/PqFGjeOWVV2jZsiVt27bliSee4M0332TXXXctPW7UqFG0a9eO0aNHM2vWLEaMGAHAJ598Ql5eHnl5ebRp04ZJkyZt9JwPP/wwc+bMYcyYsms2VqGxeRs/ZpPer6hCh2WjfAEeeughdtllF7p27brp11aGE5SZVbuIYMiQIZx88snce2+yyOyMGTNYsWIF9957LxdffDEAa9asoaCggClTpgCw11578frrrwNwyimnMGjQIIYNG7bWe69evZoGDdb/1TZ48GAGDx5cRVeVO6qyfDfmoYceYtCgQZWSoNzEZ2bV7tlnn6Vhw4acccYZpdv22WcfbrjhBu6777sFbp9//nk6dOhAhw4dNvh++fn5jBo1iu7duzNu3DgeeeQRDjjgAPbdd1/69evHggULABg/fjxnn302kHwBn3vuufTq1YtOnTpRUFBQBVeaHZVdvk8++SQHHngg++23H8cddxzLly8HYMyYMXTt2pW9996bCy64gKlTp/Lwww9z4YUX0q1bN95///0tug7XoMys2r3xxhvsv//+62zfa6+9qFevHjNmzGCfffbh3nvvZfjw4RV6z+Li4tI5MRcvXsxLL72EJP7yl7/whz/8gWuvvXad18ybN48XXniBt956i8GDB29ybSFXVWb5Lly4kMsuu4xJkybRrFkzrrrqKq677jrOOussJkyYwFtvvYUklixZQsuWLRk8ePBm1bzWxwnKzHLK8OHDuffee9ljjz146KGHuOSSSyr0uuOPP7708dy5czn++OOZN28excXF5d53c8wxx1CvXj26du1aWsuq7Ta1fF966SXmzJlD7969geQPgQMPPJC8vDyaNGnCqaeeyqBBgxg0aFClx+omPjOrdnvssQfTp09f774TTjiB+++/n0mTJrH33nvTtm1bJkyYQLdu3ejWrVu5Kwc0a9as9PE555zD2WefzaxZs7j11lvLvfemcePGpY9r0+KtlVm+EcFhhx3G66+/zuuvv86cOXO4/fbbadCgAS+//DLDhg3j0UcfZcCAAZV+HU5QZlbt+vbty8qVK7nttttKt82cOZPJkyfTuXNn2rRpw5gxY0qbn4YMGVL6Bdm9e/eNvn9RURHbb789AHfddVfVXEQOq8zy7dmzJ1OmTOG9994D4KuvvuKdd95h+fLlFBUVceSRR/LHP/6RGTNmANCiRQuWLVtWKdfhJj6zOq6iw8IrkyQmTJjAqFGjuOqqq2jSpAkdO3bk+uuvB5JmqDFjxnDsscdu1vuPHTuW4447jm222Ya+ffvy4YcfVmL0mxpMxYaFV6bKLN9tt92W8ePHM3z4cFauXAnAZZddRosWLTj66KP55ptviAiuu+46IKmhnXbaadxwww0UFBTQuXPnzb+O2lStzZbu3buHFyy0muLNN99k9913z3YYVket7/MnaXpErFM1dhOfmZnlJCcoMzPLSU5QZnWQm/YtGzb1c+cEZVbHNGnShC+//NJJyqpVyXpQTZo0qfBrPIrPrI7ZYYcdmDt3Ll988UW2Q7E6pmRF3YpygjKrYxo2bFjhFU3NsslNfGZmlpOcoMzMLCc5QZmZWU5ygjIzs5zkBGVmZjnJCcrMzHKSE5SZmeUk3wdVGT5/DcbmZTsKM6vJsrAsR65zDcrMzHKSE5SZWQ0wevRo+vTpw4gRI1i1alXp9gkTJpCfn09+fj7t27dn3LhxrFixonRbjx492HfffYFkIce99tqL/Px8zj///GxdSoW5ic/MLMfNmDGDzz77jMmTJ3P55ZdTUFCw1nLtQ4YMAeDQQw/lmGOOoWnTphQWFgIwfvx4Pv7449L3+v3vf8+gQYOq/Ro2h2tQZmY5burUqfTv3x+AAQMGMGXKlHWOmT9/PitXrqRDhw5rbf/nP//Jj370o9Lnv/3tbznkkEN45plnqjboSuAEZWaW4xYvXszWW28NQF5eHosWLVrnmAcffJChQ4eutW3JkiXMnz+/dIn1c889l9dee43777+f8847j+Li4qoPfgs4QZmZ5biWLVuydOlSAIqKimjVqtU6xxQUFDBs2LC1tv3rX//i6KOPLn1e8rq2bduy++67M3fu3CqMess5QZmZ5bhevXoxadIkACZOnEjv3r3X2r9gwYIKNe+VJLmvv/6at956i3bt2lVx5FvGCcrMLMd169aNtm3b0qdPH2bPns3QoUMZOXJk6f71Ne8VFRUxf/58dtttt9JtF154Ib169eLQQw/loosuomnTptV2DZtDXvZ5y3Xfrn5MO715tsMwsxomf/xX3z3peNBa+0pG4dUFkqZHRPey212DMjOznOT7oMzMsqTwlGbfPRlbmLU4cpVrUGZmlpOcoMzMLCc5QZmZWU5ygjIzs5zkBGVmZjnJCcrMzHKSExQgSem/TbIdi5mZJep8gpKkiAhJhwOjJDXOdkxmZuYbdUmT0wHAQKAgIlZW5HWSTgdOB6i/9bZ0/ObOKozSzGq9MY+td/NHVw6s5kByR52uQUmqL6kR8HegPzClZPvGXhsRt0VE94joXn+rvCqO1Mys7qmTCaqkzwmoHxHFwIFAE+AqgIj4VlKdLBszq1lGjx5Nnz59GDFiBKtWrSrdPmHCBPLz88nPz6d9+/aMGzcOgGuvvZbevXtz+OGHM2/ePABuuOEGOnbsuM56UtlW576EM/qc+gF/kvRzoC3QHThe0uUAEbEmm3GamW3MjBkz+Oyzz5g8eTK77bYbBQUFpfuGDBlCYWEhhYWFdO7cmWOOOYb58+fz2GOP8cILL3DppZdy6aWXAnDCCSfw9NNPZ+syylXnElSanH4IjAMKgOOBsyNiIUmSOlPS1dmM0cysIqZOnUr//v0BGDBgAFOmTFnnmPnz55cuZvjxxx+zxx57IIn99tuPyZMnA/C9732P+vU32rNR7erMIImSmlP6tBvwU2AN0By4DCAiFkjqCnTNSpBmZptg8eLFpavi5uXlsWjRonWOyVzMsHPnzkybNo2VK1fy7LPPrvf4XFLrE5SkFkDHiJiVjtb7FJgP3AasAgZGxHxJRwE7RsTNwPwyCc3MLOe0bNmydBn3oqIiWrVqtc4xBQUF3HlnMsq4TZs2nHnmmfTv359u3bqttdpuLqoLTXzNgDsk/Rm4HmgDvAy8CzyUJqcewJXAByUvcnIys1zXq1cvJk2aBMDEiRPp3bv3WvsXLFhQ2rxX4ic/+QnPPfccQ4YMIT8/vzrD3WS1OkGltaD5wJ+AU4AXI2Im8BHwT2AHSf8hSVy/jognshSqmdkm69atG23btqVPnz7Mnj2boUOHMnLkyNL9mc17JU444QT69u3LXXfdxfnnnw/Avffey0knncTkyZPp168fa9bkxhgx1faKgqR2wD4kfU23AhdHxJ/SfVsB9YFmaU1qs5r1GrfrEu1Ovr4SozYzS9SFG3UlTY+I7mW319oalKR6krYF/gLsEBEFJDfjXiLpJEm7AzcAq9Jalpv1zMxySK0dJJHex/SFpL8A/yNpTUTcIWkA8FdgGXBFRHyT1UDNzGy9amWCSmtHg4GrI2KCpFXA2ZKaRMTNknqRNOvN9Wg9M7PcVOua+NJpjHYGdgd+Kal+RDwK3AtcJeknQFFEzAU365mZ5apakaAy1nPaDmgTEY8A/wB2BM5PD5sCvAjM8jRGZma5r1YkqHT6osHA08C/JN0JfAJMAvaW9CTwMHBVRLyWxVDNzKyCakUflKRdSKYuOgGYBfwfMBL4NUnN6Ujgg4iYmrUgzcxsk9ToGpQSbYErgHbAirT57jSSiV8viIhFEXG3k5OZWc1SIxNUSZ9TJBYAN5HMr3eIpB3TJFUy156ZmdVANbKJL+1zGgQcDbwK3A9cA/wcOEjSSySDI87NXpRmZrYlauRUR5LakyzT/giwDbATcA7QEbiUZMbyByPi8eq4z6l79+4xbdq0qjyFmVmtVeOnOsoYSr4rcAhwV0T8AbgRmJP+OxcYCzQC2kna1vc5mZnVTDUiQWUs034YyXDxc4GfSGobEZ+T9De9D9wcES8BE4EeQHHWgjYzsy2S0wlKUn0o7XPaF/gFMBA4AFhIMlPEthlLaoxOj7+HZARfUXYiNzOzLZWzCUpSG+A1STummw4G+pCsjrsGOAPoDPy/NEnNi4h3JNUDiIjlWQnczMwqRc4mqIhYCDwDPC/p+xExjmSk3i8k9YyI/wJnkQyMaJ3xOk9jZGZWC+RkgsqoBY0imeT1FUntIuISkqQ1WlLv9B6oYyPirexFa2ZmVSEn74OKiDWS6kXEmoj4dTqA7z+SekTEtZIaA/8r6USSdZ3MzKyWyckEBeUmqamSDoqIKyTtFBFLshymmZlVkZxJUBlDyetHxLdQmqTqR8S3aZJqDEyX1CEiPsxyyGZmVoVyIkFlJKf+QG9JXwIPRMRnEfFtRk3ql5JujQjf32RmVsvlxCCJNDkdClwPPAGMAX4uqUG6f03JwAngnexEaWZm1SknElTqcJLJXr8FPgNuiYjVGTfrrkn/9dRFZmZ1QNab+CR1BL4E3iSZwmgHYFhEzJU0AmgI3JG9CM3MLBuyWoOStA3we2BPkpVwdwT+EBEfS9oH+BXweRZDNDOzLKn2GlSZUXqLJb1KkqQOA/4CHCnpdKAZ8JuIeKK6YzQzs+yrtgSVzjy+IB2V1xloHBFzIuLqdL69gyPiVkmPAwLqRcSH1bGek5mZ5Z5qSVDpWk63SRoNvEcySq+epDUkCw1+CQwGno6ITzJf6+RkZlY3VduKuumQ8U7AyRHxv+ls5bcAi4D5wK9JBkc8XC0BVaLu29WPaac3z3YYZlaTjPVqQCXKW1G3SmtQkpoBKyNiNdCeZPj4SElNI+KXwHGSBgLfJ1lc8L9VGY+ZmdUcVT2KrxdQIGkwMJ7kHqcuwNGSbgKIiMci4naSdZ5equJ4zMxy3ujRo+nTpw8jRoxg1apVpdsnTJhAfn4++fn5tG/fnnHjxgEwd+5cBg8ezKGHHsrFF18MwCWXXELPnj3p2bMnd999d1auY0tVeROfpCeBfOCIiHg63bYN8BIwOSJ+lm6rl84YUeMGRbiJz8w2WTlNfDNmzODqq6/m7rvv5vLLL6dTp04MHz58neMOPfRQxo8fT4cOHRg+fDjXXHMN22+/fen+Dz74gE6dOlFcXMz+++/PzJkzSSfdzjnlNfFVWQ1K35XEBOBx4FpJW0EyvJxk2faDJe2RJiXPFGFmdd7UqVPp378/AAMGDGDKlCnrHDN//nxWrlxJhw4dWLVqFR999BHnn38+ffv2ZerUqQB06tQJgIYNG1K/fv3qu4BKVOl9UBk1oJ3TWtEtwC2SbgdeAfZIb8LdHtjVCcnM7DuLFy+mXbt2AOTl5bFo0aJ1jnnwwQcZOnQoAAsXLuT111/nvvvuo1GjRhx11FG88sorpcdef/31DBs2LGdrTxtS6Qkqnfj1KOBy4C1JTYFzIuJUSX+VNBOoD/zKycnMbG0tW7Zk6dKlABQVFdGqVat1jikoKODOO+8sPX7nnXemffv2QFJjWr16NQ0aNODJJ59k8uTJFBQUVN8FVKJKb+KT9APgdyQzQzwA9AGukNQpIn5CMt/e8Ih4rLLPbWZW0/Xq1YtJkyYBMHHiRHr37r3W/gULFpQ27wE0bdqU1q1bs2TJEr766itWrlxJgwYNmDVrFpdeeil//etfqVcvl+YFr7hKjVpSF+B14FRgL+B8YB+gOXCvpL0iojAiZlbmec3Maotu3brRtm1b+vTpw+zZsxk6dCgjR44s3Z/ZvFfiiiuu4KijjqJv375ccsklAIwaNYpFixYxaNAg8vPzKSqqefddVdooPkkHAfcBB0bEJ5IuAZanUxkNB84juUn37Uo5YQ7xKD4zW5/88V+Vv7PjQeXuKiwsrPxgcliV3qgraTfgKuBnGVMVzQTOSWeQGAycXxuTk5mZVY3KGiSRB2wNnEIypBzgP0Aj4Gjg0ohYd6ykmVktVnhKs/J3ji2stjhqqs3qgyq5x0lSh3Tww39IktNqSZcCRMTciPgH8OOI+Ldq4hhHMzPLms1KUOlQ8qOB+4E/SLoL+Aa4AdhR0tUZx5as/eQh5WZmVmGbW4PaFRgF9AMmkozU+wiYBtxKkqR2qZwQzcysLqpQH5SkRhFRnD5uDBQBTwEnAz8Gjo2IryTtFxEvSpoTETVvTKOZmeWMjSaodBTeAEkBLAQGAA8CewM7AydGxAeS+gHXSDo6Ij6uyqDNzKz222CCkpQXEUWS3gPuAdoC/SNilqRngYYkS2d8DfwcGO3kZGZmlaHcPqi0Ke9VSedFxByStZzeB7oDRMStwN9IFhrcATg3Ih6tyaP1JPWRtG224zAzsw3UoCJipaSTgIclfRYR+0vqDvxW0jYRcR3JoIiXI+LzjNfVyNF6knoB44DTgC+yHI6ZWZ23wSa+dMDDQOBJSa0i4rZ0JdxRkvYEdgPOBD7f0PvkOkk7ARcBf4+I6ZLqlwyP38BrTgdOB6i/9bZ0/ObOaojUzGqNMeXPl/3RlQOrMZDctdFh5hHxMsnM5FdJ+llEPAmMBlYCv4uIGVUcY3VoASwChkraZWPJCSAibouI7hHRvf5WeVUfoZlZHVOh+6Ai4hWgP8myGedExKyIODMinqiJfU4ZM2HsmS4P8jHwW2AS8HNJnTOPMzPLFaNHj6ZPnz6MGDGCVatWlW6fMGEC+fn55Ofn0759e8aNGwfATTfdRI8ePejRowcPPPAAAP/85z/Zdddd6d59nflZc0qFb9RNk9RRJEmqg6T66fYa1+eUzoRxBMlMGEOBOSTzCf4LWAz8StLONfHazKz2mjFjBp999hmTJ09mt912W2shwiFDhlBYWEhhYSGdO3fmmGOOAeDmm29m6tSpFBYWcsUVVwDQt29fZs2alY1L2CSbNJNEOufe9hHxcUWawXKVpB2AX5Dc0/UksAz4JCKmkyyyuIhkolszs5wxdepU+vfvD8CAAQOYMmXdObjnz5+/1oKGnTp1YsWKFSxbtoyWLVsC0Lp1axo1yv2vuM2ZzXwZJM1fNaWGkQ6Zrx8RX0vaBvgS+DdwInAsMDAilkgakm7/XUSsyF7EZmbrWrx4Me3atQMgLy+PRYsWrXNM2QUNBw4cyO677863337L7bffXm2xVoZNTlAlSakGJaf6QA+gq6RlQG/geqAnyRyC/SLis7Qv6nLgw4h4PUvhmpmVq2XLlixduhSAoqIiWrVqtc4xBQUF3HlnMqp46dKl3HLLLbz77rsUFxfTt29fjjjiCGpK93rNXKh+E6RNkV+Q1JSuBZ6KiHeBK4CvgbMlXQPcDoxxcjKzXNWrVy8mTZoEwMSJE+ndu/da+xcsWLBW8169evVo2rQpTZo0oVmzZhQXF1ND6hZALU9QGaPw3gbeI1lEcU9J20fETOBHwGzgTeDMiHjYI/fMLFd169aNtm3b0qdPH2bPns3QoUMZOXJk6f6yzXvNmzfn2GOP5cADD6RXr16cddZZ1KtXj8LCQvr168c777xDv379+Pzz3LyVVTUpm24OSfsCv42IYyV1JFkm5JuIGCOpNdAxHRyx2Rq36xLtTr5+i2M1s7pt/j1jAOjZqTWFhYXZDaYaSZoeEeuMea/VNajU60AjSQ9FxEckTXlNJP2LpEZVWcvem5lZJaq1NShJ2wFF6TpV9YB/AFtFxFGSWgFHkwwtf3pLz+UalJlVpro21VGdqUFJqi+pDXAvcKKkZhGxhmRIeZ6kRyJiUUTcGRFPu8/JzCw31ZoEVZJoIuLbiFgI/AY4AThOUvN0NN+DJMvRl2bqmjJc3sysrqk1/S8Z0xcNIxmx9yhwLnAjSc1pGTAQOD4i3s5epGZmVhG1qQbVFfg18AbJ4op/B1oBZwHfI+lzutnJycysZqgVNai0yW4CcGlE3JZumwVcDBxP0tzXICJW1aQpmszM6rJaUYOKiGnAAuDnGZufTbc1j8Sq9FgnJzOzGqBGJqiM9Zz2kfRDgHSIYrGkxyV9D9gDOIhkMUIzM6thamQTXzog4jDgBmCZpJeBayKih6QpJNMXjQdGRMQbWQzVzMw2U426Ubek/0hSA+BC4DGSZHQdycCIGyPiQ0mPAY0i4rDM11VVXN27d49p06ZV1dubmdVqteJG3TQ5HUMyK8QQoEt6f9MVJNfyK0ldImIg0F7S30tel62Yzcxs89SoBJUOJR8F3EdSe7pEUu+IWABcCawBmgBExK7A/2YpVDMz20I5naAktZV0Yvq4PXAJ8E5EFETEJSQ34d4g6ZCImA/8IiJmSWoIkE4Oa2ZmNVDOJqh0pN5BwEBJJwGfkvQ3bSepj6T6EXEryWCIm9Ol3L8FKBlSbmZmNVfOjuJL+5ueARoC/YBvSGpQY0kWGlwj6aWIuFHSvyJicfaiNTOzypaTCSpj4tfFkh4nqekNSHdfQjIzxE8BAS9ExCdZCdTMzKpMTiWodJ2mphHxmVIRUZQOG4ckSa0BLiNJVK41mZnVUjmToCQ1Bc4Bmkq6KSI+XU+SWgMMBepHxG+zGrCZmVWpnBkkERErgJK7XU+R1L7k/qWSJAU8ATwOvJWlMM3MrJpkPUFJaidpf4CIeAx4ANiGJEl1LJOklgD/iIhZWQvYzMyqRdYSlKR6krYG5gCvSBor6ZfA+yQ1pVXATyRtnzkTRDpzhJmZ1XLZrEFFRCwFzgOKSRYVBHgYGAR0A3YBzpT0PU9XZGZWt2QlQUlqC7wsKS8i/gqcBgwHHgGOIll8sAlJkjof2DobcZqZWfZkZRRfRCyQ9C7wkqQDIuJvkpoDLwNHRcSzkl4BVgO7RMR72YjTzMyyp1prUJK2KnkcEScCk4HXJG0dEbcAo4GHJOVHxPKI+CYiZqavVXXGamZm2VVtCUrSDsDbkm6UdA5ARJwO3EOSpFpExG3AxcBjklpmJiX3QZmZ1S3V0sQnqQnQCviAZCn2fpL2SLf9BugNPC5pQETcJOnhdEi5mZnVUVW+om46IKKAZGTe/sChwJfAq0AXoCfQgmQC2HeBrmlcq6t6JdzK0n27+jHt9ObZDsPMct3YomxHkJPKW1G3ympQGcmlIVCczgTxTNoPdWh67puBO4BtgReBWZn3OdWE5GRmZlWjKvug8tJ/A2hasjEiHgUmATsCI4GdIuK/wA0R8YwHQ5hZXTR69Gj69OnDiBEjWLXquyXtJkyYQH5+Pvn5+bRv355x48YB0KVLl9LtTz31FACXXXYZBx98MD/4wQ+48cYbs3IdlalKmvgkNSaZIeJGklrSbGDfiFiecUxfkma9D4AbgJU1tcbkJj4zq5BymvhmzJjB1Vdfzd13383ll19Op06dGD58+DrHHXrooYwfP54OHTrQvXt3pk2bttb+4uJiGjVqxOrVq9l7772ZNWsW9evXr5JLqUzlNfFVSQ0qIlYCJwEXAWOAqcAukvaS1EVSC5IE9jzwr3Q4eY1MTmZmW2rq1Kn0798fgAEDBjBlypR1jpk/fz4rV66kQ4cOACxfvpxDDjmEE088kUWLFgHQqFEjAFauXEnnzp1rRHLakCpr4ouIF4EjSJbQGEHSnHdP+vME8BzwYkS8XVUxmJnVBIsXL2brrZMJc/Ly8koTTqYHH3yQoUOHlj6fMmUKzz33HAMGDODiiy8u3T5q1Ci6dOlCz549qz7wKlal90FFxHTgYOALYFpE7BURPyAZJHF4RHxYlec3M6sJWrZsydKlSwEoKiqiVatW6xxTUFDAsGHDSp+3bt0agGHDhjFjxozS7ddffz0ffPABEyZMYN68eVUcedWq8ht1I+JNkiHmV0i6IN1WDHiZdjMzoFevXkyaNAmAiRMn0rt377X2L1iwYK3mveLiYlauXAnA5MmT2XnnnQFKtzVu3JitttqKJk2aVNclVIlquVE3Il6RNAiYJOl+4NOIWFMd5zYzy3XdunWjbdu29OnTh/bt23PBBRcwcuRIbr31VmDd5r3Fixdz5JFH0qxZMxo3bswdd9wBwHnnncdbb71FcXExJ510Ettss01WrqeyVPmNumudLJlzb2m1nbCaeBSfmZXIH/9V+Ts7HlTursLCwsoPpoao1lF8G7AsDcb3OpmZ2QZV63IbJUPJPaTczGqrwlOalb9zbGG1xVEbZHNFXTMzs3I5QZmZWU5ygjIzs5zkBGVmZjnJCcrMzHKSE1QGD383M8sdTlCslZhq9m3XZma1iBMUyX1Zkg4HHpSU55qUmVn2VeuNurlKUlfgfOCCiCiqSIKSdDpwOkD9rbel4zd3VnGUZlbjjXlso4d8dOXAagikZqjzNShJTUnWrdob6AgVm+kiIm6LiO4R0b3+VnkbO9zMzDZRnUxQJTUkSVsBK4FbSJadP0LSwdmMzcysIkaPHk2fPn0YMWIEq1atKt0+YcIE8vPzyc/Pp3379owbNw6Am266iR49etCjRw8eeOCBbIW9Sepkgkr7nI4G/gY8DPQHCoA3gBMkHZrN+MzMNmTGjBl89tlnTJ48md12242CgoLSfUOGDKGwsJDCwkI6d+7MMcccA8DNN9/M1KlTKSws5IorrshS5JumTiYoSb2A0cAZwELgFxHxDvAg8ClwkiSP6DOznDR16lT69+8PwIABA5gyZco6x8yfP3+tRQ47derEihUrWLZsGS1btqzOcDdbnRgkIal+RHybsakVSZPewcAuwI/T7YuBPwMtI2Jx9UZpZlYxixcvpl27dgDk5eWxaNGidY4pu8jhwIED2X333fn222+5/fbbqy3WLVHrE5SkxkAvSa+SDIJoBzQG/gdoCpwYER9JGgYMB06KiA+zFa+Z2ca0bNmSpUuTtV+Liopo1arVOscUFBRw553J6OKlS5dyyy238O6771JcXEzfvn054ogjyPU7aupCE19zkqR0P/AQMDv992tgDtBMUn9gLHBnRKzISpRmZhXUq1cvJk2aBMDEiRPp3bv3WvsXLFiwVvNevXr1aNq0KU2aNKFZs2YUFxdTE5blq/UJKiK+BL4Afgi8CqxOm/tOTQ85FzgH+FVEPOqbdM0s13Xr1o22bdvSp08fZs+ezdChQxk5cmTp/rLNe82bN+fYY4/lwAMPpFevXpx11lnUq5f7X/+qCVl0S0g6APgAaAscBnQB7oiIaZK2Ixkk0Tgilm3uORq36xLtTr6+MsI1szpq/j1jAOjZqTUAhYWFWYymekmaHhHdy27P/RS65Y4GXiRJRH8nqU2dIunnwDVA6y1JTmZmVjVq7SAJSQ0iYnVEXCSpGJgIHA78CTgROBm4IiLmZTNOMzOA7594JQCFnuqoVK2sQUnaH7hIUhuAiBhLckPuQ0CDiBgHHBYR/3Kfk5lZbqqVCQooAvoCZ0pqnW67iWR4+ROSmgBfQcXm3TMzs+pXq5r40ppTQ5LZII4hmcpIkm4iuQfqceCBiPgmWzGamVnF1PgEJUnp3Hp9SAZBPE9y31MBSV/TbcB1JPPt/TQipmctWDMzq7Aan6DS5NQXGAIMi4iXJe0CjAf+C/wU2Ba4KiLmZC9SMzPbFDW6DypjgMMQ4CzShJtO/DoOOCAiVkTEJ05OZmY1S42sQZU065FM+vplRJwjqR7wd0m7RkQxybXtmc7FV+zBEGZmNUuNrEGlzXpHAhMk3Smpa0ScBUwC/ivpN8BxwF8iYqWTk5lZzVOjalAZAyLaAucB/49kpohRkv4WEadJWkqy1lN+REyX1DAiVm3ofbfUXtvnMc0315mZVaoalaDS5NQbaAnMjohCSc8BvwFGpLNHnC+pIfCwpC4R8XU2YzYzs81TI5r4JNVP/z0I+CcwCDhD0mmRuJRkxN5PJLWIiHNJhpm3y1rQZma2RXK6BiWpFbA8IooldQNOAs6IiIclFQDXpK1+f4mI30jauWTi14g4L4uhm5nZFsrZGpSkrYALgYslNQD2AfYDuqW1pKeB84ExkkYCRMR7WQvYzMwqVc4mKKAYeAnYCjg3Iu4C/gh0Bg6W1CwingHOIFkZ18zMapGcbeKLiNWSHiNJVIMk/TIirpPUFBgKNJL0ZERMym6kZmZWFXI2QUFpknoqfZqZpBqRJKkppLOSm5lZ7ZLTCUpSvTJJ6ghJYyLiSkmPRsR/sxqgmZlVmZxJUBk34ZbeWBsRayTVz0hSDYCBknaKiA+zG7GZmVWlnElQaXIaCIyUVAhMi4jnI+LbjCT1OPBiRCzMbrRmZlbVcmYUn6RdgZHA0yQj906VdDhAZpJycjIzqxtyIkFJ2ptkocF/RcQ4koUHpwDD01oVEfFtFkM0M7NqlhMJKiJmAjOBX6XPPwSeAKYBP5bUJovhmZlZFmQlQZUsNCipo6S9ACLiMOA9SVPT558AjwCj3axnZlb3ZCVBpQMiBgMPApdJGi+pTUQMBOZJmpEe93FEfJqNGM3MLLuqLUFJapLxuBfwW+Bw4CGSJdt/L2nbiBgKfJoeY2ZmdVS1JChJ2wD/kNQi3bQQOAvYn2QuvQOBXYB7JO0QEYMiYmp1xGZmZrmpyhOUpEYRsRj4ObCdpAMj4h2SARA/BO6IiDkkI/e2BppWdUxmZpb7qjRBpaPv7pD0g4iYBxwGTEifrwHeBI6TdA4wAvhFRLxblTGZmVnNUKUzSUTEQkmfAKMkXR0Rf5L0LfB3ScOBCUAjYCBwlZv1zMysRJXVoNJFBiFZon1nkqS0b0TcAlwP3AXsEhF/BoZFxKMlw8/NzMyqLEGlc+f1A+4ALiFpzrsqTVI3A7eTDJxoCZRMDhtVFY+ZmdUsVZKgMmpChwKPRcS/I2IYMB34q6TuEfFHID8ilqT9UWZmZqWqJEFl1ITmAM1LpiqKiF+n5/y1pK0B34RrZmbrVWmDJDLWc+oDbA/8l6TGNBw4UtIr6fneAq6JiKWVdW4zM6t9KqUGlS6FEZL6A7cCAUwCdiIZEHEQcBnJvU53RsSLlXFeMzOrvbaoBiWpVUQsStdr2gb4KfAjkhtuZwEzIuJzSZNJ1nhqFRHvb3HUZmZW6212DUpSR2C6pN8DpLNFvAqcAlwLHJsmp1OB/SNisZOTmZlV1JY08a1OX58v6bp02yqSPqdTI+L9dCHCXwINtyxMMzOraza7iS8i5kq6EagPbCvpqogYLWk3YKykYqAr8OuIeK6S4jUzszpikxKUpE5Aj4i4N900g2Tww6+BoZIujYiRkvYBWgOLIuL1khF+lRq5mZnVaqpo3pDUCHgHaA9cCbwMFAKDgJbAZOA84KuIOKcKYs1ZkpYBb2c7jhzVhmR5FVuXy2b9XC7lq61l0yEiti27scI1qIgolnQ0yQKDvYFXgEdJalFbpRPBXk8yMWzXdAmNuuLtiOie7SBykaRpLpv1c9msn8ulfHWtbDZpkEREzACOBvYEWpAskQHwfUldgDeA8+pYcjIzsyqwyYMkImKmpAEkN+KeGRFnpRO+Lk/n1FtWyTGamVkdtFmj+CLilXSm8scktYmIGyo5rprmtmwHkMNcNuVz2ayfy6V8dapsKjxIYr0vlg4gqUntAcz1rORmZlZZtihBAUja2hO/mplZZauMyWKXwVprQJmZmW2xLU5QJTfg1oUbcSUNkPS2pPckjVnP/saS7kv3/yedr7DWq0C5/FLSHEkzJT0tqUM24syGjZVNxnFDJYWkOjOEuCJlI+lH6WdntqR7qjvGbKnA/6n2kp6V9Fr6/+rIbMRZ5SLCPxX4IZnS6X2gE9CI5P6vrmWO+Tnw5/TxCcB92Y47R8rlUJJ75QDOrAvlUtGySY9rATwPvAR0z3bcuVI2QBfgNWCb9Pn3sh13DpXNbSSjqCGZUu6jbMddFT9VsqJuLdUDeC8iPoiIYuBeknvCMh0N3JU+LgB+WAeaPjdaLhHxbER8nT59CdihmmPMlop8ZgAuBa4CvqnO4LKsImVzGnBTJCslEBH/reYYs6UiZRMkyxoB5AGfV2N81cYJquK2Z+0l6uem29Z7TESsBopI5iSszSpSLplOBR6v0ohyx0bLRtJ+wI4R8Vh1BpYDKvK52QXYRdIUSS+l91/WBRUpm7HASZLmAv8GauX0cpW25LvZxkg6CegOHJLtWHKBpHrAdSRrqNm6GpA08+WT1Lqfl7RXRCzJZlA5YjgwPiKulXQg8DdJe0Ytu9XHNaiK+wzYMeP5Dum29R4jqQFJ1fvLaokueypSLqQ3dv8vMDgiVlZTbNm2sbJpQTJtWKGkj4CewMN1ZKBERT43c4GHI2JVRHxIMll1l2qKL5sqUjanAvcDRMSLQBOSiWRrFSeoinsF6CJpp3Rm9xOAh8sc8zBwcvp4GPBMpL2YtdhGy0XSvsCtJMmprvQjwEbKJiKKIqJNRHSMiI4k/XODI2JadsKtVhX5//QQSe0JSW1Imvw+qMYYs6UiZfMJ8EMASbuTJKgvqjXKauAEVUFpn9LZwETgTeD+iJgt6XeSBqeH3Q60lvQeyUrC5Q4rri0qWC5XA82Bf0p6XVLZ/2y1UgXLpk6qYNlMBL6UNAd4FrgwImp7i0RFy+Z84DRJM4B/AKfUxj+Gt3gmCTMzs6rgGpSZmeUkJygzM8tJTlBmZpaTnKDMzCwnOUGZmVlOcoIyq0SSvk2H0r8h6RFJLTP27SHpmXSW6ncl/TZzrkZJR0ials7e/ZqkazdwnockvVRm23hJw8psW57xeBdJ/07P/aqk+yW1LXN8PUk3pPHPkvSKpJ22oEjMNpsTlFnlWhER3SJiT2ARcBaApKYkN1teGRG7AvsAvUhmwEfSnsCfgJMioivJlFDvre8EadLbH8iT1KkiQUlqAjwG3BIRXSJiP+BmYNsyhx4PbAfsHRF7AUOAJRW79HLP7SnVbLM4QZlVnRf5bpLPE4EpEfEkQDq7+9l8dzP3r4DLI+KtdP+3EXFLOe97LPAIySzXJ1QwlhOBFyPikZINEVEYEW+UOa4dMK9kTreImFsym3i6RtGrkmZIejrd1iqtzc1MJ3TdO90+VtLfJE0hmSduW0kPpDWyVyT1rmDcVoc5QZlVAUn1SaaiKZk1Yw9geuYxEfE+0FzS1iRz8q21fwOGk8we8I/0cUVU9P3vB45KmymvTaepQtK2wP8BQyNiH+C49PhLgNciYm/gIuCvGe/VFegXEcOBccAfI+IHwFDgLxWM2+owV73NKldTSa+T1JzeBJ6qzDdP+4y6AC9EREhalc5i/QbJGkFlbdJUMRExV9KuQN/052lJxwFbAc+nk7YSEYvSlxxEknCIiGcktU4TLiQTva5IH/cDumZ0uW0tqXlElPaRmZXlGpRZ5VoREd2ADoBI+6CAOST9RqXS/qPlEbEUmF12fzl+BGwDfJjOgN6R72pRX6b7St6/FbAwfVrR9yciVkbE4xFxIXAFcExFXrceX2U8rgf0TPvnukXE9k5OtjFOUGZVIO1jOhc4Px0k8HfgoHTZkZJBEzcAf0hfcjVwkaRd0v31JJ2xnrceDgzImAF9f77rhyoEjk9nwIZknaln08f3AL0kDSx5I0kHp4MzyNi2n6TtSmIA9gY+Jplp/eCSEX1p8gOYDPw43ZYPLEwTbllPkrGonqRu6znGbC1OUGZVJCJeA2YCw9OmrqOB30h6G5hFsqzCn9JjZwKjgH9IehN4A1hrhJ6kjiQ1s9Lh5WmTW5GkAyLiUZKEMT1tZuwNjE6PWwEMAs5Jh5nPIRlBWHaJhu8Bj0h6I419NfCniPgCOB14MJ1B+770+LHA/pJmAlfy3XIzZZ0LdE8HU8wB1pd8zdbi2czNzCwnuQZlZmY5yQnKzMxykhOUmZnlJCcoMzPLSU5QZmaWk5ygzMwsJzlBmZlZTvr/tB0XRBGIzhUAAAAASUVORK5CYII=\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-03-15T09:32:25.260269\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, "metadata": { "needs_background": "light" - } + }, + "output_type": "display_data" } ], "source": [ @@ -216,15 +2980,1322 @@ "metadata": {}, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-03-15T09:32:52.987873\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAA350lEQVR4nO3deXhV1dn+8e/NjIAgYCkOgCAOOKFSRBAbKSIKIghWsVJ9axXn0moLdvgJRX21DhWtWn2rUGut2ijWoYqiRhGkCiogOM8ooMhskYA8vz/2TjwEAgGSnJPk/lxXLs7Ze52zn71yOE/W2muvpYjAzMws19TKdgBmZmab4gRlZmY5yQnKzMxykhOUmZnlJCcoMzPLSU5QZmaWk5ygrMaR9CNJT2Y7ji2RFJL23MbXPi7p9PKOqaaStEpS+y2U6SnprcqKqSZwgrJtIulDSb2zcNwzJL2wFeXbpV/0dYq2RcTfI6JPOcf1o/RLbJWk1ZLWZzxfVZ7HKouIODYi/lrZx90cSXtJ+qekxZKWS5ot6ReSamc7ti2JiMYR8f4WykyJiL0rK6aawAnKrBykSa9xRDQGjgU+K3qebqvRJHUA/gN8AhwQEU2Bk4AuQJNsxrY5mX/YWOVzgrLtlrZqpkr6o6Rlkt6X1D3d/omkzzO7myRNkPRnSU9JWinpOUlt030btXgkFUj6qaR9gT8Dh6ctk2Xp/n6SXpW0Ij3e6Izwnk//XZa+5vCSrbA01pfTv+pfltS9xLHHpue3UtKTklpuZf3sIukBSV9I+kDSRRn7akv6taT30vefKWn3jJf3lvROWq83S1JGnb8g6VpJS9P3PbZknWUc49q05fK+pPMz67hka1jSaEl3ZzzvJmlaGsMsSXkZ+85I33NlGsOPSqmGMcC0iPhFRCwAiIi3IuLUiFiWvtcASXPT4xSkv++i43wo6Zdpq+srSXdIaqWkK3OlpMmSdkrLFn2Gzpb0maQFki7JeK+ukl5Mj7NA0p8k1cvYH2kdvQO8k7Ftz/TxcZLmpcf9tOi9JeVJmp/xPvum57EsPa8BGfsmpL/Px9L3+Y+SJG6ZIsI//tnqH+BDoHf6+AxgHfA/QG3gcuBj4GagPtAHWAk0TstPSJ8fme4fB7yQ7msHBFAn41gFwE8zjvVCiVjygANI/uA6EFgEDNzM+52RcbzmwFJgGFAHGJo+b5Fx7PeAvYCG6fOrtlA3ecD89HEtYCbw/4B6QHvgfeCYdP8vgTnA3oCAgzKOHcCjQDOgDfAF0DfjHNYCZ6V1fi7wGaBN1Nk5wJvA7un5PptZJ5m/y/T5aODu9PGuwJfAcem5HJ0+3xloBKwA9k7Ltgb2K6VOFgL/s5k62wv4Kn3/usCvgHeBehkxTgdapTF9DrwCHAw0AJ4BLivxO/9HGuMBad0VfV4PBbqlv+92wBvAiIxYAngqrauGGdv2TB8vAHqmj3cCDtnE771uGv+v0997L5LPfFFdTUjrsWsax9+Be7P9/zrXftyCsvLyQUSMj4hvgPtIvgx/HxFrIuJJoBDIvOD/WEQ8HxFrgN+QtIp23/httywiCiJiTkSsj4jZJF9M3y/jy/sB70TE3yJiXUT8g+TL/PiMMuMj4u2IWA3cD3TeivC+B+wcEb+PiMJIrmP8H3BKuv+nwG8jaU1ERMyKiC8zXn9VRCyLiI9JEkvmsT+KiP9L6/yvJAmi1SZi+CFwQ0R8EhFLgP/divhPA/4dEf9O6/cpYAZJwgJYD+wvqWFELIiIuaW8TwuSL/bSnEzymXgqItYC15L8QdA9o8xNEbEoIj4FpgD/iYhXI+JrYCJJsso0JiK+iog5wHiSPz6IiJkRMT39fX8I3MbGn5f/jYgl6e+8pLVAJ0k7RsTSiHhlE2W6AY1Jfn+FEfEMyR8bQzPKTIyIlyJiHUmC6ryZ+qmRnKCsvCzKeLwaICJKbsu8FvNJ0YOIWAUsAXbZlgNLOkzSs2kX2nKSFkNZu+F2AT4qse0jkr/SiyzMePxfNjyPLWkL7JJ28yxLuyV/zbeJZHeSFlppNnfs4n0R8d/04aZi24WM+mbj892ctsBJJeI/AmgdEV+RJJZzgAVpd9U+pbzPlyQJtDQb/B4iYn0ac+bvoeTnaXOfL9j4nHeB4sEaj0paKGkFcCUbf14+oXSDSRL0R0q6pw8v5Xw+Sc8jM4by+lzVCE5Qli3FrSVJjUm6Uz4j6eYB2CGj7HczHm9q+v17gIeB3SO5+P5nku6y0spn+ozkSzhTG+DTLbyurD4haV02y/hpEhHHZeyv6GsPC8iob5Lzy/QVpdf3J8DfSsTfKCKuAoiISRFxNEnyeZOkdbgpk0m+2Euzwe8hvda2O9v3eyh5zp+lj28libVjROxI8geDSry21M9NRLwcEScA3wEeImlVl/QZsLukzO/Y8vxc1QhOUJYtx0k6Ir04PRaYnnZBfUHyn/i09OL+T9jwC3wRsFvmRW2SUWBLIuJrSV2BUzP2fUHSDVXaPSz/BvaSdKqkOpJOBjqRdMeUh5eAlZJGSmqYntP+kr6X7v8LMFZSRyUOlNSinI5d5H7gIkm7pQMJRpXY/xpwiqS6kroAQzL23Q0cL+mYNPYG6WCA3dJBCidIagSsAVaR1PWmXAZ0l3SNpO8CSNpT0t2SmqUx9pP0A0l1gYvT95y2Hef9O0k7SNqP5Profen2JiTXzlalLb5zy/qGkuopuaWgadoVuYJNn/N/SFpFv0rrNY+k2/jebT6bGsgJyrLlHpIvrSUkF61Py9h3FsnggS+B/djwS+oZYC6wUNLidNt5wO8lrSQZjFD8F23a9XUFMDXtouqWGUR6vac/yRfilyQX5/tHxGLKQXp9qD/J9YUPgMUkSalpWuT6NN4nSb7s7iC59lKe/g+YBMwiGVjwYIn9vyP5I2ApyWi7ezLi/wQ4gaSV8QVJi+qXJN8dtYBfkLQWlpBcx9nkl31EvAccTjIoYW7aFfsAyfWslRHxFsln4CaSOjoeOD4iCrfjvJ8jGajwNHBtei0U4BKSP2JWktTNfZt+eamGAR+m3YPnABuNXEzjPp7kloPFwC3AjyPizW04jxqraMSPWaWRNIFktNNvsx1LTSSpHUmyrJteoK9Wqvv51SRuQZmZWU5ygjIzs5zkLj4zM8tJbkGZmVlO8kSI5aBly5bRrl27bIdhZlYlzZw5c3FE7FxyuxNUOWjXrh0zZszIdhhmZlWSpE3ObuIuPjMzy0lOUGZmlpOcoMzMLCf5GpRZDbN27Vrmz5/P119/ne1QrIZp0KABu+22G3Xr1i1TeScosxpm/vz5NGnShHbt2pFMGm5W8SKCL7/8kvnz57PHHnuU6TXu4jOrYb7++mtatGjh5GSVShItWrTYqpa7E5RZDeTkZNmwtZ87JygzM8tJvgZlVsO1G/VYub7fh1f1K1O5hQsXMmLECF5++WWaNWtGq1ateOKJJ3jjjTfYe++9i8uNGDGC1q1bM3LkSObMmcOwYcMA+Pjjj2natClNmzalZcuWTJ48eYvHfPjhh5k3bx6jRpVcs7ECjW665TJb9X7Ly1QsG/UL8NBDD7HXXnvRqVOnrT+3EpygzKzSRQSDBg3i9NNP5957k0VmZ82axerVq7n33nu57LLLAFi/fj35+flMnToVgAMOOIDXXnsNgDPOOIP+/fszZMiQDd573bp11Kmz6a+2AQMGMGDAgAo6q9xRkfW7JQ899BD9+/cvlwTlLj4zq3TPPvssdevW5ZxzzinedtBBB3HjjTdy333fLnD7/PPP07ZtW9q2bbvZ98vLy2PEiBF06dKFcePG8cgjj3DYYYdx8MEH07t3bxYtWgTAhAkTuOCCC4DkC/iiiy6ie/futG/fnvz8/Ao40+wo7/p98sknOfzwwznkkEM46aSTWLVqFQCjRo2iU6dOHHjggVxyySVMmzaNhx9+mF/+8pd07tyZ9957b7vOwy0oM6t0r7/+OoceeuhG2w844ABq1arFrFmzOOigg7j33nsZOnRomd6zsLCweE7MpUuXMn36dCTxl7/8hT/84Q9cd911G71mwYIFvPDCC7z55psMGDBgq1sLuao863fx4sVcfvnlTJ48mUaNGnH11Vdz/fXXc/755zNx4kTefPNNJLFs2TKaNWvGgAEDtqnltSlOUGaWU4YOHcq9997Lfvvtx0MPPcSYMWPK9LqTTz65+PH8+fM5+eSTWbBgAYWFhaXedzNw4EBq1apFp06diltZ1d3W1u/06dOZN28ePXr0AJI/BA4//HCaNm1KgwYNOPPMM+nfvz/9+/cv91jdxWdmlW6//fZj5syZm9x3yimncP/99zN58mQOPPBAWrVqxcSJE+ncuTOdO3cudeWARo0aFT++8MILueCCC5gzZw633XZbqffe1K9fv/hxdVq8tTzrNyI4+uijee2113jttdeYN28ed9xxB3Xq1OGll15iyJAhPProo/Tt27fcz8MJyswqXa9evVizZg2333578bbZs2czZcoUOnToQMuWLRk1alRx99OgQYOKvyC7dOmyxfdfvnw5u+66KwB//etfK+Ykclh51m+3bt2YOnUq7777LgBfffUVb7/9NqtWrWL58uUcd9xx/PGPf2TWrFkANGnShJUrV5bLebiLz6yGK+uw8PIkiYkTJzJixAiuvvpqGjRoQLt27bjhhhuApBtq1KhRnHjiidv0/qNHj+akk05ip512olevXnzwwQflGP3WBlO2YeHlqTzrd+edd2bChAkMHTqUNWvWAHD55ZfTpEkTTjjhBL7++msiguuvvx5IWmhnnXUWN954I/n5+XTo0GHbz6M6NWuzpUuXLuEFC62qeOONN9h3332zHYbVUJv6/EmaGREbNY3dxWdmZjnJCcrMzHKSE5RZDeSufcuGrf3cOUGZ1TANGjTgyy+/dJKySlW0HlSDBg3K/BqP4jOrYXbbbTfmz5/PF198ke1QrIYpWlG3rJygzGqYunXrlnlFU7NschefmZnlJCcoMzPLSU5QZmaWk5ygzMwsJzlBmZlZTnKCMjOznOQEZWZmOcn3QZWHz16F0U2zHYWZVUVZWI6jqnALyszMcpITlJlZDhs5ciQ9e/Zk2LBhrF27tnj7xIkTycvLIy8vjzZt2jBu3DhWr15dvK1r164cfPDBAIwZM4Zu3brRrVs37r777mydylZzgjIzy1GzZs3i008/ZcqUKeyzzz7k5+cX7xs0aBAFBQUUFBTQoUMHBg4cSMOGDYu3nXfeeQwcOBCAYcOGMX36dJ5//nmuvvrqKjNRsBOUmVmOmjZtGn369AGgb9++TJ06daMyCxcuZM2aNbRt23aD7f/85z/54Q9/CED79u2BZB7G2rVrV3DU5ccJyswsRy1dupQdd9wRgKZNm7JkyZKNyjz44IMMHjx4g23Lli1j4cKFGy2tfsMNNzBkyBAkVVzQ5cij+MzMclSzZs1YsWIFAMuXL6d58+YblcnPz2f8+PEbbPvXv/7FCSecsMG2J598kilTpmzQTZjr3IIyM8tR3bt3Z/LkyQBMmjSJHj16bLB/0aJFW+zeA5gzZw5jx47lrrvuolatqvO1X3UiNTOrYTp37kyrVq3o2bMnc+fOZfDgwQwfPrx4/6a695YvX87ChQvZZ599ireNGDGCJUuW0L9/f/Ly8li+vGrce6WqMpojl3XZpXbMOLtxtsMwsxyXN+GrjTe2O2KTZQsKCio2mBwiaWZEdCm53S0oMzPLSR4kYWZWSQrOaLTxxtEFlR5HVeEWlJmZ5SQnKDMzy0lOUGZmlpOcoMzMLCc5QZmZWU5ygjIzs5zkBAUonTlRUoNsx2JmZokan6AkKSJC0jHACEn1sx2TmZn5Rl3S5HQY0A/Ij4g1ZXmdpLOBswFq77gz7b4ev4VXmJltwqjHSt314VX9KjGQ3FOjW1CSakuqB/wd6ANMLdq+pddGxO0R0SUiutTeoWkFR2pmVvPUyASlb1frqh0RhcDhQAPgaoCI+EZSjawbM6s6Ro4cSc+ePRk2bBhr164t3j5x4kTy8vLIy8ujTZs2jBs3DoDrrruOHj16cMwxx7BgwQIAhg8fTvfu3enWrRtPPfVUVs6jNDXuSzjjmlNv4E+SzgNaAV2AkyVdARAR67MZp5nZ5syaNYtPP/2UKVOmsM8++2ywEOGgQYMoKCigoKCADh06MHDgQBYuXMhjjz3GCy+8wNixYxk7diyQJLlp06bx+OOP85vf/CZbp7NJNS5BpcnpB8A4IB84GbggIhaTJKlzJV2TzRjNzLZk2rRp9OnTB4C+ffsyderUjcosXLiweEHDjz76iP322w9JHHLIIUyZMgWA9u3bA1C/fv2cWwq+xgySKGo5pU87Az8B1gONgcsBImKRpE5Ap6wEaWZWRkuXLqV169YANG3alCVLlmxUJnNBww4dOjBjxgzWrFnDs88+u1H5Sy+9lAsvvLDiA98K1T5BSWoCtIuIOelovU+AhcDtwFqgX0QslHQ8sHtE3AIsLJHQzMxySrNmzVixYgWQrKLbvHnzjcrk5+czfnwywrhly5ace+659OnTh86dO2+w4u6dd97JunXrOO200yon+DKqCV18jYA7Jf0ZuAFoCbwEvAM8lCanrsBVwPtFL3JyMrNc1r17dyZPngzApEmT6NGjxwb7Fy1aVNy9V+THP/4xzz33HIMGDSIvLw+AyZMn88ADDxQPpMgl1boFlbaCFkr6E3AbcEtEzJZUF/gncJSk/wDfAJdGxBPZjNfMrKw6d+5Mq1at6NmzJ23atOGSSy5h+PDh3HbbbcCG3XtFTjnlFD7//HPatm3LzTffDCSj+Jo1a0bv3r1p2LAhjz/+eKWfS2lU3RsKkloDB5Fca7oNuCwi/pTu2wGoDTRKE9k2devVb90xWp9+QzlGbWY13cJ7RtGtfQsKCgqyHUqFkzQzIrqU3F5tu/gk1ZK0M/AXYLeIyCe5GXeMpNMk7QvcCKyNiIXgbj0zs1xSbbv40vuYvpD0F+B/JK2PiDsl9QXuAlYCV0bE11kN1MxsE7576lUU1PCpjqplgkpbRwOAayJioqS1wAWSGkTELZK6k3TrzfdoPTOz3FTtuvjSaYz2BPYFfiGpdkQ8CtwLXC3px8DyiJgP7tYzM8tV1SJBZazntAvQMiIeAf4B7A5cnBabCrwIzPE0RmZmua9aJKh0+qIBwNPAvySNBz4GJgMHSnoSeBi4OiJezWKoZmZWRtXiGpSkvUimLjoFmAP8HzAcuJSk5XQc8H5ETMtakGZmtlWqdAtKiVbAlUBrYHXafXcWycSvl0TEkoi428nJzKxqqZIJquiaUyQWATeTzK/3fUm7p0mqaK49MzOrgqpkF196zak/cALwCnA/cC1wHnCEpOkkgyMuyl6UZma2ParkVEeS2pAs0/4IsBOwB3Ah0A4YSzJj+YMR8Xhl3OfUpUuXmDFjRkUewsys2qryUx1lDCXfG/g+8NeI+ANwEzAv/Xc+MBqoB7SWtLPvczIzq5qqRILKWKb9aJLh4hcBP5bUKiI+I7ne9B7JbOXTgUlAV6Awa0Gbmdl2yekEJak2FF9zOhj4OdAPOAxYTDJTxM7pZK9/Akam5e8hGcG3PDuRm5nZ9srZBCWpJfCqpN3TTUcCPUlWx10PnAN0AP5fmqQWRMTbkmoBRMSqrARuZmblImcTVEQsBp4Bnpf03YgYRzJS7+eSukXE58D5JAMjWmS8ztMYmZlVAzmZoDJaQSNIJnl9WVLriBhDkrRGSuqR3gN1YkS8mb1ozcysIuTkfVARsV5SrYhYHxGXpgP4/iOpa0RcJ6k+8BtJp5Ks62RmZtVMTiYoKDVJTZN0RERcKWmPiFiW5TDNzKyC5EyCyhhKXjsivoHiJFU7Ir5Jk1R9YKakthHxQZZDNjOzCpQTCSojOfUBekj6EnggIj6NiG8yWlK/kHRbRPj+JjOzai4nBkmkyeko4AbgCWAUcJ6kOun+9UUDJ4C3sxOlmZlVppxIUKljSCZ7/Qb4FLg1ItZl3Ky7Pv3XUxeZmdUAWe/ik9QO+BJ4g2QKo92AIRExX9IwoC5wZ/YiNDOzbMhqC0rSTsD/AvuTrIS7O/CHiPhI0kHAr4DPshiimZllSaW3oEqM0lsq6RWSJHU08BfgOElnA42A30bEE5Udo5mZZV+lJah05vFF6ai8DkD9iJgXEdek8+0dGRG3SXocEFArIj6ojPWczMws91RKgkrXcrpd0kjgXZJRerUkrSdZaPBLYADwdER8nPlaJyczs5qp0lbUTYeMtwdOj4jfpLOV3wosARYCl5IMjni4UgIqR112qR0zzm6c7TDMrKoY7ZWAMpW2om6FtqAkNQLWRMQ6oA3J8PHhkhpGxC+AkyT1A75Lsrjg5xUZj5mZVR0VPYqvO5AvaQAwgeQep47ACZJuBoiIxyLiDpJ1nqZXcDxmZjlt5MiR9OzZk2HDhrF27dri7RMnTiQvL4+8vDzatGnDuHHjAJg/fz4DBgzgqKOO4rLLLgNgzJgxdOvWjW7dunH33Xdn5TzKQ4V38Ul6EsgDjo2Ip9NtOwHTgSkR8dN0W610xogqNyjCXXxmtlVK6eKbNWsW11xzDXfffTdXXHEF7du3Z+jQoRuVO+qoo5gwYQJt27Zl6NChXHvttey6667F+99//33at29PYWEhhx56KLNnzyadcDsnldbFV2EtKH1bGxOBx4HrJO0AyfBykmXbj5S0X5qUPFOEmdVo06ZNo0+fPgD07duXqVOnblRm4cKFrFmzhrZt27J27Vo+/PBDLr74Ynr16sW0adMAaN++PQB169aldu3alXcC5azcr0FltID2TFtFtwK3SroDeBnYL70Jd1dgbyckM7PE0qVLad26NQBNmzZlyZIlG5V58MEHGTx4MACLFy/mtdde47777qNevXocf/zxvPzyy8Vlb7jhBoYMGZLTrafNKfcElU78ejxwBfCmpIbAhRFxpqS7JM0GagO/cnIyM/tWs2bNWLFiBQDLly+nefPmG5XJz89n/PjxxeX33HNP2rRpAyQtpnXr1lGnTh2efPJJpkyZQn5+fuWdQDkr9y4+Sd8Dfk8yM8QDQE/gSkntI+LHJPPtDY2Ix8r72GZmVVn37t2ZPHkyAJMmTaJHjx4b7F+0aFFx9x5Aw4YNadGiBcuWLeOrr75izZo11KlThzlz5jB27FjuuusuatXKpTnBt065Ri6pI/AacCZwAHAxcBDQGLhX0gERURARs8vzuGZm1UHnzp1p1aoVPXv2ZO7cuQwePJjhw4cX78/s3ity5ZVXcvzxx9OrVy/GjBkDwIgRI1iyZAn9+/cnLy+P5cur5n1X5TaKT9IRwH3A4RHxsaQxwKp0KqOhwM9IbtJ9q1wOmEM8is/MSsqb8FXpO9sdUequgoKC8g8mx1XojbqS9gGuBn6aMVXRbODCdAaJAcDF1TE5mZlZxSivQRJNgR2BM0iGlAP8B6gHnACMjYiNx0uamVVTBWc0Kn3n6IJKi6Mq26ZrUEX3OElqmw5++A9JclonaSxARMyPiH8AP4qIf6uqjnM0M7Os2KYElQ4lPwG4H/iDpL8CXwM3ArtLuiajbNHaTx5SbmZmZbatLai9gRFAb2ASyUi9D4EZwG0kSWqv8gnRzMxqojJdg5JULyIK08f1geXAU8DpwI+AEyPiK0mHRMSLkuZFRNUc12hmZjlhiwkqHYXXV1IAi4G+wIPAgcCewKkR8b6k3sC1kk6IiI8qMmgzM6v+NpugJDWNiOWS3gXuAVoBfSJijqRngbokS2f8FzgPGOnkZGZm5aHUa1BpV94rkn4WEfNI1nJ6D+gCEBG3AX8jWWhwN+CiiHi0Ko/Wk9RT0s7ZjsPMzDbTgoqINZJOAx6W9GlEHCqpC/A7STtFxPUkgyJeiojPMl5XJUfrSeoOjAPOAr7IcjhmZjXeZrv40gEP/YAnJTWPiNvTlXBHSNof2Ac4F/hsc++T6yTtAfwa+HtEzJRUu2h4/GZeczZwNkDtHXem3dfjKyFSM6sWRpU+V/aHV/WrxEBy2xaHmUfESyQzk18t6acR8SQwElgD/D4iZlVwjJWhCbAEGCxpry0lJ4CIuD0iukREl9o7NK34CM3Mapgy3QcVES8DfUiWzbgwIuZExLkR8URVvOaUMRPG/unyIB8BvwMmA+dJ6pBZzswsF4wcOZKePXsybNgw1q5dW7x94sSJ5OXlkZeXR5s2bRg3bhwAN998M127dqVr16488MADAPzzn/9k7733pkuXjeZmzTllvlE3TVLHkySptpJqp9ur3DWndCaMY0lmwhgMzCOZT/BfwFLgV5L2rIrnZmbV06xZs/j000+ZMmUK++yzzwYLEQ4aNIiCggIKCgro0KEDAwcOBOCWW25h2rRpFBQUcOWVVwLQq1cv5syZk41T2GpbNZNEOuferhHxUVm6wXKVpN2An5Pc0/UksBL4OCJmkiyyuIRkolszs5wwbdo0+vTpA0Dfvn2ZOnXj+bcXLly4wYKG7du3Z/Xq1axcuZJmzZoB0KJFC+rVqxpfb9sym/lKSLq/qkoLIx0yXzsi/itpJ+BL4N/AqcCJQL+IWCZpULr99xGxOnsRm5ltaOnSpbRu3RqApk2bsmTJko3KlFzQsF+/fuy7775888033HHHHZUWa3nZ6gRVlJSqUHKqDXQFOklaCfQAbgC6kcwh2DsiPk2vRV0BfBARr2UpXDOzTWrWrBkrVqwAYPny5TRv3nyjMvn5+Ywfn4woXrFiBbfeeivvvPMOhYWF9OrVi2OPPZaqdGm96i5WX0ZpV+QXJC2l64CnIuId4Ergv8AFkq4F7gBGOTmZWS7q3r07kydPBmDSpEn06NFjg/2LFi3aoHuvVq1aNGzYkAYNGtCoUSMKCwupIu2KYtU6QWWMwnsLeJdkEcX9Je0aEbOBHwJzgTeAcyPiYY/cM7Nc1LlzZ1q1akXPnj2ZO3cugwcPZvjw4cX7S3bvNW7cmBNPPJHDDz+c7t27c/7551OrVi0KCgro3bs3b7/9Nr179+azz3L3NlZVtYy6tSQdDPwuIk6U1I5kmZCvI2KUpBZAu3RwxDar37pjtD79hu2O1cxqroX3jAKgW/sWFBQUZDeYSiZpZkRsNO69WregUq8B9SQ9FBEfknTlNZD0L5IWVXkte29mZuWo2ragJO0CLE/XqaoF/APYISKOl9QcOIFkaPnT23sst6DMrLzUxKmOakwLSlJtSS2Be4FTJTWKiPUkQ8qbSnokIpZExPiIeNrXnMzMclO1SVBFiSYivomIxcBvgVOAkyQ1TkfzPUiyHH1xpq4qw+XNzGqaanP9JWP6oiEkI/YeBS4CbiJpOa0E+gEnR8Rb2YvUzMzKojq1oDoBlwKvkyyu+HegOXA+8B2Sa063ODmZmVUN1aIFlXbZTQTGRsTt6bY5wGXAySTdfXUiYm1VmqLJzKwmqxYtqIiYASwCzsvY/Gy6rXEk1qZlnZzMzKqAKpmgMtZzOkjSDwDSIYqFkh6X9B1gP+AIksUIzcysiqmSXXzpgIijgRuBlZJeAq6NiK6SppJMXzQBGBYRr2cxVDMz20ZV6kbdoutHkuoAvwQeI0lG15MMjLgpIj6Q9BhQLyKOznxdRcXVpUuXmDFjRkW9vZlZtVYtbtRNk9NAklkhBgEd0/ubriQ5l19J6hgR/YA2kv5e9LpsxWxmZtumSiWodCj5COA+ktbTGEk9ImIRcBWwHmgAEBF7A7/JUqhmZradcjpBSWol6dT0cRtgDPB2RORHxBiSm3BvlPT9iFgI/Dwi5kiqC5BODmtmZlVQziaodKTeEUA/SacBn5Bcb9pFUk9JtSPiNpLBELekS7l/A1A0pNzMzKqunB3Fl15vegaoC/QGviZpQY0mWWhwvaTpEXGTpH9FxNLsRWtmZuUtJxNUxsSvSyU9TtLS65vuHkMyM8RPAAEvRMTHWQnUzMwqTE4lqHSdpoYR8alSEbE8HTYOSZJaD1xOkqjcajIzq6ZyJkFJaghcCDSUdHNEfLKJJLUeGAzUjojfZTVgMzOrUDkzSCIiVgNFd7ueIalN0f1LRUkKeAJ4HHgzS2GamVklyXqCktRa0qEAEfEY8ACwE0mSalciSS0D/hERc7IWsJmZVYqsJShJtSTtCMwDXpY0WtIvgPdIWkprgR9L2jVzJoh05ggzM6vmstmCiohYAfwMKCRZVBDgYaA/0BnYCzhX0nc8XZGZWc2SlQQlqRXwkqSmEXEXcBYwFHgEOJ5k8cEGJEnqYmDHbMRpZmbZk5VRfBGxSNI7wHRJh0XE3yQ1Bl4Cjo+IZyW9DKwD9oqId7MRp5mZZU+ltqAk7VD0OCJOBaYAr0raMSJuBUYCD0nKi4hVEfF1RMxOX6vKjNXMzLKr0hKUpN2AtyTdJOlCgIg4G7iHJEk1iYjbgcuAxyQ1y0xKvgZlZlazVEoXn6QGQHPgfZKl2HtL2i/d9lugB/C4pL4RcbOkh9Mh5WZmVkNV+Iq66YCIfJKReYcCRwFfAq8AHYFuQBOSCWDfATqlca2r6JVwy0uXXWrHjLMbZzsMM8tlo5dnO4KcVdqKuhXWgspILnWBwnQmiGfS61BHpce+BbgT2Bl4EZiTeZ9TVUhOZmZWMSryGlTT9N8AGhZtjIhHgcnA7sBwYI+I+By4MSKe8WAIM6tpRo4cSc+ePRk2bBhr1367nN3EiRPJy8sjLy+PNm3aMG7cOAA6duxYvP2pp54C4PLLL+fII4/ke9/7HjfddFNWzqO8VUgXn6T6JDNE3ETSSpoLHBwRqzLK9CLp1nsfuBFYU1VbTO7iM7MtKqWLb9asWVxzzTXcfffdXHHFFbRv356hQ4duVO6oo45iwoQJtG3bli5dujBjxowN9hcWFlKvXj3WrVvHgQceyJw5c6hdu3aFnEp5K62Lr0JaUBGxBjgN+DUwCpgG7CXpAEkdJTUhSWDPA/9Kh5NXyeRkZrY9pk2bRp8+fQDo27cvU6dO3ajMwoULWbNmDW3btgVg1apVfP/73+fUU09lyZIlANSrVw+ANWvW0KFDhyqTnDanwrr4IuJF4FiSJTSGkXTn3ZP+PAE8B7wYEW9VVAxmZrlu6dKl7LhjMllO06ZNixNOpgcffJDBgwcXP586dSrPPfccffv25bLLLivePmLECDp27Ei3bt0qPvBKUKH3QUXETOBI4AtgRkQcEBHfIxkkcUxEfFCRxzczy3XNmjVjxYoVACxfvpzmzZtvVCY/P58hQ4YUP2/RogUAQ4YMYdasWcXbb7jhBt5//30mTpzIggULKjjyilfhN+pGxBskQ8yvlHRJuq0Q8DLtZlbjde/encmTJwMwadIkevToscH+RYsWbdC9V1hYyJo1awCYMmUKe+65J0Dxtvr167PDDjvQoEGDyjqFClMpN+pGxMuS+gOTJd0PfBIR6yvj2GZmuaxz5860atWKnj170qZNGy655BKGDx/ObbfdBmzcvbd06VKOO+44GjVqRP369bnzzjsB+NnPfsabb75JYWEhp512GjvttFNWzqc8VfiNuhscLJlzb0WlHbCSeBSfmQHkTfiq9J3tjih1V0FBQfkHU4VU6ii+zViZBuN7nczMbLMqdbmNoqHkHlJuZtVRwRmNSt85uqDS4qgusrmirpmZWamcoMzMLCc5QZmZWU5ygjIzs5zkBGVmZjnJCSqDh7+bmeUOJyg2SExV/9ZrM7NqwgmK5L4sSccAD0pq6paUmVn2VeqNurlKUifgYuCSiFhelgQl6WzgbIDaO+5Mu6/HV3CUZlaljXpsi0U+vKpfJQRSddT4FpSkhiTrVh0ItIOyzXQREbdHRJeI6FJ7h6ZbKm5mZlupRiaoohaSpB2ANcCtJMvOHyvpyGzGZma2JSNHjqRnz54MGzaMtWvXFm+fOHEieXl55OXl0aZNG8aNGwfAzTffTNeuXenatSsPPPBAtsLeajUyQaXXnE4A/gY8DPQB8oHXgVMkHZXN+MzMSjNr1iw+/fRTpkyZwj777EN+fn7xvkGDBlFQUEBBQQEdOnRg4MCBANxyyy1MmzaNgoICrrzyyixFvvVqZIKS1B0YCZwDLAZ+HhFvAw8CnwCnSfKIPjPLOdOmTaNPnz4A9O3bl6lTp25UZuHChRsscti+fXtWr17NypUradasWWWGu11qxCAJSbUj4puMTc1JuvSOBPYCfpRuXwr8GWgWEUsrN0ozsy1bunQprVu3BqBp06YsWbJkozIlFzns168f++67L9988w133HFHpcW6vap9gpJUH+gu6RWSQRCtgfrA/wANgVMj4kNJQ4ChwGkR8UG24jUz25xmzZqxYkWy7uvy5ctp3rz5RmXy8/MZPz4ZWbxixQpuvfVW3nnnHQoLC+nVqxfHHnssVeFumprQxdeYJCndDzwEzE3//S8wD2gkqQ8wGhgfEauzEqWZWRl0796dyZMnAzBp0iR69Oixwf5FixZt0L1Xq1YtGjZsSIMGDWjUqBGFhYVUlSX5qn2CiogvgS+AHwCvAOvS7r4z0yIXARcCv4qIR32Trpnlss6dO9OqVSt69uzJ3LlzGTx4MMOHDy/eX7J7r3Hjxpx44okcfvjhdO/enfPPP59atarGV7+qSibdVpIOA94HWgFHAx2BOyNihqRdSAZJ1I+Ildt6jPqtO0br028oj3DNrAZaeM8oALq1bwFAQUFBFqOpfJJmRkSXkturRhrdPicAL5Ikor+TtKbOkHQecC3QYnuSk5mZVYxqO0hCUp2IWBcRv5ZUCEwCjgH+BJwKnA5cGRELshmnmdl3T70KgAJPdbSBatmCknQo8GtJLQEiYjTJDbkPAXUiYhxwdET8y9eczMxyU7VMUMByoBdwrqQW6babSYaXPyGpAfAVlG3ePTMzq3zVqosvbTnVJZkNYiDJVEaSdDPJPVCPAw9ExNfZitHMzMqmyicoSUrn1utJMgjieZL7nvJJrjXdDlxPMt/eTyJiZtaCNTOzMqvyCSpNTr2AQcCQiHhJ0l7ABOBz4CfAzsDVETEve5GamdnWqNLXoDIGOAwCzidNuOnEr+OAwyJidUR87ORkZla1VMkWVFG3Hsmkr19GxIWSagF/l7R3RBSSnNv+6Vx8hR4MYWZWtVTJFlTarXccMFHSeEmdIuJ8YDLwuaTfAicBf4mINU5OZmZVT5VqQWUMiGgF/Az4fyQzRYyQ9LeIOEvSCpK1nvIiYqakuhGxdnPvu70O2LUpM3yDnZlZuapSCSpNTj2AZsDciCiQ9BzwW2BYOnvExZLqAg9L6hgR/81mzGZmtm2qRBefpNrpv0cA/wT6A+dIOisSY0lG7P1YUpOIuIhkmHnrrAVtZmbbJadbUJKaA6siolBSZ+A04JyIeFhSPnBt2uv3l4j4raQ9iyZ+jYifZTF0MzPbTjnbgpK0A/BL4DJJdYCDgEOAzmkr6WngYmCUpOEAEfFu1gI2M7NylbMJCigEpgM7ABdFxF+BPwIdgCMlNYqIZ4BzSFbGNTOzaiRnu/giYp2kx0gSVX9Jv4iI6yU1BAYD9SQ9GRGTsxupmZlVhJxNUFCcpJ5Kn2YmqXokSWoq6azkZmZWveR0gpJUq0SSOlbSqIi4StKjEfF5VgM0M7MKkzMJKuMm3OIbayNivaTaGUmqDtBP0h4R8UF2IzYzs4qUMwkqTU79gOGSCoAZEfF8RHyTkaQeB16MiMXZjdbMzCpazozik7Q3MBx4mmTk3pmSjgHITFJOTmZmNUNOJChJB5IsNPiviBhHsvDgVGBo2qoiIr7JYohmZlbJciJBRcRsYDbwq/T5B8ATwAzgR5JaZjE8MzPLgqwkqKKFBiW1k3QAQEQcDbwraVr6/GPgEWCku/XMzGqerCSodEDEAOBB4HJJEyS1jIh+wAJJs9JyH0XEJ9mI0czMsqvSEpSkBhmPuwO/A44BHiJZsv1/Je0cEYOBT9IyZmZWQ1VKgpK0E/APSU3STYuB84FDSebSOxzYC7hH0m4R0T8iplVGbGZmlpsqPEFJqhcRS4HzgF0kHR4Rb5MMgPgBcGdEzCMZubcj0LCiYzIzs9xXoQkqHX13p6TvRcQC4GhgYvp8PfAGcJKkC4FhwM8j4p2KjMnMzKqGCp1JIiIWS/oYGCHpmoj4k6RvgL9LGgpMBOoB/YCr3a1nZmZFKqwFlS4yCMkS7XuSJKWDI+JW4Abgr8BeEfFnYEhEPFo0/NzMzKzCElQ6d15v4E5gDEl33tVpkroFuINk4EQzoGhy2KioeMzMrGqpkASV0RI6CngsIv4dEUOAmcBdkrpExB+BvIhYll6PMjMzK1YhCSqjJTQPaFw0VVFEXJoe81JJOwK+CdfMzDap3AZJZKzn1BPYFficpMU0FDhO0svp8d4Ero2IFeV1bDMzq37KpQWVLoURkvoAtwEBTAb2IBkQcQRwOcm9TuMj4sXyOK6ZmVVf29WCktQ8Ipak6zXtBPwE+CHJDbdzgFkR8ZmkKSRrPDWPiPe2O2ozM6v2trkFJakdMFPS/wKks0W8ApwBXAecmCanM4FDI2Kpk5OZmZXV9nTxrUtfnyfp+nTbWpJrTmdGxHvpQoS/AOpuX5hmZlbTbHMXX0TMl3QTUBvYWdLVETFS0j7AaEmFQCfg0oh4rpziNTOzGmKrEpSk9kDXiLg33TSLZPDDpcBgSWMjYrikg4AWwJKIeK1ohF+5Rm5mZtWaypo3JNUD3gbaAFcBLwEFQH+gGTAF+BnwVURcWAGx5ixJK4G3sh1HDmpJsrSKbcx1UzrXzaZV53ppGxE7l9xY5hZURBRKOoFkgcEewMvAoyStqB3SiWBvIJkYtlO6hEZN8VZEdMl2ELlG0gzXy6a5bkrnutm0mlgvWzVIIiJmAScA+wNNSJbIAPiupI7A68DPalhyMjOzCrDVgyQiYrakviQ34p4bEeenE76uSufUW1nOMZqZWQ20TaP4IuLldKbyxyS1jIgbyzmuqub2bAeQo1wvpXPdlM51s2k1rl7KPEhiky+WDiNpSe0HzPes5GZmVl62K0EBSNrRE7+amVl5K4/JYlfCBmtAmZmZbbftTlBFN+DWhBtxJfWV9JakdyWN2sT++pLuS/f/J52vsNorQ738QtI8SbMlPS2pbTbizIYt1U1GucGSQlKNGEZclnqR9MP0czNX0j2VHWO2lOH/UxtJz0p6Nf0/dVw24qwUEeGfMvyQTOn0HtAeqEdy/1enEmXOA/6cPj4FuC/bcedIvRxFcq8cwLk1oV7KWjdpuSbA88B0oEu2486FegE6Aq8CO6XPv5PtuHOobm4nGUENyXRyH2Y77or6qZAVdauprsC7EfF+RBQC95LcE5bpBOCv6eN84Ac1oOtzi/USEc9GxH/Tp9OB3So5xmwpy2cGYCxwNfB1ZQaXRWWpl7OAmyNZJYGI+LySY8yWstRNkCxpBNAU+KwS46tUTlBltysbLlE/P922yTIRsQ5YTjInYXVWlnrJdCbweIVGlDu2WDeSDgF2j4jHKjOwLCvLZ2YvYC9JUyVNT++9rAnKUjejgdMkzQf+DVTbqeXKbcl3sy2RdBrQBfh+tmPJBZJqAdeTrKFmG6pD0s2XR9Lifl7SARGxLJtB5YihwISIuE7S4cDfJO0f1fA2H7egyu5TYPeM57ul2zZZRlIdkub3l5USXfaUpV5Ib+z+DTAgItZUUmzZtqW6aUIybViBpA+BbsDDNWCgRFk+M/OBhyNibUR8QDJRdcdKii+bylI3ZwL3A0TEi0ADkolkqx0nqLJ7GegoaY90ZvdTgIdLlHkYOD19PAR4JtIrmdXYFutF0sHAbSTJqaZcS4At1E1ELI+IlhHRLiLakVyfGxARM7ITbqUpy/+lh0haT0hqSdLl934lxpgtZambj4EfAEjalyRBfVGpUVYSJ6gySq8pXQBMAt4A7o+IuZJ+L2lAWuwOoIWkd0lWEi51WHF1UcZ6uQZoDPxT0muSSv6Hq5bKWDc1ThnrZRLwpaR5wLPALyOiuvdGlLVuLgbOkjQL+AdwRnX9Q3i7Z5IwMzOrCG5BmZlZTnKCMjOznOQEZWZmOckJyszMcpITlJmZ5SQnKLNyJOmbdCj965IekdQsY99+kp5JZ6p+R9LvMudqlHSspBnpDN6vSrpuM8d5SNL0EtsmSBpSYtuqjMd7Sfp3euxXJN0vqVWJ8rUk3ZjGP0fSy5L22I4qMdtmTlBm5Wt1RHSOiP2BJcD5AJIaktxweVVE7A0cBHQnmQEfSfsDfwJOi4hOJFNCvbupA6RJ71CgqaT2ZQlKUgPgMeDWiOgYEYcAtwA7lyh6MrALcGBEHAAMApaV7dRLPbanVLNt4gRlVnFe5NuJPk8FpkbEkwDp7O4X8O3N3L8CroiIN9P930TEraW874nAIyQzXZ9SxlhOBV6MiEeKNkREQUS8XqJca2BB0bxuETG/aEbxdJ2iVyTNkvR0uq152pqbnU7qemC6fbSkv0maSjJX3M6SHkhbZC9L6lHGuK0Gc4IyqwCSapNMR1M0a8Z+wMzMMhHxHtBY0o4kc/JtsH8zhpLMIPCP9HFZlPX97weOT7spr0unqULSzsD/AYMj4iDgpLT8GODViDgQ+DVwV8Z7dQJ6R8RQYBzwx4j4HjAY+EsZ47YazE1vs/LVUNJrJC2nN4CnyvPN02tGHYEXIiIkrU1nsn6dZJ2gkrZqqpiImC9pb6BX+vO0pJOAHYDn04lbiYgl6UuOIEk4RMQzklqkCReSyV5Xp497A50yLrntKKlxRBRfIzMryS0os/K1OiI6A20BkV6DAuaRXDcqll4/WhURK4C5JfeX4ofATsAH6Qzo7fi2FfVluq/o/ZsDi9OnZX1/ImJNRDweEb8ErgQGluV1m/BVxuNaQLf0+lzniNjVycm2xAnKrAKk15guAi5OBwn8HTgiXXakaNDEjcAf0pdcA/xa0l7p/lqSztnEWw8F+mbMgH4o316HKgBOTmfBhmSdqWfTx/cA3SX1K3ojSUemgzPI2HaIpF2KYgAOBD4imWn9yKIRfWnyA5gC/CjdlgcsThNuSU+SsbCepM6bKGO2AScoswoSEa8Cs4GhaVfXCcBvJb0FzCFZWuFPadnZwAjgH5LeAF4HNhihJ6kdScuseHh52uW2XNJhEfEoScKYmXYz9gBGpuVWA/2BC9Nh5vNIRhCWXKbhO8Ajkl5PY18H/CkivgDOBh5MZ9G+Ly0/GjhU0mzgKr5dbqaki4Au6WCKecCmkq/ZBjybuZmZ5SS3oMzMLCc5QZmZWU5ygjIzs5zkBGVmZjnJCcrMzHKSE5SZmeUkJygzM8tJ/x8tPzDPLM4bygAAAABJRU5ErkJggg==\n" + "image/png": "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\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-03-15T09:32:52.987873\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, "metadata": { "needs_background": "light" - } + }, + "output_type": "display_data" } ], "source": [ @@ -244,15 +4315,8 @@ "## Scikit Learn Compatible Imputers. \n", "\n", "You can also use any other scikit-learn compatible imputer as an imputing strategy.\n", - "eg. [feature engine](https://feature-engine.readthedocs.io/en/latest/index.html) library provides a host of other imputing stratgies as well. You can pass them for comparision as well." + "e.g. [feature engine](https://feature-engine.readthedocs.io/en/latest/index.html) library provides a host of other imputing stratgies as well. You can pass them for comparision as well." ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] } ], "metadata": { @@ -276,4 +4340,4 @@ }, "nbformat": 4, "nbformat_minor": 4 -} \ No newline at end of file +} diff --git a/docs/tutorials/nb_metric_volatility.ipynb b/docs/tutorials/nb_metric_volatility.ipynb index 56f3f1ed..c6c2eac0 100644 --- a/docs/tutorials/nb_metric_volatility.ipynb +++ b/docs/tutorials/nb_metric_volatility.ipynb @@ -13,7 +13,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The estimation of AUC of your model could be influenced by, for instance, how you split your data. If other random seed was used, your AUC could be 3% lower. In order to understand how stable your model evaluation is, and what performance you can expect on average from your model, you can use the `metric_volatility` module.\n", + "The estimation of AUC of your model could be influenced by, for instance, how you split your data. If another random seed was used, your AUC could be 3% lower. In order to understand how stable your model evaluation is, and what performance you can expect on average from your model, you can use the `metric_volatility` module.\n", "\n", "### Setup" ] @@ -49,11 +49,11 @@ "### TrainTestVolatility\n", "The class that provides a wide functionality for experimentation with metric volatility is TrainTestVolatility. Please refer to the API reference for full description of available parameters.\n", "\n", - "By default, the class performs a simple experiment, in which it computes the metrics on data split into train and test set with different random seed at each iteration. Having computed the mean and standard deviation of the metrics, you can analyse the impact of random seed setting on your results and get a better estimation of performance on this dataset.\n", + "By default, the class performs a simple experiment, in which it computes the metrics on data split into train and test set with a different random seed at each iteration. Having computed the mean and standard deviation of the metrics, you can analyse the impact of random seed setting on your results and get a better estimation of performance on this dataset.\n", "\n", "When you run the `fit()` and `compute()` or `fit_compute()`, the experiment described above is performed and the report is returned. The `train_mean` and and `test_mean` show an averaged performance of the model, and `delta_mean` indicates on average how much the model overfits on the data. \n", "\n", - "By looking at `train_std`, `test_std`, `delta_std`, you can assess the stability of these scores overall. High volatility on some of the splits, may indicate the need to change the sizes of these splits or make changes to the model." + "By looking at `train_std`, `test_std`, `delta_std`, you can assess the stability of these scores overall. High volatility on some of the splits may indicate the need to change the sizes of these splits or make changes to the model." ] }, { @@ -128,7 +128,7 @@ "source": [ "The results above show quite unstable results, due to high `train_std` and `test_std`. However, the `delta_mean` is relatively, which indicates that the model might underfit and increasing the complexity of the model could bring improvements to the results.\n", "\n", - "One can also present the distributions of train, test and deltas for each metric. The plots allows for a sensitivity analysis of the " + "One can also present the distributions of train, test and deltas for each metric. The plots allows for a sensitivity analysis." ] }, { @@ -157,7 +157,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "In order to simplify the use of this class for the user, two convenience classes have been created to perform the main types of analyses with less parameters to set by the user." + "In order to simplify the use of this class for the user, two convenience classes have been created to perform the main types of analyses with less parameters needed to be set by the user." ] }, { @@ -238,9 +238,9 @@ "source": [ "### BootstrappedVolatility\n", "\n", - "This class allows to perform a different experiment. At each iteration, the train test split is the same, however, the samples in both splits are bootstrapped (sampled with replacement). Thus, some of the samples might be omitted, and some will be used multiple times in a given run. \n", + "This class allows to perform a different experiment. At each iteration, the train-test split is the same, however, the samples in both splits are bootstrapped (sampled with replacement). Thus, some of the samples might be omitted, and some will be used multiple times in a given run. \n", "\n", - "With this experiment, you can estimate an average performance for a specific train test split, as well as indicate how volatile the scores are towards certain samples within your splits. Moreover, you can experiment with the amount of data sampled in each split, to tweak the test split size." + "With this experiment, you can estimate an average performance for a specific train-test split, as well as indicate how volatile the scores are towards certain samples within your splits. Moreover, you can experiment with the amount of data sampled in each split, to tweak the test split size." ] }, { @@ -338,4 +338,4 @@ }, "nbformat": 4, "nbformat_minor": 2 -} \ No newline at end of file +} diff --git a/docs/tutorials/nb_sample_similarity.ipynb b/docs/tutorials/nb_sample_similarity.ipynb index 8172a830..cfd9456f 100644 --- a/docs/tutorials/nb_sample_similarity.ipynb +++ b/docs/tutorials/nb_sample_similarity.ipynb @@ -12,17 +12,17 @@ "\n", "\n", "\n", - "Having two samples `X1` and `X2` with the same set of features, one can analyse how well a model can recognize which dataset a randomly selected row comes from. The Resemblance model assigns label `0` to `X1` dataset, and label `1` to `X2`. Then, the data is shuffled and split into Train split (`X_train`, `y_train`) and Test split (`X_test`, `y_test`).\n", + "Having two samples `X1` and `X2` with the same set of features, one can analyse how well a model can recognize which dataset a randomly selected row comes from. The Resemblance model assigns label `0` to `X1` dataset, and label `1` to `X2`. Then, the data is shuffled and split into a Train split (`X_train`, `y_train`) and a Test split (`X_test`, `y_test`).\n", "\n", "The user provides a binary classifier that is then fitted on the Train split and evaluated on both Train and Test.\n", - "Interpreting such model allows to understand, which features and interactions between them differ between these two samples.\n", + "Interpreting such a model allows us to understand which features and interactions between them differ between these two samples.\n", "\n", - "It is crucial that the model does not overfit or underfit, because interpretation of such model will lead to wrong conclusions. Therefore, you should try fitting the model with a couple different hyperparameter settings, and make sure that `Train AUC` is not significantly higher than `Test AUC`\n", + "It is crucial that the model does not overfit or underfit, because the interpretation of such a model will lead to the wrong conclusions. Therefore, you should try fitting the model with a couple of different hyperparameter settings, and subsequently make sure that the `Train AUC` is not significantly higher than the `Test AUC`\n", "\n", - "Once you have the final Resemblance Model, the `Test AUC` significantly above 0.5 indicates predictive power of the model, as well as the change in the distribution between `X1` and `X2`. The higher the `Test AUC`, the larger the difference between two datasets.\n", + "Once you have the final Resemblance Model, a `Test AUC` significantly above 0.5 indicates the predictive power of the model, as well as the change in the distribution between `X1` and `X2`. The higher the `Test AUC`, the larger the difference between two datasets.\n", "\n", - "Then you can use interpret the model, in order to understand the patterns that the model has learned.\n", - "There are two classes in `probatus` that allow you to analyse, which features have changed between two samples:\n", + "You can then further inspect the model, in order to understand the patterns that the model has learned.\n", + "There are two classes in `probatus` that allow you to analyse which features have changed between two samples:\n", "\n", "- **SHAPImportanceResemblance (Recommended)** - Trains a Resemblance model based on a **tree classifier**, then it uses SHAP library to analyse the differences in features between the two samples. The main advantage of using this method is its high speed, better understanding of the relations in the data and handling of categorical features and missing values.\n", "\n", @@ -223,7 +223,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "We can see that second sample have higher values in all the features." + "We can see that the second sample have higher values in all the features." ] }, { @@ -314,7 +314,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Same as before, we can get more insights into the importance of the features. However, now we can also analyse standard deviation of the permutation importance. High std might indicate that permutation of this feature has a higher or lower impact only in part of the available samples, while low std, indicates a consistent effect." + "Same as before, we can get more insights into the importance of the features. However, now we can also analyse the standard deviation of the permutation importance. A high standard deviation might indicate that the permutation of this feature has a higher or lower impact only in part of the available samples, while a low standard deviation indicates a consistent effect." ] }, { @@ -395,4 +395,4 @@ }, "nbformat": 4, "nbformat_minor": 4 -} \ No newline at end of file +} diff --git a/docs/tutorials/nb_shap_dependence.ipynb b/docs/tutorials/nb_shap_dependence.ipynb index 3d5eca14..0c98a6b1 100644 --- a/docs/tutorials/nb_shap_dependence.ipynb +++ b/docs/tutorials/nb_shap_dependence.ipynb @@ -6912,13 +6912,6 @@ "source": [ "fig = tdp.plot(feature='sepal_length', figsize=(7, 4), type_binning='agglomerative')" ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] } ], "metadata": { diff --git a/docs/tutorials/nb_shap_feature_elimination.ipynb b/docs/tutorials/nb_shap_feature_elimination.ipynb index b230009d..32f3c75e 100644 --- a/docs/tutorials/nb_shap_feature_elimination.ipynb +++ b/docs/tutorials/nb_shap_feature_elimination.ipynb @@ -8,7 +8,7 @@ "\n", "[![open in colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ing-bank/probatus/blob/master/docs/tutorials/nb_shap_feature_elimination.ipynb)\n", "\n", - "Recursive Feature Elimination allows to efficiently reduce the number of features in your dataset, without losing the predictive power of the model. `probatus` implements the following feature elimination routine for **tree-based & linear models**:\n", + "Recursive Feature Elimination allows you to efficiently reduce the number of features in your dataset, without losing the predictive power of the model. `probatus` implements the following feature elimination routine for **tree-based & linear models**:\n", "\n", "\n", " While any features left, iterate:\n", @@ -18,7 +18,7 @@ " 3. Remove `step` lowest importance features.\n", "\n", "\n", - "The functionality is similar to [RFECV](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFECV.html), yet, it removes the lowest importance features, based on SHAP features importance. It also supports the use of any hyperparameter search schema that is consistent with sklearn API e.g. [GridSearchCV](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html), [RandomizedSearchCV](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RandomizedSearchCV.html) and [BayesSearchCV](https://scikit-optimize.github.io/stable/modules/generated/skopt.BayesSearchCV.html#skopt.BayesSearchCV) passed as a `clf`, thanks to which you can perform hyperparameter optimization at each step of the search.\n", + "The functionality is similar to [RFECV](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFECV.html), yet it removes the lowest importance features, based on SHAP features importance. It also supports the use of any hyperparameter search schema that is consistent with sklearn API e.g. [GridSearchCV](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html), [RandomizedSearchCV](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RandomizedSearchCV.html) and [BayesSearchCV](https://scikit-optimize.github.io/stable/modules/generated/skopt.BayesSearchCV.html#skopt.BayesSearchCV) passed as a `clf`, thanks to which you can perform hyperparameter optimization at each step of the search.\n", "hyperparameters of the model at each round, to tune the model for each features set. Lastly, it supports categorical features (`object` and `category` dtype) and missing values in the data, as long as the model supports them.\n", " \n", "The main advantages of using this routine are:\n", @@ -26,11 +26,11 @@ "- It uses a tree-based or a linear model to detect the complex relations between features and the target.\n", "- It uses SHAP importance, which is one of the most reliable ways to estimate features importance. Unlike many other techniques, it works with missing values and categorical variables.\n", "- Supports the use of sklearn compatible hyperparameter search schemas e.g. [GridSearchCV](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html), [RandomizedSearchCV](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RandomizedSearchCV.html) and [BayesSearchCV](https://scikit-optimize.github.io/stable/modules/generated/skopt.BayesSearchCV.html#skopt.BayesSearchCV), in order to optimize hyperparameters at each iteration. This way you can assess if the removal of a given feature reduces the predictive power, or simply requires additional tuning of the model.\n", - "- You can also provide a list of features, that should not be eliminated. E.g incase of prior knowledge.\n", + "- You can also provide a list of features that should not be eliminated e.g. incase of prior knowledge.\n", "\n", "The disadvantages are:\n", "\n", - "- Removing lowest [SHAP](https://shap.readthedocs.io/en/latest/) importance feature does not always translate to choosing the feature with lowest impact on model's performance. Shap importance illustrates how strongly a given feature affects the output of the model, while disregarding correctness of this prediction.\n", + "- Removing lowest [SHAP](https://shap.readthedocs.io/en/latest/) importance feature does not always translate to choosing the feature with the lowest impact on a model's performance. Shap importance illustrates how strongly a given feature affects the output of the model, while disregarding correctness of this prediction.\n", "- Currently, the functionality only supports tree-based & linear binary classifiers, in the future the scope might be extended.\n", "\n", "## Setup the dataset\n", @@ -1961,9 +1961,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "You can also provide a list of features that sholud not be eliminated. \n", - "Say based on your prior knowledge you know that features `f10,f19,f15` are important and sholud not be eliminated.\n", - "This can be done by providing a list of columns to `columns_to_keep` parameter in the fit function.\n" + "You can also provide a list of features that should not be eliminated. \n", + "Say based on your prior knowledge you know that the features `f10,f19,f15` are important and should not be eliminated. This can be done by providing a list of columns to `columns_to_keep` parameter in the `fit()` function.\n" ] }, { @@ -6242,17 +6241,10 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "As shown in the plot, ShapRFECV provides superior results for both: CV Validation and Test AUC, compared to RFECV and the baseline model with all the available features. Not only the introduced method allows to eliminate features without the loss in performance, but also it may increase the performance of the model.\n", + "As shown in the plot, ShapRFECV provides superior results for both: CV Validation and Test AUC, compared to RFECV and the baseline model with all the available features. Not only does the introduced method enable the elimination of features without the loss in performance, but it may also increase the performance of the model.\n", "\n", - "When it comes to time required to perform the feature selection in the experiment above, RFECV takes 6.11 s ± 33.7 ms, while ShapRFECV takes 10.1 s ± 72.8 mss, which shows that the latter is more computation expensive, due to SHAP values calculation." + "When it comes to the time required to perform the feature selection in the experiment above, RFECV takes 6.11 s ± 33.7 ms, while ShapRFECV takes 10.1 s ± 72.8 ms, which shows that the latter is more computationally expensive, due to calculation of the SHAP values." ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] } ], "metadata": { diff --git a/docs/tutorials/nb_shap_model_interpreter.ipynb b/docs/tutorials/nb_shap_model_interpreter.ipynb index 79f0e1a0..497362bf 100644 --- a/docs/tutorials/nb_shap_model_interpreter.ipynb +++ b/docs/tutorials/nb_shap_model_interpreter.ipynb @@ -74,7 +74,7 @@ "source": [ "## ShapModelInterpreter\n", "\n", - "The `ShapModelInterpreter` class in Probatus, is a convenience wrapper class that allows to easily interpret the ML models. \n", + "The `ShapModelInterpreter` class in Probatus is a convenience wrapper class that allows us to easily interpret the ML models. \n", "Currently it supports only **tree-based & linear models**.\n", "\n", "### Feature importance\n", @@ -82,8 +82,8 @@ "\n", "- `mean_abs_shap_value_test` - SHAP feature importance computed on the test set. It is an unbiased measurement of feature importance of the model on unseen data.\n", "- `mean_abs_shap_value_train` - SHAP feature importance computed on the test set. It is a biased measurement, because the model has used this data to train. However, the a significant difference between this metric, and the `mean_abs_shap_value_test` might indicate a shift in the data distribution, target distribution, or overfitting of the model.\n", - "- `mean_shap_value_test` - This metric presents how strongly values of a given feature in the test set, push the prediction towards one class or the other. The positive value, indicates that this feature increases probability of positive class, and negative indicates that it decreases it. In the balanced setting it is typically around 0, and imbalanced it has is relative to the majority class. It is crucial to compare it with `mean_shap_value_train`, if it differs significantly, there is possibly a shift in data or target distribution in the test set.\n", - "- `mean_shap_value_train` - This metric presents how strongly values of a given feature in the train set, similarly to `mean_shap_value_test`.\n" + "- `mean_shap_value_test` - This metric presents how strongly values of a given feature in the test set push the prediction towards one class or the other. A positive value indicates that this feature increases the probability of the positive class, and negative indicates that it decreases it. In the balanced setting it is typically around 0, while for imbalanced the value it has is relative to the majority class. It is crucial to compare it with `mean_shap_value_train` - if it differs significantly, there is possibly a shift in data or target distribution in the test set.\n", + "- `mean_shap_value_train` - This metric presents how strongly the values of a given feature in the train set push the prediction towards one class or the other, similarly to `mean_shap_value_test`.\n" ] }, { @@ -94,8 +94,65 @@ }, "outputs": [ { - "output_type": "execute_result", "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
mean_abs_shap_value_testmean_abs_shap_value_trainmean_shap_value_testmean_shap_value_train
f10.3440570.3298130.3440570.329813
f40.1184550.1269200.1184550.126920
f30.0355740.024811-0.029377-0.016373
f20.0073890.0090720.0073890.009072
\n", + "
" + ], "text/plain": [ " mean_abs_shap_value_test mean_abs_shap_value_train mean_shap_value_test \\\n", "f1 0.344057 0.329813 0.344057 \n", @@ -108,11 +165,11 @@ "f4 0.126920 \n", "f3 -0.016373 \n", "f2 0.009072 " - ], - "text/html": "
\n\n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n
mean_abs_shap_value_testmean_abs_shap_value_trainmean_shap_value_testmean_shap_value_train
f10.3440570.3298130.3440570.329813
f40.1184550.1269200.1184550.126920
f30.0355740.024811-0.029377-0.016373
f20.0073890.0090720.0073890.009072
\n
" + ] }, + "execution_count": 2, "metadata": {}, - "execution_count": 2 + "output_type": "execute_result" } ], "source": [ @@ -134,15 +191,1194 @@ "metadata": {}, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-02-11T16:04:50.793625\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "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\n" + "image/png": "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\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-02-11T16:04:50.793625\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, "metadata": { "needs_background": "light" - } + }, + "output_type": "display_data" } ], "source": [ @@ -153,11 +1389,11 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The AUC on train and test sets is illustrated in each plot, to indicate if the model overfits. If you see that Test AUC is significantly lower than Train AUC, this is a sign that the model might be overfitting. In such case, the interpretation of the model might be misleading. In such situations we recommend retraining the model with more regularization.\n", + "The AUC on train and test sets is illustrated in each plot, to indicate if the model overfits. If you see that Test AUC is significantly lower than Train AUC, this is a sign that the model might be overfitting. In such cases, the interpretation of the model might be misleading. In these situations we recommend retraining the model with more regularization.\n", "\n", "### Summary plot\n", "\n", - "Summary plot gives you more insights into how different feature values affect predictions made. This is a very crucial plot to make for every model. Each dot on the X-axis represents a sample in the data, and strongly it affected the prediction (together with predictions direction). The colours of the dots, present the values of that feature. For each model try to analyse this plot with Subject Matter Expert, in order to make sure that the relations that the model has learned make sense." + "Summary plot gives you more insights into how different feature values affect the predictions made. This is a very crucial plot to make for every model. Each dot on the X-axis represents a sample in the data, and how strongly it affected the prediction (together with predictions direction). The colours of the dots present the values of that feature. For each model try to analyse this plot with Subject Matter Expert, in order to make sure that the relations that the model has learned make sense." ] }, { @@ -166,15 +1402,3667 @@ "metadata": {}, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-02-11T16:04:53.255478\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "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\n" + "image/png": "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\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-02-11T16:04:53.255478\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, "metadata": { "needs_background": "light" - } + }, + "output_type": "display_data" } ], "source": [ @@ -189,7 +5077,7 @@ "\n", "This plot allows you to understand how the model reacts for different feature values. You can plot it for each feature in your model, or at least the top 10 features. This can provide you with further insights on how the model uses each of the features. Moreover, one can detect anomalies, as well as the effect of the outliers on the model. \n", "\n", - "As an addition, the bottom plot, presents the feature distribution histogram, and the target rate for different buckets within that feature values. This allows to further analyse how the feature correlates with the target variable." + "As an addition, the bottom plot presents the feature distribution histogram, and the target rate for different buckets within that feature values. This allows you to further analyse how the feature correlates with the target variable." ] }, { @@ -198,15 +5086,2362 @@ "metadata": {}, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-02-11T16:04:54.775424\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "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\n" + "image/png": "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\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-02-11T16:04:54.775424\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, "metadata": { "needs_background": "light" - } + }, + "output_type": "display_data" } ], "source": [ @@ -219,7 +7454,7 @@ "source": [ "### Sample explanation\n", "\n", - "In order to explain predictions for specific samples from your test set, you can use sample plot. For a given sample, the plot presents the force and direction of prediction shift that each feature value causes." + "In order to explain predictions for specific samples from your test set, you can use a sample plot. For a given sample, the plot presents the force and direction of the prediction shift that each feature value causes." ] }, { @@ -228,22 +7463,2462 @@ "metadata": {}, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-02-11T16:04:56.377013\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "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\n" + "image/png": "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\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-02-11T16:04:56.377013\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" }, { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-02-11T16:04:56.909876\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "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\n" + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeoAAAEUCAYAAADk9XVTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAA2G0lEQVR4nO3deXwU9f3H8dcnCQkhIeEUREAUgYpyWPHGq7VaLWjPn1KP2mut2vZnW63VatVWsdrLttZjf623VntYqyJWq8UDr9paqUQUkUNAkIQjCZCEJN/fH98J2exujiXJziR5Px+PeWR35juzn/lmdz4z3zm+5pxDREREoikn7ABERESkdUrUIiIiEaZELSIiEmFK1CIiIhGmRC0iIhJhStQiIiIRpkQtoTOzBWb2lbDjSGRmK8zsuBA+d6yZVZtZbpY/d4SZPWtmVWb2s2x+dhSZ2TFmtnoX5y00s0fMbIuZ/bELYjnSzN7qxPzOzPbpbBwSHiXqLDKzmWb2QvAD3mhmC83soGDa2Wb2fJp5UhJGsBFxZnZx0vhxwfjqYFhhZt9rI54vm9mSYOO83sweM7OBXbW+3S3N+jYNp4YdW0cl/3+dc6ucc8XOuYYshxIDyoES59x3kiea2R1mdnVnPyThf5bX2WVF2GeBEcBQ59znOrsw59xzzrlJnQ+r+5nZpUm/xe1m1mhmw4LpQ8zsATOrMLNyM7vXzErCjjvqlKizJPgyPgr8GhgC7AFcBdTuwuK+AGwEzmpl+iDnXDEwB/iBmX08TTxHA3OBOc65gcC+wAO7EEsUDAqSW9PQU9cjTHsCZU5PQOoKewJvO+fqM52xp+/AOOfmJv4WgeuABc658qDI1cBgYC9gPH6H5spQgu1JnHMasjAAM4DNbUw/G3g+zfgVwHEJ74uAKuA0oA6YkTBtHOCAvIRx/wQuTLPcC4GH2ojnE8BrQCXwHnBlms/5YjBtE/A14CBgEbAZuDFp3RYCNwJbgCXARxOmLwC+kvD+S8CbwXL/BuzZSowp65swLR/4D/CN4H1uEMMPgvdXAn/C75xUAf8GpqWrd+Bg4MVgvd4P1iM/oawL1n9pUOY3gAXTxgNPAxX4I9Z78TsWAHcDjcB2oBr4bvI6AaOAh/E7Zu8AX0343CuBPwB3BeuwOPH7kKZODg++D1uCv4cH4+8AduC/T9UkfN+C6bGk6Y8kxPZnYAOwHPhmwjwHA6/ivz/rgZ8H41cF61cdDIeliTPtvMG0PwLrgnV4FtgvYdodwE3A/GDZC4GRwA3479IS4ICk//ElQFkw/XagfzDtGGB1QtlW1zUp9quCetoRxPBl/AHRZcBK4IPg/1Wa9B3+clA3z6ZZZnIsK/C/30VBPTzQFHcw/SL893Qt/rfkgH2CaQXAT4PPWg/cAhQG0x4DfpawnPuB2zqxzTPgXeALCePmA+clvD8f+NuufkZfGUIPoK8MQAl+Y30ncCIwOGn62XQsUZ8Z/AhzgUeAXydMa/rR5wU/kiOAbSQkxYSyR+ITxFVBuYKk6ccAU4KNzNTgR/3JpM+5BegPHA/UAA8Bu+FbCz4Ajk5Yt3rgW0A/4NRgAzMkmL6AIFEDp+AT0r7BelwGvNBKne5c31am74/fAO8LfB94CcgNpl2J35h+NojpQvwGuF9yvQMHAocG8YzD70RckPA5Dt9aMggYi9+YfzyYtg/wMfwGcjg+udzQxv+3xToF5W8K6nl6sOyPJKxDDXAS/vtwLfBSK3UxJKiLM4P1mBO8HxpMvwO4uo3vb4vpwffiX8AP8DtFe+M3yicE018EzgxeFwOHduR/1ta8wfsvAQOD+rwB+E9SjOXB/6s/fgdpOb7lKRd/NPePpLp/AxgT1M/CpnUkITm2t65p4r8SuCcp5neC+YqBB4G7k+rjLvxOeGGa5e2MJSHuV/A7D0Pw38evBdM+jv+t7h8s7z5aJupf4Hf8hgT1+AhwbTBtJP53+xHg9GAdBwbTZuJ3QlsbZqaJ+yj8zkpxwrhZ+B2CwcHwNAm/JQ2t/CbCDqAvDfiEcQewGp+4HgZGBNPODsYl/wAaabkh/zvBhh6/sd1Ac3Jp+tFvxm+E36SVPf+g/InBD3Vz8IP6OUEiS1P2BuAXSZ+zR8L0CuDUhPd/bvoBBuu2luAoMxj3Cs0b4wU0J+r5wJcTyuXgdzb2TBNT4vomDvsmlPkO8FZQHxMSxl9JQlILPud94Mjg/QqSjiwTyl4A/CXhvUvcUOGPcr/XyryfBF5LeN/ic2i5szUGaCDYWAbTrwXuSFiHvydMmwxsb+VzzwReSRr3InB28PoOMkvUhwCrkspcAtwevH4WvxM4rJX/WVuJOu28acoNCpZVmhDj/yVM/wbwZsL7KSS0agV1/7WE9ycBy4LXx9CcqNtc1zRxXUnLRP0ULY8iJ+F3Ept2/BywdxvruTOWhLjPSHh/PXBL8Po24McJ0yYGy98Hv/O+FRifMP0wYHnC+8/gW8nKSZN8MxmA3zV9VxPGjcJvwxqD4UkSWqc0pB90jjqLnHNvOufOds6Nxu/xjsInwCYvOecGJQ74JioAzGwMcCy++RTgr/gjh08kfdQw59xg59y+zrlftRHPfOfcbPze9Sn4hPqV4LMOMbN/mNkGM9uCb9odlrSI9Qmvt6d5X5zwfo0LfqmBlcH6J9sT+KWZbTazzfgmX8MfpbdmWFK9vZkw7c5gmY8555Ymzfde0wvnXCN+ByolJjObaGaPmtk6M6vEn9tProt1Ca+3Eax7cDX1/Wa2Jpj3njTztmYUsNE5V5UwbiUt6yL5c/u3cp5zVDBvouRlZWJPYFTT/yn4X12KP+cIvil3IrDEzP5pZrMyWHbaec0s18x+bGbLgrpcEZRPrM9MvpOQ8B2g7e9kW+vanuS6X4lP0onzv0dm0n7fgs9KXqcmw4EBwL8S1uPxYHyTR/CtD28551Iubu0oMxsAfA7/+0v0B+Bt/NF8CbAM/5uQNihRh8Q5twR/BLB/BrOdif+fPWJm6/BNU/3xF5d1JpZG59xT+Gaopnjuwx/xj3HOleKbua0TH7OHmSXOPxZ/lJ3sPeCcpMRb6Jx7YRc/9yZ8s/QJZjYzadqYphdmlgOMbiWmm/HnNyc450rwG+mO1sVc/BHNlGDeM5LmdWnn8tYCQ5KuxB8LrOngZycva8+kcZksKznO9/BHYon/p4HOuZMAnHNLnXNz8KdCrgP+ZGZFaZaT+kGtz/t5/A7lcUAp/mgUOve9HJPwuq3vZKvr2gHJdT8W33qWuBPRbr100PukrlOTcvzOyn4J61Hq/EVfTa7Bt8TtbmZzmkYGt4gl312ROByZFMen8DvZC5LGTwdudc5tdc5V47crHa3HPkuJOkvM7ENm9h0zGx28H4Nvun4pg8V8Ad8kOD1h+AxwkpkNzTCeU8zsNDMbbN7BwNEJ8QzEH83VBNM+n8ny09gN+KaZ9TOzz+FPAzyWptwtwCVmtl8QZ2lQPmNmdib+fOXZwDeBO80scaN0oJl9OjgCvQB/BX66/8dA/IVN1Wb2IeDcDMIYiD+tsMXM9sBf6JNoPf7cZQrn3HvAC8C1ZtbfzKbijzZ35QjkMWCimX3ezPKCW9gm43diOiI5zleAKjO72Px9w7lmtr813254hpkND1oqNgfzNOJP1TTSyjq3M+9A/P+oAn9kOLeDsbflfDMbbWZD8NcxpLtjoM117YDfA98ys72C799c4AG3C1eFd8AfgLPNbHJwVHtF04SgPv8P+IWZ7QZgZnuY2QnB66PwF4iehd/W/Dr4zuL8LWLFbQzPJcXxBeCupFY08BcxfiWox0L8hYqLuroSehsl6uypwp/retnMtuITwhv4c6jtMrND8Xvlv3HOrUsYHsZfqDKn7SWk2AR8FX+lclOT7E+cc03N6ucBPzSzKvxFNH/IcPnJXgYm4PfqrwE+65yrSC7knPsL/ijq/qB58w38ufS2bE7au/+2mY3Fn1Y4yzlX7Zy7D38l8S8S5vsr/sK2TfjWik8753akWf6F+B2VKvyGLpPbv64CPoy/eG4e/kKiRNcClwVNkRemmX8O/shxLfAX4Arn3N8z+HwAgrqehf++VeCvMJ/lmm+bac/vgMlBnA85f5/3LPzO4nL8//W3+CNd8Bc1LTazauCXwGnOue3OuW34///CYFmHpvmstPPiL7haiW8FKCOzndzW3Ac8gW+dWoa/4KyFDqxre27DX+H/bDB/Df78eZdzzs3Hf++fxm8Xnk4qcnEw/qXg9/V3YJL520fvAr7unFsTJN7fAbcntYS1K0juHwmWl+xL+O/zavz/cW862SLYF1jqDo9I1zKzs/EXiyU3PYfGzK7EXwl7RtixSDjMbAX+e5nxjo9INumIWkREJMKUqEVERCJMTd8iIiIRpiNqERGRCFOiFhERiTAlahERkQjr0V2qiYhI3xA8lOZ3+E6AyoFLgucjtFZ+X3xPdgfiH7RzUfCcBszsHuCjQFFxcfEHF110UeXIkSMnNS33nHPOieE74ml6KM0aF/QJHo/HU5Ybi8X+EkzbuVz8Y16vj8Viv22KKR6PL0hebiwWa7evcR1Ri4hIT/AbfBeiI/C9e93c9ATDZMHTBv+Kf/LeEPwT0O4xs4lBkWuBcc65kquuumrJ+++/v+8111xzQtNyx40bNwD/8JemJ681Jem0y43H4y2WG4vFSoCTgavj8fiBSeF9PRaLFQdDu0kalKhFRCTigme9fwa4PHjS4PP4vgjObGWWD+E7KPmFc67BOfc0vhvTMwGcc4udc7XxeLyoqKjo+Iceeqhq1apVo2Kx2PPAwzNnzhzZ3nJjsVhDLBZrsdxYLLY4FovVBmVdMIzv7Pqr6VtERLLOzB7F93OdzvPOucQe1yYC9c65txPGvY7vn6CFeDz+6M0333xUTU1NUWFh4aZ4PA7AL3/5y6I1a9ZMAS4PPv+msWPHfvGiiy7qt27dutU09z3w+ogRI07EP2P/x/hucr/vnFvQ2qqQ0LlSPB6/Cd+/QCHwGql9Glwbj8d3LjcWi7W23J2UqEVEJOuSEnF7ivF9EiTagu+opYVYLDbLzPrhE+Et+Of7H4tvrl503XXXNX3+eTfeeOMDOTk5D+Kfwd90JLxljz32WInvB7wOOA3fY+H0W2+99S3gA+CieDzetNyjgX8kfP558Xj8G/i+vo9JWC74Z62XJS43Ho9Pj8Viy9paeTV9i4hI1FXj+69OVILvKCdF0LnOJ4FP4C/q+g6+Y6HVieXy8/Mr8/LyCvFd3Db1ildSVFS01jlX5Zyrdc7diW/ePikWi3VouUGz+PNJyyUWi70ci8WqYrFYbSwW27nc9lZeR9QiIpJ1ZjYfSO7HuslzzrnEXvPeBvLMbIJzbmkwbhqwOHnGeDw+Hzjy1ltvTRx9RF1dXeHmzZvfTCr+NpC31157DVm+fHnTueR0y3UE/Z7HYrFFJDS5x+PxF4A7W1mPPNo+R71zuW3RI0RFRCTyzOx+fGL7Cr7L0ceAw51zKck6KD8Vn4hz8N32no+/GKwU3w3no8D2a6655qkVK1bMfPDBB+fMnTt3rXPusfvvv/+iBQsW3IW/jepUIA4c4Jx7Ox6Pd2i5wHH4JvU5sVjs4Xg8Pgjf1fEzycuNxWKJ595TqOlbRER6gvPwF2h9APweODcxSZvZfDO7NKH8mcD7QfmPAh9zztXik/25+ObqTddee+1uo0eP/u/cuXPvAH5fXl7+3QULFnwVf490+YUXXvib7373uw8kXMiWstzgSu8WywV+ClwQi8UeDubrh+/vfAP+fu1vAJ9sL0mDjqhFREQiTUfUIiIiEaZELSIiEmFK1CIiIhGmRC0iIhJhStQiIiIRpkQtIiISYUrUIiISGWZ2d9gxRI3uoxYRkcgws63OuaKw44gSHVGLiIhEWNidcuhwXkREdvrc5z4HvT83tNsRR4vCITd99/Z/hoiIZGDZsmWMH99Wh1O9QkaJWk3fIiISGSNGjAg7hMhRohYRkcjYuHFj2CFEjhK1iIhExo4dO8IOIXJ0jlpERCKjtraWgoKCsMPobjpHLSIiPdPq1avDDiFywr49S6RrLFsH/10ZdhQi0lGHToSRg1NGFxcXhxBMtClRS+/wmevhnfchV41EIpHX0Agf3huevSZlUl6e0lIy1Yj0DnX1sLU27ChEpKNeXgqV26BkQIvRmzdvZujQoSEFFU06/BARkewr6Afz/50yeuTIkSEEE21K1CIikn1V2+GeZ1JGl5eXhxBMtClRi4hIOP6+CGpb3jfd2NgYUjDRpUQtIiLhyM+Dpxa1GKWm71RK1CIiEo7K7XDfcy1GrV27NqRgokuJWkREwvPwP6GhYefbkpKSEIOJJiVqEREJ1wtvhR1BpClRi4hIeLbVwgPP73xbWVkZYjDRpEQtIiLhaWiEP7wAQQdRo0aNCjmg6FGiFhGRcG2rhddXALBu3bpwY4kgJWoREQlXXT388QUAcnKUlpKpRkREJFw7GuC+ZwEYNmxYyMFET8aJetasWUWTJk1aNWDAgMbi4mI9QkZEus7gYnjwYqi+D1bcCnOObLv8AXvDMz+Cqnth3W3wzU80T1t+C2z7vZ9WdS/87QfdG7t0zvotsGydmr7TyDhRb9my5boNGzbsfuyxxw6vrq7OAfjwhz/83O67774tLy/P7bXXXhVdH6aI9CpXnOqHZL/5qm8GHfElOP0XcHMMJo9Jv4yhA+Hxy+HWJ2HoF2Cf8+CJ/7QsM/taGHi6H074YZevhnQh5+DPLzJo0KCwI4mcjBN1TU3NpNLS0sp58+btTMiFhYVvT5w48aeTJ0/+b9eGJyJ9xoAC+MyhcPl9sLUGFi7xD8M48+j05b99MvztNd9kWlcP1TWwZE12Y5auU7MD7n6G+vr6sCOJnIwS9dSpUxe99tprx61atWpIQUGBmzx58jsACxcu/PIzzzzzg/z8/I3dE6aI9HoTR0F9Iyx9v3nc6ythv1aOqA+dCBurYeFcWH87PHwJjEk6v3nvBfDB7b7Ze+q47opcusrStWxfrkeIJssoUS9atGjq9OnTnxk7duzG2tpaKysr26czH15VVaXXet0lrxvU407PV9wfKre1HLdlKwwsTF9+9FD4wrHwv7fB2Bgs/wB+/+3m6affAOO+BnueA/94A/52OZQO6LbwpQsU9GPIquYHnkRh29IdrzNlLrjJvKNmzJixoKKiYsry5cuHZjKtFZl9uEhrJn8T3lwddhTSlkcuhZn7+tf9+/m/NUEXh8+/CZf/3h8dF81pnufbJ8Mx+8HJ16Yu7z8/h3+/C1+60b8fUgwVd0HpGakJH+DNX8FFd8Gjr3bdOknXKsjj3ZevYO9p+4UdSXezTArndVcUIiItzJ7b/LrpQrKrHmgeN6AA8nJgn93hnaD5e9o4WPxe+uUtWrHzaVZA+7v9DrCMto+SbYdMJG/wwLCjiBzdRy0i0bCtFh58GX54mk/ah38ITjkI7n4mffnbn4ZPHeKTeV4uXP45eK7MH02PGebn75cHBf3gwlNg2EBY+GZWV0kyUNQfzjyaIUOGhB1J5HTJEfWsWbMKgfzGxsZc55zNmjWrFGh89NFHd71RXkT6nvPicNv5/gKwiio4Nw5lwRH1zH1h/mX+Vivw550vvRfmfd8n9uffhM//wk8bWOhv7Ro/0jev/2c5nHi1v/hMoqm+Hk45mPXr11NcXBx2NJHSJeeoJ0+e/M6bb745PrHc4MGDGzZu3NjejoDOUUvX0DlqkZ5t+jh47eds3LixLxxVd+856ldfffWY5HGdvfpbRET6sMJ8OOsYAOrq6sKNJYJ0jlpERMLlHHzqUAC2bt0acjDRo0QtIiLh2mMojNsNgNGjR4ccTPQoUYuISHjy8+CMo3a+Xb1a15okU6IWEZHw5OXCZw/b+bagoCDEYKJJiVpERMJTUgj7jW1+W1ISYjDRpEQtIiLhyM2BU49o8cS4DRs2hBhQNClRi4hIOAbkw6kzW4waOrSjXUX0HUrUIiISjpwcOGRCi1HbtqXpUKWPU6IWEZHsM/PPas9pmYa2b98eUkDRpUQtIiLZN7A/zDkyZbTuo06lRC0iItlX3+D7Gk+i+6hTqT9q6R1GlMK76yFX/Q2LRF6Dg9kHQX6/lEmFhYUhBBRtGfee1cXUe5Z0je21sEnPCBbpMXYr9Q87SbJ582YGDRqU/Xiyq3t7zxKJpMICP4hIj1ZRUdEXEnVGdI5aREQiY/jw4WGHEDlK1CIiEhmVlZVhhxA5StQiIhIZtbW1YYcQObqYTEREIqO2trYv9KCV0cVkOqIWEZHI0H3UqZSoRUQkMoqKisIOIXJ0e1a2NTTAx66C98rDjqTn+unZcMrBYUchIt0gPz8/7BAiR4k62+rq4dkyaGgMO5Ke60d/UKIW6aU2bdrEkCFDwg4jUtT0HQY95bJz/rsKKqrCjkJEusGIESPCDiFylKil5+mXBw+/EnYUItINNm7cGHYIkaNELT3P1hq4+5mwoxCRbrBjx46wQ4gcJWrpmV5Y4hO2iPQq6o86lRK19Ez5/eDx18KOQkS6mO6jTqVELT1T1Xa4R83fIr1NcXFx2CFEjhK19FxP/AfqdD5LpDfJy9Ndw8mUqKXnys2FBYvDjkJEutDmzZvDDiFylKil56qugfueCzsKEelCI0eODDuEyFGilp7LOXjoZWjUU95Eeovycj1eOZkStfRsjY3w8tKwoxCRLtKoHe8UStTSs22rgweeDzsKEekiavpOpUTd1w0uhgcvhur7YMWtMOfI1steeAr89waovBfevdm/T/T0VfDB7bDlHvjPz+Hkg7o1dMB3bvLAQt8MLiI93tq1a8MOIXIyvg5+1qxZRUuXLn3zvffeG52Tk0N1dbWSfU9wxan+71UPtBz/m6/6Hr1GfAmmj4N534fXV0DZe6nLMIOzfgWLVsD4kfDEFb67zgcW+un/e5ufr6ERDp4Af78SJn4d1m3qvvUCf0/14lWw/57d+zki0u1KSkrCDiFyMk6yW7ZsuW7Dhg27H3vsscOTk/SJJ544raioqHHo0KF1XReidJsBBfCZQ+Hy+/zjOBcugYf/CWcenb78Tx6C1971ifjttfDXV+CIfZun/3dlc/edzkG/XBgztNtXgx0N8KcXu/9zRERCkHGirqmpmVRaWlo5b968iuRpS5cu/dtuu+3WzYdP0mUmjoL6Rlj6fvO411fCfmM6Nv+R+/oj2USPXArb74dXrvf3OL+6rOvibU1dPdzzbPd/joh0u8rKyrBDiJyMmr6nTp26qKysbIpzjoKCAjd+/PhlZWVl+wDMnDnzFudczqhRo/5SWVl5VveEK12quD9Ubms5bstWGFjY/rxXngo5OXD70y3Hz54Leblw3FTYd3T2zh2vqYBVG2Ds8Ox8noh0i1GjRoUdQuRkdES9aNGiqdOnT39m7NixG2tra60pSZ944olTFi9e/JUJEyacnMnyqqqq+uzrrHjkUth0tx++9yk/NL1/5FL/wJCSAS3nKRngz/m25fwT4axj4BPX+KPZZPUNvsOM46fD7CxcUAa4Rueb8onG/1ev9Vqvd+31ypUrQ48hatt/cxke8cyYMWNBRUXFlOXLl+88+ThhwoT3hw0b9tyLL774P0ccccRvlyxZclZFRUV+BxbX9y7V3V4LJaf7JudsSncx2YAC2HQX7HcBvBM0f9/5TVi7ES65J/1yvvgR+OEcOOoyWL6+7c988gqY9y+44dFOh9+uA/aCf/+s+z9HRLrVypUr2XPPXn9hqGVSuNNXbB955JE31tTUFA0bNmxOZ5clWbatFh58GX54mk/ah38ITjkI7m6lV6rPHwVzT4ePXZmapCftAR8/APrn+6bv04+CoybDM2XdvhoU5vsjfBHp8YYNGxZ2CJHT6W5KKioqPv7BBx8M/Mc//rGjuLiY+vp6q6uro7i4uPGwww479cknn/xjVwQq3eS8ONx2vr//uaIKzo0335o1c1+YfxkMPN2/v3oODB0I/7y+ef57noVzb/X7h1eeCpPH+Cu/l74Pp/7cXyXe3ZyDTx3S/Z8jIt1u3bp1jB8/PuwwIqXTTd8nnXTSOOfczl2gjRs3XrZ06dJPHHLIIYfl5uYufvTRR9s64ammb+m8fXaHpb8JOwoR6QIVFRUMHTq0/YI9W0ZN350+on7sscdWACua3h9xxBHlZubmz5//ameXLdKu/Dw486iwoxCRLlJfn+YC1T4u4yPqLqYjaumcAQW+KX5yB+/9FpFIW7ZsWV9o+s7uxWQioRo0wN+vLSK9wujR+j0nU6KWnis3B06b6Z9BLiK9wurVq8MOIXKUqKXnKsyH/zki7ChEpAv169cv7BAiR4laeq68XDhon7CjEJEuNGTIkLBDiBwlaumZzODTh/rnjYtIr7F+fTtPPOyDtJWTnqm4P8yZGXYUItLFBg8eHHYIkaNELT1TQyMcvV/YUYhIF6urqws7hMhRopae6eMHQL9OP69HRCJm69atYYcQOUrU0vMMLIQz9DQykd5I91GnUqKWnqeuHk44IOwoRKQb6D7qVErUYWjse09O7VIz9/WPDhWRXqegQL/tZDrJl20F/eBbs2FVediR9Fzf/WTYEYhINykpKQk7hMhRpxwiIhIZ6pQjlZq+RUQkMvpAX9QZU6IWEZHI2LZtW9ghRI4StYiIRMb27dvDDiFydI5aREQio7a2ti9c+a1z1CIi0jPpPupUStQiIhIZhYWFYYcQOUrU3ck5WL857ChERHqMAQMGhB1C5ChRd6d7n4WRXwo7ChGRHqOioiLsECJHibo7DQyacNZuDDcOEZEeYvjw4WGHEDlK1N0pL6je+f8ONw4RkR6isrIy7BAiR4k6G/7wQtgRiIj0CLW1tWGHEDlK1N1tQAE8WwY76sOOREQk8tQfdSol6u7WLxfyc2HhkrAjERGJPN1HnUqJOhuqa+Chl8OOQkQk8oqKisIOIXKUqLOh0cFflKhFRNqTn58fdgiRo0SdLR9sgZUfhB2FiEikbdq0KewQIkeJOltycuAx3aYlItKWESNGhB1C5ChRZ8u2WvjDwrCjEBGJtI0b9YCoZErU2fTi21BTF3YUIiKRtWPHjrBDiBwl6mwqyPP3VIuISFq6jzqVEnU2Vek2LRGRtug+6lQZJ+pZs2YVTZo0adWAAQMai4uLG7sjqF7LOXjolc4vZ2MVfOrHUDQH9ozBfc+2/ZkX3wVDz/LDxXf5cU1iN8Okr0POZ+COpzsfm4hIJxQXF4cdQuRknKi3bNly3YYNG3Y/9thjh1dXV+ecdNJJkydPnrxs8ODB9QUFBW7o0KF1Bx100OOzZ8+27gi4x9u8Fd55v2Nlr7zfD8nO/z/Iz4P1t8G934Jz47B4VfplxJ/wOwev/xwW/QIeeRVufaJ5+rRxcFMMPrx3xqsiItLV8vLywg4hcjJO1DU1NZNKS0sr582bVwHQ2Ng4vLi4+O1DDjnk2OOPPz7nwAMP/J9ly5Yd98EHHzzU5dH2Fo/9a9fn3VoDf34JfvR5KC6EmfvCyQfB3c+kL3/nAvjOyTB6GOwx1L9OPHI+/0T46FTo32/XYxIR6SKbN28OO4TIyWjXZerUqYvKysqmOOcoKChw48ePX1ZWVrYPkJglHjrggAP+WVFRcUjXhtpLbK+DBxbCN2ft2vxvr/XdZ04c1Txu2p7wTCsXqS1+zx817yw7zo8TEYmgkSNHhh1C5GR0RL1o0aKp06dPf2bs2LEba2trLUjSLcyePTt3w4YN+5eWlr7d3vKqqqp6/WuXeD64yRurOjRvbW1d6vjqGhoHFrYcX1oEVdvTL6e6hq15zTFU5zr/7HHnWpSvb2jokvXVa73Wa73uzOs1a9aEHkN3v86UpU0kbZgxY8aCioqKKcuXLx+abvq0adNeW7t27X4HH3zw+Hnz5r3XzuIy+/CeZt6rcPoNsGVby/EfmwZPXJF+nlnXwPNv+tc1wf2ETc3SM/eFH82BIy6FbQnnrn/2V1iwGB65NHV5pWfAk1fAwRP8+38tg2Muh6r7WpabeSl85Tg4+yMZraKISFdavnw5e+21V9hhdLeMruHq0rP206dP/9fatWv3mzFjxuEdSNJ9U1EB/M/hrU9/9PvNr5suJLvytOZxW2ugvhGWroUJQfP36ytgvzHpl7ffGD+9KVG3VVZEJGRq+k7VJfdRz549O3f//fdf8v777+83Y8aMA+fPn/9qVyy3V6pvhJMO3PX5i/rDpw+BH9zvk/bCN+Gv/4Qzj05f/qxj4OcPw5oKWLsRfvZwy6Pmuh3+aWnOwY4G/7pRd92JSDjWrl0bdgiR0+lEPWvWrIKlS5e+W1FRMfbAAw+cOn/+/P92RWC91h5DYNSQzi3jppi/KG23L8KcX8DNMdhvrJ/2XBkUf7657DnHw+yDYMq3YP8L4BMH+nFNjv8hFJ4GL7zl76kuPE1PTxOR0JSUlIQdQuR0uul727Zt57z11ltj8/LyeOqpp94qKCgAYI899ih/9913h3c6wt4kNwc+c1jHyyc2eScaMhAe+l76aUdOhuqE889mcP1ZfkhnwY86Ho+IiGRdxheTdbG+dTHZwEKYfxkcsW+oYYmIRNWyZcsYP3582GF0t4wuJtOzvrOpoREOmRh2FCIikTVq1Kj2C/UxStTZ9JEpkJcbdhQiIpG1bt26sEOIHCXqbCnu3/ZtWSIiQk6O0lIy1Ui27KiHE6aHHYWISKQNGzYs7BAiR4k6W/YaAbsNCjsKEZFIU9N3KiXqbOiXC5/N4LYsEZE+atCgQWGHEDlK1NnQP993RSkiIm2qr68PO4TIUaLubrVBxxoH9vr7AkVEOq26ujrsECJHibq71ezwvWXpSkYRkXaNHj067BAiR9kjGz6n27JERDpi9erVYYcQOUrU3ampP+njp4Ubh4hID9GvX7+wQ4gcJeru1C/XD0MGhh2JiEiPMGRIJ3sX7IXUKUd3q9sB+dpDFBHpCHXKkUpH1N1NSVpEpMMGDx4cdgiRo0QtIiKRUVdXF3YIkaNELSIikbF169awQ4gcnaMWEZHIqK2tpaCgIOwwupvOUYuISM+k+6hTKVFnmZp1RERa1weOpjOmRJ1lDQ0NYYcgIhJZJSUlYYcQOUrUWRR/vYGn3t4SdhgiIpG1YcOGsEOIHCXqLDrnScfdq4aFHYaISGQNHTo07BAiR4k6Sxoa/QXuEwfWhhyJiEh0bdu2LewQIkeJOkve2uj/DkCJWkSkNdu3bw87hMhRos6Sf67zR9R6PJ6ISOvUH3UqJeosWbjGJ+pNmzaFHImISHTpPupUStRZ8vwaR65Bv/z8sEMREYmswsLCsEOIHCXqLNjR4HhnM+TlQL46RRcRadWAAQPCDiFylKizoKwC+udCjunJZCIibamoqAg7hMhRos6CV9c7GoPXxQMHhhqLiEiUDR8+POwQIkeJOgueW+3YusO/rtGtByIiraqsrAw7hMhRos6CF9Y29+ZZX18fYiQiItFWW6tnTSRTou5mtfWOFQmP99Z91CIirdN91KmUqLvZG+VQmNf8XvdRi4i0TvdRp+oTiXrjdsenHmqg6IZ69ry1nvvebGy1rHOOi59pYOiN9Qy9sZ6Ln2nAueam69gTDUz6XT05P63njjdaX06TV9c76hOKqa9VEZHWFRUVhR1C5GScqGfNmlU0adKkVQMGDGgsLi5uP1NFwPlPNZKfC+vPy+XeT+Ry7pONLC53acvGFzkeesfx+hdyWfSFXB5Z5rj19eay04YbNx2Xw4dHdOyzn33PsS3htHRubm5nVkVEpFfL10OhUmScqLds2XLdhg0bdj/22GOHV1dX5wDsvffe5SUlJQ39+/d3gwYNqp8+ffq/Zs2aFYn7kLbWOf78tuNHR+RQnG/MHG2cvI9xd1n6fYw7FzfynRk5jB5o7DHQ+M5BOdyxuLns+Qfk8NE9c+ifl3b2FC+933KHQD3DiIi0TqcHU2WcqGtqaiaVlpZWzps3b+dd6ePHjz//qKOOGlRTU2OHHXbYtE2bNo1fu3bt410b6q55e5N/ItjEIbZz3LThxuLy9OUXl8O03TpWtj3bdzhWVbUcV1JSsmsLExHpA0aM6GBzZR/SweNCb+rUqYvKysqmOOcoKChw48ePX1ZWVrbPk08++UBSUVddXT2u68LcddU7oCSpJaW0AKrq0jd9V++A0vyWZat3+HPXZpZ2ntYsKocBeVBZ1zzOP5lMz7IVEUln48aNFBcXhx1GpGR0RL1o0aKp06dPf2bs2LEba2trraysbJ+maVOmTHmjoKDAPf7442+Ul5eXjhkz5qr2lldVVdXp18fcX4/9NP1w2D21FPfziTJx3g2VtQzMt7TLTC5fWQvF/ZqTdGL5mpqaNmPrl0OLC9GcczQ0NHTZuuu1Xuu1Xve214mPWY5CPN3xOlOWmEg6YsaMGQsqKiqmLF++fGjytNmzZ1ttbe0p5eXlF+22227fefzxx19qZ3GZffgu2FrnGHxjA4u/mMuEwT7ZnvVYA6OK4cdH5aaUP/y+er64fw5fner3YW77byPxRY28dHrLxoeZv6/nK1NyOHv/1vd16hocxb9sYEdwirswD66b2cg3ZuhiCRGRdGpra/vC3TEZNc926e1ZjzzyiHviiSceKi4ufuWNN954siuXvauK8o1PTzB+sLCRrXWOhWscf33Hcebk9Kt+1uQcfv5qI2uqHGurHT97tZGz92suW9fgqKl3OAc7GqGm3tHYys5Ofq6xV2nLcbpQQkSkdbqPOlW33EftnOu3efPmyNwMd9NxOWyvh91uamDOow3c/LEc9hvmd2ieW+0o/mXz/VPnTDNmjzem3NnA/nc08Im9jXOmNe/8HP/HBgpvaOCFtRB7opHCGxp49r3WGwaO2KPljlMf2FMUEdllOj+dKqOLydI54YQTTqypqTmquLj4hpycnPKamprPLVmy5KtjxoxZ0QXxdYkhhcZDn0xt5gY4crRR/b/N1WBmXH90LtcfnX5ZC07LrMqO3MP4w1vNnXLk6D5qEZFW5eV1Oi31Op2uEedczrvvvnt+RUXF9xobGykqKtoxduzY10aNGjWrKwLs6WaMtBbNFtu3bQMicYu5iEjkbN68maFDUy6B6tMyvpisi4X64dlQ3+gouqGBukZ/Mdncw3ZwwSG6PUtEJJ2tW7f2hceIhncxmaTKyzEmJHSYVZ1w64GIiLRUXr6LT5jqxZSos2BmwgVlrrFHPB5dRCQUjdpGplCizoIj9jCK+/nXJaWlbRcWEenDRo4cGXYIkaNEnQUzRjYfUW/ZvDm8QEREIm7t2rVhhxA5ug4+CyYOZufTyfr37x9uMCIiEaaOi1LpiDoLcnOMDw2Bep16ERGRDClRZ8mRo40G17IjDxERaamysjLsECJHiTpLDh/lz1OXDhoUbiAiIhE2atSosEOIHCXqLGm6oKxyy5aQIxERia5169aFHULkKFFnyfhB/u/GOl2/JyLSmpwcpaVkqpEsyTF/RL26rtc/Gk9EZJcNGzYs7BAiR4k6ix7+ZA7njFOzjohIa9T0nUrtsFk0e58cKirU16qISGsG6YLbFDqizrL6+vqwQxARiSxtI1MpUWdZdXV12CGIiESWtpGp1B91ltXW1lJQUBB2GCIikdRHtpHqjzrKVq9eHXYIIiKRpW1kKiXqLOvXr1/YIYiIRJa2kamUqLNsyJAhYYcgIhJZ2kamUqLOsvXr14cdgohIZGkbmSrU+6jNbAeQG2YMHWB07UVvXb28KOnN69aVVE8dp7rqGNVTx0Whruqcc/kdLRz2Vd+RZ2avOudmdOHytjrneuVzRLu6rnor1VPHqa46RvXUcT2xrtT0LSIiEmFK1CIiIhGmRN2+eBcv78EuXl6UdHVd9Vaqp45TXXWM6qnjelxd6Ry1iIhIhOmIWkREJMKUqEVERCKszydqM8s1s5+Y2QYzqzKzP5vZsDbKf9zMFpvZdjN7w8yOz2a8YcmknszsJDN72szKzWyTmT1nZkdmO+awZPqdSpjvXDNzZnZZNuIM2y789nYzszvNrMLMKs3sP2Y2Kpsxh2UX6upCM1sWlF1qZudlM96wmNlpwfam0sza7S/TzGaY2Stmti2orzOyEWem+nyiBr4HnAIcAowOxt2drqCZ7Y2/GOxaoDT4+xczG9f9YYauw/UEDAZ+DewDDAfuA+ab2ZjuDjIiMqkrAMxsT+A7wH+7N7RIyeS31x94CqgDJgGDgNOBvtInYiZ1dTJwFXC6c24gcBbwEzP7WDYCDdkm4CbggvYKmlkpMB/4M36b9TXgFjM7rDsD3CXOuT49ACuBLye8H49/as2eacpeBTyXNO454Iqw1yNK9dTK/OuAT4e9HlGtK+DvwKnAAuCysNchavUEnAO8B/QLO+4eUFffBl5IGvcicGHY65HF+joGqG+nzBeDerWEcXcDt4cdf/LQp4+ozWwQMBb4V9M459wyoBKYlmaWaYllA/9upWyvsQv1lDz/FGAYfeBocVfqyszOAbY65x7IRoxRsAv1dCywFLgjaPpeYmbfykasYduFurofKDGzI8wsJzjtNBF4PAvh9iTTgNdckKEDkdyeh/qs7wgYGPzdkjR+M1DSSvl0Zffr0qiiJ9N62snMdsM3Lf3UObe060OLnIzqyszGApcBh3ZvWJGT6XdqGD5ZX4A/EpoKPG5mHzjn7u2mGKMi07r6APgT8A+aT29e4Jx7o1ui67la2563uU0LQ58+ogaqgr+lSeMH4fdW05XvaNneJNN6AiC40OcfwBPAJd0SWfRkWle/Ba52zq3pzqAiaFd+e2ucc790ztU5514F7sGft+3tMq2ry4HPA9OBfvgjxG+Z2Ze7Kb6eqsdsz/t0onbObQZWAR9uGhdcMFYCLEozy+uJZQMHBON7rV2oJ4IL7J4D5jvnvp7UvNRr7UJdfQyYG1whXw4cAVxiZs9lIdzQ7EI9/Yf0PR71+u/VLtTVgcBfnHNlzlsMPATM7v5oe5TX8TsziaK5PQ/7JHnYA/B94C1gL/wX/4/A462UHQ9sA+bg91TnAFuBcWGvR8Tq6UPAavyRYuixR7yuRicNLwLXAyPCXo+I1dOewW/vfHzXuNOADcCpYa9HBOvqkqDshOD9vsAy4PKw1yML9ZQL9AeOB+qD1/1JuGAsoeyg4Dt0EZAPfBR/F8FhYa9HSqxhBxD2EPxjfwqU45tCHgSGBdNOB6qTyn8cWAxsD/4eH/Y6RK2egNvxRzrVScPpYa9H1OoqzbwL6DtXfWf62zsGeA2/c7wUOD/sdYhiXeGvPfoxsCL43a0CfkYfuGIeODvY9iQP44Ajg/oYm1D+IOCVYHv+LnBG2OuQbtCzvkVERCKsT5+jFhERiTolahERkQhTohYREYkwJWoREZEIU6IWERGJMCXqbmZmQ8zsL2a21cxWmtnnd7W8mS0wsxozqw6Gt5LmvcfM3g+6eHvbzL7SXeslIiLZ0def9Z0Nv8F3zTcC/xSceWb2uvNPC9qV8l93zv22lXmvxfewU2tmHwIWmNlrzrnkjkRSxOPxIcDv8A8KKAcuicVi9+1K+Xg8vgD/7Oqm/mDXxGKxSQnz3oN/uEARvlet62OxWGvrJCLSp+mIuhuZWRHwGfwTgaqdc88DDwNndkX5ZM65xc652qa3wTC+g+Em7iCcDtwcj8fb6mykvfJfj8VixcEwKWnea4FxsVisBDgZuDoejx/YwThFRPoUJeoMmdmjZra5leHRpOIT8X2ivp0w7nVa722rI+WvDZ4LvdDMjkkT301mtg1YArwPPNbeOsXj8Z07CLFYrDoWi7W5g5Bp+WSxWGxxLBbb1R0KEZE+RU3fGXLOzcqgeDGpPbFsobnbukzLXwyU4Y9kTwMeMbPpzvdN2xTfeWb2jZ/85CfPFhYWfjgvL291PB5P/pznY7FY4npMBOpjsVjyDsLRrcTZkfLXxuPxH+OfOfz9WCy2IHEB8Xj8Jvzj/grxj4Vsd4dCRKQvUqLuXtWk9m1aQnO3dRmVd869nDD+TjObA5wE/DpxBudcA3CEmd0ClDnnftVOnN2+QxGPx6fHYrGdOxSxWOy8eDz+DeAw/DOcaxERkRRK1Bkys/n4h7un85xz7sSE928DeWY2wTm3NBg3Dd+ZRzqZlneAtRFuHh1rUu7SHYpYLNZihyIej6fdoYjFYg3A8/F4/AzgXKC9HQoRkT5HiTpDSYm4vbJbzexB4IfBrVLT8R3dH55peTMbBBwCPIO/mvpU4CjgfwHMbDfgI8CjwPbrrrvupV/96lcz8vLyauLxeHKH8c/FYrGUHYp4PD4hFot1eIcig/JdtUMhItLnqPesbmZmQ4DbgI8BFcD3nHP3JUyfjz8Sn9tWeTMbjj+P+yGgAX+x2OXOuSeD+YYDf8InzBxgJfAr59z/dSTOeDx+Pz6hNu0gPAYcHovF0ibf1soDa0jdoYgDB8Risbfj8XiLHQrgOHyXfXNisdjDHYlVRKQv0RF1N3PObQQ+2cb0E5Pepy3vnNuA7zu1teVsoPWLvzriPPwOwgf4HYRzE5N0PB6fjz8Sn9tW+Xg8Phy4mpY7FJ9MuPDM4Zu5b6F5h+ICJWkRkfR0RC0iIhJhuo9aREQkwpSoRUREIkyJWkREJMKUqEVERCJMiVpERCTClKhFREQiTIlaREQkwpSoRUREIkyJWkREJML+Hw9WSzjD6dEbAAAAAElFTkSuQmCC\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-02-11T16:04:56.909876\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ @@ -256,7 +9931,7 @@ "source": [ "## Detecting Data or Target Distribution Shift\n", "\n", - "Let's assume that there is a shift between train and test data:" + "Let's assume that there is a shift between the train and test data:" ] }, { @@ -282,8 +9957,65 @@ "metadata": {}, "outputs": [ { - "output_type": "execute_result", "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
mean_abs_shap_value_testmean_abs_shap_value_trainmean_shap_value_testmean_shap_value_train
f30.0413110.024811-0.039222-0.016373
f20.0075570.0090720.0066050.009072
f10.0000000.3298130.0000000.329813
f40.0000000.1269200.0000000.126920
\n", + "
" + ], "text/plain": [ " mean_abs_shap_value_test mean_abs_shap_value_train mean_shap_value_test \\\n", "f3 0.041311 0.024811 -0.039222 \n", @@ -296,11 +10028,11 @@ "f2 0.009072 \n", "f1 0.329813 \n", "f4 0.126920 " - ], - "text/html": "
\n\n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n
mean_abs_shap_value_testmean_abs_shap_value_trainmean_shap_value_testmean_shap_value_train
f30.0413110.024811-0.039222-0.016373
f20.0075570.0090720.0066050.009072
f10.0000000.3298130.0000000.329813
f40.0000000.1269200.0000000.126920
\n
" + ] }, + "execution_count": 8, "metadata": {}, - "execution_count": 8 + "output_type": "execute_result" } ], "source": [ @@ -324,26 +10056,5000 @@ "metadata": {}, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-02-11T16:04:59.045097\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "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\n" + "image/png": "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\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-02-11T16:04:59.045097\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, "metadata": { "needs_background": "light" - } + }, + "output_type": "display_data" }, { - "output_type": "display_data", "data": { - "text/plain": "
", - "image/svg+xml": "\n\n\n\n \n \n \n \n 2021-02-11T16:04:59.520565\n image/svg+xml\n \n \n Matplotlib v3.3.3, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAd4AAAEUCAYAAACF7VcRAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAABMlElEQVR4nO3dd5wkVdXw8d+p7p4cdjbnXZacFZosLApKFhEfECQLioqiD4KKgeCLGXNCQcngg6K4ZAEBEQmFZBZ2Wdic4+SZ7qrz/nFrdnt6Z3Z6Zme6e3rP9/Pp3eoKt25V9/Spe+vWvaKqGGOMMSY/vEJnwBhjjNmWWOA1xhhj8sgCrzHGGJNHFniNMcaYPLLAa4wxxuSRBV5jjDEmjyzwGlOEROQcEXlqENM7SUQWiUiziLx3sNLdGiJyuYhcX+h8GJNvFni3QSLyPhF5WkQ2iMhaEfm3iOwXLevxB19E5ovIkVnzDhcRFZGvZM2fHs1vjl7zReSrW8jPJ0XkTRFpEpEVInK/iNQO1vGWOhF5XETO72O1HwEXqWqNqr6Yp31ukap+R1W3Ko3+6ul7bEy+WeDdxohIHXAv8AtgJDAJuAroGEByZwNrgbN6WT5CVWuA04BvicjRPeRnJvAd4DRVrQV2Bf40gLwUlIjEC52HPkwDXh/IhiISG8A2xX4+jCkYC7zbnp0AVPUOVQ1UtU1VH1bVV/qTiIhUAx8DPgfsKCLJ3tZV1f/gfvT36GHxfsB/ukphqrpWVW9S1aZoP91KVtkl8qhk/VkRmRuVmL8tIttHJfpGEfk/ESmL1j1cRBaLyGUislJElonIR0TkWBGZE5X+L89Ie38R+Y+IrI/W/WVXWhn7/pyIzAXmisivROTarPP0dxH5Ui/nUEXkCyLyjoisFpEfikiPf5MicrCIPB/VUjwvIgdH868BDgV+GdUu/DJru3IRaQZiwMsiMi+av2t0bteLyOsi8uGMbW4Ukd9ENQ8twPuz0uxxn9nnI5r3s6iKu1FEXhCRQzPSuVJEbo2mu2pJzhaRhdH5+HpP5yJa/1gReSP6zJeIyJczlh0vIi9Fx/a0iOwVzb8FmArMivJ9WW/pGzOkVNVe29ALqAPWADcBxwANWcvPAZ7qYbv5wJEZ788EluF+0GcBv8hYNh1QIA4IcAjQChzRQ7qHAm24UvchQHnW8seB83vLX7Sfe6Lj2h1Xcn8UmAHUA28AZ0frHg6kgW8BCeACYBVwO1Abbd8GbBetvy9wYHQc04HZwBez9v0PXM1BJbA/sBTwouWjo+Me18tnocA/o+2nAnO6jjXzOKPl66JzHsfVIKwDRvV0jrawrx2i6QTwNnA5UAZ8AGgCdo6W3whsiD4PD6joIb3N9pl9PqJ5ZwCjonxfAizvSg+4Erg16zvz++hc7h19lrv2cjzLgEOj6QZgn2j6vcBK4ADcd/Ns3He3vKfvsb3sVYiXlXi3MaraCLyPTT9yq6JS2biM1Q6MSgsbX7jAkOls4E+qGuAC18dFJJG1zmpcVfT1wFdV9dEe8vMv4KPAPsB9wBoR+XE/qzd/oKqNqvo68BrwsKq+o6obgAdwP8ZdUsA1qpoC7sQFx5+palO0/Ru4H31U9QVVfUZV06o6H7gOmJm17++qK6W3qepzuIB1RLTs48DjqrpiC3n/frT9QuCnuKCa7ThgrqreEuXlDuBN4ISczs7mDgRqgO+paqeqPoa7/ZC573tU9d+qGqpqez/S3ng+AFT1VlVdE+X7WqAc2HkL218VncuXgZeJPosepIDdRKROVdep6n+j+Z8CrlPVZ9XV6NyEC+AH9uMYjBlSFni3Qao6W1XPUdXJuOrfibgf/S7PqOqIzBewsGuhiEzBVT/eFs26B6jABYhMo1W1QVV3VdWfbyE/D6jqCbiS0om40l5/Gt1kBra2Ht7XZLxfE10sdC3rafsaABHZSUTuFZHlItKIuxc9Omvfi7Le34Qr5RH9f0sfec/cfgHus8g2MVpG1rqT+ki7NxOBRaoabiG97OPKVbftROTLIjI7qiJfj6uFyD6HmZZnTLfS/bPLdDJwLLBARJ4QkYOi+dOAS7IuGqfQ83k1piAs8G7jVPVNXNViT/dfe3Mm7rszS0SWA+/gAu/ZW5mXMCoVP5aRnxagKmO18Vuzj376Da5kuaOq1uGqZiVrnezhvW4FThSRvXENxf7Wxz6mZExPxVVVZ1uKCyhkrbuklzz0ZSkwJet+cmZ6uaTZ2/KN86P7uZcBp+BuaYzA1Qhkn8N+U9XnVfVEYCzuHP9ftGgRrkYj88KxKqol2FK+jckbC7zbGBHZRUQuEZHJ0fspuCrGZ/qRzNm4e7LvyXidDBwrIqP6mZ8TReTjItIgzv646tyu/LwEfFREqkRkB+CT/Ul/K9UCjUCziOwCfKavDVR1MfA8rqT7l64q1y24NDr2KcDF9Nyi+35gJxE5XUTiInIqsBuuehhciX1GTkfkPIsrTV4mIgkRORxXbX1nP9LIZZ+1uHvqq4C4iHwLdy9+q4hImYh8QkTqo1sGjUBX6f33wIUickD0faoWkeNk0+Np/T1Xxgw6C7zbniZcw5Nnoxarz+Dui16Sy8YiciCu9PUrVV2e8fo7rsFOT/cot2QdrpHTXNwP6K3AD1W1qxr7J0An7gfzJjZVb+fDl4HTcefs9+T+mNNNwJ70Xc0Mrpr+BdwFxn3ADdkrqOoa4HjcZ7QGV4o8XlVXR6v8DPiYiKwTkV6r9DPS68QF2mNw9+F/DZwV1X7kKpd9PgQ8iGs0tgBoZ+BV2NnOBOZHtwAuBD4BoKo+7vv0S9x3623crYsu3wW+EVVDfxljCkBUrebFmMEkIofhLiCm6Rb+wEREcdXYb+ctc8aYgrMSrzGDKGrZfTFw/ZaCrjFm22WB15hBIiK7AuuBCXRvJW6MGULiugLdI2ueL67TnKujdhF9pXGliPxo6HK5iXXrZswgUdXZQHU/1t/q1r3GmC1T1W8VOg/ZrMRrjDGmZEVdoF4UTdeLyF/EDcryqIjcnFXKnSSuq9Q3ReQ+EanqJdmtMhxLvHbfbAtmzZoFwAknDLRTI2OM6ZehqbmRj/b8W69397a/P4tIZi9rO/WwzreAdaq6i4iMxD1R8JeM5Ulc//EbcK3yP4F7omFQDcfAa4wxpuT1O55/TFVf27i1iN/DOu8HPg9uQBYR+VvW8odUdX20/bPA9v3NRC6sqtkYY0wRkl5eQyqzxBwwRIVTC7zGGGOKkNfLa6s8TjR+uIiMwPUNn3cWeI0xxmwrrgbGisibwF8BH3c/N6/sHq8xxpgilHu5UFWn9zAvGU0+njG7BThNVdtFpA54CvhdtP6VWdt3ez+YLPAaY4wpQkNyP7cBeCAa77sCuF1VHxmKHW2JBV5jjDFFaPDvhKrqSmDfQU+4nyzwGmOMKUKl27GbBV5jjDFFR3sJvKUQji3wGmOMKUKlEGJ7ZoHXGGNM0dFe7vGWQji2wGuMMaYIlUKI7ZkFXmOMMUWntxJvKbDAa4wxpuiU8jB0FniNMcYUHSvxGmOMMXll93iNMcaYvLESrzHGGJNHvXWgUQos8JaYVCDEvVJulmCM2RZY4DXDQuUVTbSnZ4IIlwXNfP+kmkJnyRhjBqh0A2/pVqJvYz51WxPtgUBaIVB+8HxY6CwZY8yAKV6Pr1JgJd4ScderaQhj7k2gEC/dq0VjTOmzqmZT9Co60xDzEAUVon+MMWZ4ssBril5MlbIggFBRoDJd6BwZY8zAlUq1ck8s8JaIDgRC15pZgJRXuleLxpjSZyVeU/RSCQ/t3NQOMIyV7tWiMQYaL3+Qzu8+Dggt1fVMXva/xGrLCp2tQVPKgdd+nUtEAKQrEoSeEMQ8C7zGlLBUc4rG7z7NSsawhpGUtzTz2p53Fjpbg0x6eQ1/VuItERWdIc1VHukKd8XrhfY4kTGlas7//JU0o6milQbWogjBgrWFztagCku4XFi6R7aNqQmVeLAp2FakLfAaU4q0sZVRDz2DR8BEllBNKzW0MIZVAIRrmglXNBY4l4NhGy7xJpPJkcAdwIHA277v7zvkuTL9FhAyoaUd9YSYKh2eXVMZU4qC7z9MK7VU0IaXMWptgoDVcglKAiHEK/eoeetyyqbVFTC3AxeWSJDtSS5VzRcCNcAo3/fTyWTyeuAgYGfgRt/3zx/KDJrctJTHqeqExZXlxEJlRns7QaA8+fB6mppSzJ/bTnWNx7mfn4g3SEE5nQoREcJAicUFL9b/P5SWxjQd7SEjx265UUhne0i8TPC2srV2Z3tAWUVsq9IYSkHK1VTEEnbhVGhhcydezZa/lx2Lm5Eyj7KxVWhHGjxBEpt/v9pXtEIAZWPKSa1qxyuPkRhVsXF5qjVNoipOkArZsKSV9LPvUPnl23iN8SwaN4UdDqhjrx++n3U/9Al//RJNUounCVqoppoWFPAQ2qknTRmglHW0sXSPPzD16bOIbWijbGw58RkNSNzlT4OQ1Ny1JGY0IGXF+DexbQfeGcBs3/e7ngx9BbgL+PSQ5cr0mxcoiyvdH3LgCWsSCS7/7Ns0rUsjgKfua/zF5+byoeMa+OesNYwck+CCr05h7MTynPbx2nON3PnLJQDUj/BYvqgTUGKh4gmcfNFk9jtyVJ/prFjYzs3XvMO6pR0br9crqjyu+NPePa7/q8+9wfJ32/A8OP1b27Pz/iM2W+e+X87nxYdWMWZqJadduRN1Y7r/YKY7A35x3is0rk4RTwif/OluzHl6HU/fuYTa0eWccvVOjJlW1T3NX8znhftXgioTt6/iE9/dlcq63v9knrt7Gf/84yIq6+LsdeRonr97GWVVMU76xk5M3bN7qWP2Yyt5+Nq5eDGP476+MzMOGMnd33yDOf9aA8DE3WpZs7QDgOMu3YGdDhnF7H+v5e8/eZd0R0i8vZNxUyv46Hd3p25sxWZ5GagV77Tw5yveonltikPPmMzBp00atLQLQeevJv3hX6NzVuB9Zibxn5wCQNOqDu7+ymusmd/Czgc30P7cMpoWthDWl9OeKGOPYB361mrmjhqHKIypi3P47TOp26We5V94nHXXvUo1TcQ7UwCEdeV4ja3UswFJeJSdtCPxvz8JExt4ZqcjeetNpb6xnYlrmwHoIE6nxIiPqeDtKXV0dirbr3+HmUv+TaXEIKhkedlUXp82laAljv9IC7VjvkZT+xQW1W0PDR4z1qxmGRMZyVrK6cTDo5wO0iQAIU0Zy6SK1fvcTWusHE9DJqdXs8Pdx5JuqGHtzJvwcM/9j3niLCoOm1agT6ln2+w93mQyOQs4Gzg7mUw2J5PJq3zf/7nv+w8BpXAToWSUByGebqp2qggCmtal8XABV7tenSH/uHs16ZSycmkn992+Mud9/OnXS2htDmhtSrugK4AIgSeowr03LM0pnQdvXsqaZZ0om+7adLSGzHlhw2brvvLEGpa/2wZAGMLff7lws3UWvNqIf+9KgpSyfF4rT96xZLN1nrhtKY2r3Y9kOqX89QfzePLmxaQ7lXVL23ns+kXd03ytCf++lai6xxqWzm3l2buX9XpMrRtSPPzbBaTaQxpXdvLU7UvpbAtpXpPioZ+/221dVeXBH8yhozmgbUOKh6+dC8Ccf63ZeD6WvtFEe2Oa9sY09//obQBm/fRd2psD0iml3Yuz8u0Wnr5p8/OxNR67bgHrl3WQ7gj55w0L2bCiY1DTz7fgqnvRV5dAR5rwp48SPus+i2duWcjKuc0EKWXhX+ezYV4TYSqE1W2ELSle7qhhzujxqAihJ6xd18mrV71E69NLWfuLl9DOgHhnJ13f4FhjO3U04hEiqTSp/3sdbe+Ed1Yw5fHHABi7oXXj51tOGk+V1YHQ2amgyuFLnqI8TFEZtFNOE6+PnkKqopwwESMoizOvfipt5UpHRRkpzyNOB6DECOn6KXd/6+53QBEaK6pojbkL61A8VnkjWHXOvaw++e6N1dQCrPn43fn6SAx9BF7f908AbgNu8n2/xvf9K/KTrd41NTXZdA/TMfE4cF0TozpTTGnrYLvWju4VNdGb7CZX6WBTF1d97auvih+JqoFzyXPPCchm6yeyqsC82ObrIN1zJt7m68TLuq/jZfVlHYTdz0NPx5pKdfZ4LBvPT28nyOtp/U0r97pdth5W7M85z2U6HQYZiW/aZbF8z/s9nX3Oou9Gzue827bQ2tq68W33JHofirOr99Zws31mz9j0XoFUPKN2RYQqbSNNDFSJdwasYiRrqWMFozbuvZUKNlBDgNAm5XTE45BxQS4oiAfZt4Uy/h76e56HSojX46sUDLujqK2ttekepr0woCoI2W99C3s3tdIYjxEr2/TxKhCIsN/MOo4+ZQzllR4Tp5Vz0tmTckof4OMXTaZ2RJzahgSTZkTVm6p4oRJPCKd+cUpO6Rx7zkTGTa1AMr59lbUxdtqnbrP1dz1oBNvvUwcCZRUep3xlu83WmbZHLQecNI6yKo9Ju1Rz2GkTN1vnsNMnMW67yo37OuWbO/L+T06hvCrGmOmVHPWZGd3Wn7pHLQd8ZBxezP12T92tmkNPm9ZtnczpqvoER180nYqaGA0TK5h51mQqa+PUjyvnmItndFu/rq6O4y7fmaoRCWrGlHHUpTsBsOsRYzbWTEzZu57qkQmqRyY4/is7AvCRS7ajuiFBolyolIDxu9Ry8NlTczrnuU4f/bntGTW1kvKqGEd+ehp1Y8sHNf18T8euOgHZdyrUVuBd9iG8/aZTW1vLgWdNY8LutSQqY2x3ygwadqknXhVDxlUTH1FOcjrsllpHTJSYKmPGVbDnle9l9JE7MPKSffBqEqRrKvBIu9e4SjZQT4CHVpRRdubeSF0V7DqZps98lDiwqq6KEAgQmhLlBDFhTIVQXZ8AER6bfChtsXKaE1U8P3Yfdl67aGPQ9MKA8U3rWF/TwA6NC4ing41dKq6nlsWMZyljWM5o0sRZVVHJm2PHMLaxhZp0BzENKA9TjNVGxtx1EmMf+jih5wEKMWHcY2cO+DwPndJt1SyqvV+pASSTyRuBdHYjqt7m54GN8t6D3T+3jMZYnMUV5cRU2bO5lRd/M4Fbr1tGS1PIOReNJ1HmbXXjpExhoIgHMqDig9PRFpDqCKkZkRi0fA1n2tXtp3X5OSxoELqaAc+LpqXXz07Tbt1gdRsysgIv7nX72wkDJdWWJkiFVDWUs/6+2cz+4gM0BWWM/tge7HHF+1j0l3d550v/wVsbUEkndbjbMAFCB+4iqYIOducJXuF91Jy6GxO+si+Vk6qQEeV4Zd3bKGgq6LExWD8NyZd1uVzR42/9eL1q2P9xWAcaJaKlzGNxwv3hBSIsrCinrS3kjE9PGLJ9DqQVc7byyhjllcXYorIwLOAOL5LRQ5z00VucxN3y+LjqHpd7MaG8ZtMF6IjjduWg43btts72Z+1Ipefx+pmPM5YNG7tVVCBOihjKWN5hIbsw5qAxTL7z6C3naeuD7pAp5S4j+x14k8lkGa6KOgZoMpmsAELf9zu3vKUZSuIJoopGV9AeyqKFHey0c2WBc2aMGUzjT9+O1859Ek1Lt9BUdthodnziHLQjxYiWFN7Iql7TGA5sdKLuHgZmZrw/B3gCOHwQ8mMGKN0ZsB0dLE/EKQ+VCg0ZP6F0Okw3xjie5xFkPYsvQMfuro2FlCeQ8uF/66aU7yn2GXh93z8n6/3hQ5UZM3Dlnse8mPtja/WEDjzq6oq3GskYM3AVkyvhnTRdP+Et8QTTTyqu53C3VimXeEv3yLYxHQqoUhOElIVKaIMkGFOy3vfiSbRPqKJSViDxRiYdVUb9B3codLYGlSI9vkqBNa4qES0oO3R20uh5jFClI2EfrTGlKlGXYM+lFxc6G0NqW++r2QwDsaoEi9uUMGpc1RBYidcYM3yVSum2JxZ4S0SbSreecTrssRRjzDBWyoHX7vGWioSQiL6nokpLCbRqNMZsu0r5Hq8F3hJx5p5xWmor6KxI0FFTTqooh/kyxpjcFGtfzSLyQRG5QURmRe+TIvKB/qRR+KMwg+I3p9ZSE6ShLA7i8bUDSuPK0BizbdJeXoUkIp8HfgPMBQ6LZrcB/69f6fTVV3MRGnYZzqe//X0WMYETTjih0FkxxmwbhuQqf45c2+Nv/U56ScFKFSIyDzhCVeeLyDpVbRCRGLBSVfsejDxijatKzCB0n2yMMQVXpI8T1QJdg3d3XRgkgH51mWxVzcYYY4pOkTauehL4ata8LwD/7E8iVuI1xhhTdIq0xPt5YJaIXADUishbQBNwfH8SscBrjDGm6BRB6XYzqrpMRPYD9gem4qqdn1PVfvVYZIG3hDS2hdw+dxrTalqwplXGmOGsSEu8qGuR/Gz0GhALvCWitSNgytfWE8pkRJUHv7Oely4fUehsGWPMgBRjiVdEFtHLkzWqOjXXdCzwloiTr2+muaqMMBEHVd5e01HoLBljzIAVaYn3jKz3E4CLgTv7k4gF3hLxn+W4oAsgQmuldRlpjBm+irHEq6pPZM8TkceBB4Gf5ZqOBd4SEXhARwr3LLuiNkiCMWYY0+HztGsHsF1/NrDAWyKCVAgSBxFAILQOvowxw1cx/oKJyNVZs6qAY4EH+pOOBd4SUQ20SUYp1wq8xphhrEjv8U7Jet8C/Bi4pT+JWOAtEVUaUh4GdMRioEqVDU5kjBnGijHwquq5g5GOBd4SEU+HHNTYwgbPo0wVPMF1K2ryafFF/yD89SMEEmfk/RdQf9T0QmfJmGGpWAJvrkP+qepjuaZpgbdEpBHeqixjfWU5XqiMb2kvdJa2OZ1rWhjxq9spI8DTgFVH/5R6/Wmhs2XMsFRErZpvyGEdBWbkmqAF3hLRBqyur97YmnnZsGkQWDpWfPOfVNHAWkYBSjUrC50lY4atYinxqmq/WiznIqfAm0wmRwJ3AAcCb/u+v+9gZ8RsnbDM6/YIURCzyJtJn3gDPfybAIS7Tyf+2rU5bxt2Bqz42wL0xaWMOHY6VYdO3ris8eO3EDy1gMrLZ/LiXat4D11DcgotjCbnATqNMd0UUYl30OVa4r0QqAFG+b6f7pqZTCb3AnzgSd/3jxyC/JkctSXilKUDOuOucZWosmJFJ1+6bEn0ZJESL/OYMRpqyoSjTmygujpOEEBbW8BLfgs77VrJwTPrmTe7hVf9Zt57UB3Tdqhk2aIOGtelGT85wewXWygrE3bYo5o//3IR77zWTEVdjOQHR7PjHjU0rk/z7N9XEHQEHHfBZBbcu4Tld7xD7dpWRh06jjcWp1mR8qjWkFoCRtfF2PN77+Wh788hWNLM9tuVs2AtVI8u4/Czp9Cw7yjmv9rM8rlNbPjnMhpXtBPboZ5jv7YznStbefZ7r9MRCLH17QStAbsfPopYfRkrfvMyGxLl1E2pxHttOWNaFzA9Kol6r6/moYm/YtxnDqLhfRMIEnESi9ez4sE5jHjtLZbXjmfdcg9WdVLZ3oGnIal4gnTMY/QvXmBaeikxArwOCChncflYGi97k+YRdaQ8j0To+ksPsRZuxgxUsZR4M4lIHXAlMBMYTcbzI/3pMlJcf89blkwmrwfwff/8jHlx4BlgA6B5DLzF+HhXwdVe2UhzGKO8MyAUIVUZ57S3lxN4UclXlYp0ikSo7llfVRKhEgKBJxvn7bZrBe+81rIx3aM+MpJH/rYWVfA80HSIqBLXAE9lY9opT0jH4ogq1Z0pPFUkVHZ5+V2qO9wY0YnWgLL2gPaKOM/O3IVUeQJUGbNiHZ015QDEUumoYRjEO9MEDZW0Bh7bv7GEUaubAVg7uoZl00cSb2mndUQtI1auZ9XEUSBCRWsHH3j2FUZ0tAMh5bjrxBjtTOI1PEJCYqzyJjBru5m0NNQSJOKIhpz4xr282rAby6rHM25lE+NXu/NQQys1tLO0qp7VZbXMbHqczmA8XX9z66llZUU9C0Y3UNPZzrTGNYR4VLd3sLNeNnQfujHFYUgi5D/ljz3+1r9fzy1YRBaRW4HJwE+AW3FdSF4K/EVVf5JrOn3WRyaTyVnA2cDZyWSyOZlMXhUt+hrwPPCvfubdDAFJh9Q1tlPRlqKqtZOKls7uVygixLuCbvQ+JPqLyZj31pvdG2U9/cgGuq7NwoyBryRzEKyNaUM8VLxoA/WE1RNGbFwtVeGhQFtNhQu6XfnwNpUMM6vL02VxWtPugmBkFHQB6ta1AtBRXUEsHdJcX7XxGNqrymmvdkE8zqZMNrAYL3rvEVAfNpLwQoKom00Vj5fH7cmy6vGgytjVmy4+WqgAYFxrE1PXr2eZziDzt0YkZFldHfF0SHNZBa+PnkSjVNEWbWeM6b8Q6fFVYB8CTlbVe4Ag+v9U4Mz+JNJn4PV9/wTgNuAm3/drfN+/IplM7gmcA3yl39neSk1NTTbdw3QiHSIKobgqgcr2NLEwZGPU1Iyg2/W+h+mKiu5f7DETynreJpMqobgdh550C/gVrZ0bp8XFZqqb2omlg43zq5t6aYHdlX8RWmo2BbH26jJQJRaEhDGhrD21aR9hSHWbGyAizPh6x0iTKU2cMJRuxzSyfS2iIYiQSsQytnUBO8BDgNawBsHlX4FO3DnyFMpSIfF0SFNVOUHG/ovle2LTNj3Y00NFkR5fBebhankBmkWkHlgG7NCfRHKtar4RSPu+f35GFfOVvu/fm0wmrwTeZ1XNhTXt0jWsjCVoL09AqIxrbuM3h8Nf/rQW9TwCEVcNHAbEY8LUKWWMGuVKe0uWdrJ6dZq6+hjf+t5U7rllJXPfaGX399ZywmljePCu1axbk2L02DivPtNIokzY//313Pv7ZWhUXV09royJ0ytZuqCTlhXtJIKQ7XasoPbZRZS/uRovHRLEYqQScdaPrqG1soxYSqls6ySx70iWLU8jqtQ1tbKhrpqydMB7pnuM/Mh2zH65hdVvbqDm5RWgyprx9cy8aAbN/nJefLaFsmZXuk8nYmy3fC1jmxqpaW6hvSzB0tH1VHZ2csCKZ2gIVkQ9Wcd4uu5gFo2dQsXBE0nHY8QWrWfKyy8QS6d5q2EG2hFn0oomvDCk3OukJtVJB2WAkPZSzAjnkgA6qWQeM1hXVcWG2koAgpgQI2Dfpe+yvX69MF8IY/JnSKLhQ7Gbe/ytPyo4q5BVzY8C31HVR0XkDiAEmoF9VTWZazoDeZzoMmCu7/v3DmBbM0RCT13QBfCE5soyTjqxlvfPrGfRog522aWSWEzwchg84bRPT+z2/sNnjN04fdxp4zZOH3rsmBxytku3dx2tAenOkOoRuY+etM/JvSw4ZTJH5LB9Z0uaNYevRP1GQGiPJThwxUUcUpGdh6MAeG/0ru25ZZAQ4tuP5LnT/on3r8VU7TmSfZ88EREI5i6jYu9vsnvHSyxkCq+H29FZ7qq5x7auZzRrcz5GY0x3WvDCbY8uYNOFxsXAd4ARwFn9SWQggfdDwD7JZHJ19L4KiEfvd/J9335tCqA95nWrTk5H14ojRsQZMaJ4Htcur4pRnuf+LMuq40x4/lLC/8wjfG0JFWcfjJT1fU4q95+wcfqQ+47abHls50nE2v9AAtgVqDnlLjbc9zYeyqTW1daq2ZitEMaLMvIuUNUAQFVXAuf3sX6PBvKL/D9Aecb7/wX2A04D1g8kE2brqQiVrZ10VCTwQqU8476ncbyDtoeDth+y9OsvPoD2u+YwhnWExFnGKBqGbG/GlDaNFWXgXS4idwG3q+pTA02k34HX9/1Vme+TyWQj0OH7/uKBZsJsvTDmEQ8DyqKGSh215X1sYQZb3SFTWd5Qw8J1FXgodccNeoc3xmwzguIs8X4IV8i8XUQC4E5cEH61P4nk1LiqyAy7DOfDqK+uobGmGglDQk+ob0ux5pq6Qmdrm5Ra106sIo5XWTxV/MYMoSGJkPc03N7jb/2J604viogsIjNxQfhkYJmq7pXrtvbLUCJEYfrKJtKJGGWBsjaHRlRmaCQa7PldY7ZWWJxVzZneBGYDC4Ed+7OhBd4SURXC2yNrIPqyVrSm+9jCGGOKV1CEgVdERuBKuKfjxi54GPg+8Pf+pGOBt0Ssq4hBfFOHDR1WzWmMGcaK9HGipcDTwO24HqzWDyQR+3UuEVoR63b3u0hbBBpjTE6C4rxdtr2qLtvaRGzsuBIxttzbWM2MgD1CaowZztSTHl8FzdMgBF2wwFsybjopDhpCmQdxoawj6HsjY4wpUqEnPb5KgQXeEnHo9uV8ba+QurZ2xra2sOhL9hyvMWb4CqXnVymwe7wl5DvHV3GQzgJgbN0JBc6NMcYMXKmUbntiJV5jjDFFJxTp8VVI4lwgIo+JyCvRvMNE5JT+pGOB1xhjTNEp0nu8VwOfBH4HTI3mLaafY9NbVbMxxpiiowUu3fbiHOC9qrpaRH4TzXsXmNGfRCzwGmOMKTpF2mVkDDfwPWzqOaEmY15OrKq5hPz3az4rL0qz5gsddGzoLHR2jDFmwFSkx1eBPQD8WETKwd3zBb4NzOpPIhZ4S8Syx5Yw9w/zGNnSTn1Tipv3vr/QWTLGmAHTmNfjq8C+BIwHNgD1uJLuNOwe77bpb5e8zPgg3Pi+tsNKvMaY4avQvVRlE5EY8DHcAAl1uIC7SFWX9zctC7wloiUNd35gHzqrKqhpbed9L75V6CwZY8yAhV5x9XurqoGI/FhV/wC0AysHmlbBy+1mcLwzYQzx8gS1QYCUJ3hx9+0KnSVjjBmwYuyrGZglIlvdO5GVeEuEet3HRQji9tEaY4avIgiyPakA/iwi/wEWkTEmnKqelWsi9utcIvZ6ayH/GjOSOIKqMmZ9Y6GzZIwxA1YELZh78lr02ioWeEtEg6bZY8lyWsvKiIcBZUG60FkyxpgBCwvfgnkzqnrVYKRjgbdEjGvawLs6ierOqDWzbnl9Y4wpZsVY4hWRD/S2TFUfyzUdC7wlIkTYfcEi3po8kUQ6YEl1baGzZIwxAxZ6xVfiBW7Iej8GKMP115xzt5E5Bd5kMjkSuAM4EHjb9/19c92ByY95DSOpiMHoVeuYM2kCk1atL3SWjDFmwIqxxKuq3R4XiZ7t/QbQ1J90ci3xXojrj3IUUJdMJp8EdsG18FoF/BG4xvf9baKCc966gLKYMKVu8yuyls4QAarKNl8WquL148u00y/amLtOAOXfZ8c4eFpZr+s2E2PRmDHEA4WUIkHQe8JhSPCj+5Cpo/E+fhCkA3hlPuwyCaoqcs6fMcYMlWK8x5sterb3GlyJ98e5bpdr4J0BzPZ9P51MJluAzwBzfN9PJZPJ7YD7cQ8T/66f+c6bZc3KT14IKfPg30uU11bD0dvBjcfEiHlCY3vIh/+a5rkl0JlWAhXo6qQ7HbonntPRdUUi+kIoVBMQ94Tt65T/LgmgLDql7Wm3vRe9JPofiKfT1MVgWm1IDNhuVIy/zxdSAexel2Z+q0eTJIDYxk/okD8G0LwecIEYBcIQUaVSQz6441ROfXEeICRSaZaNq2elfANBUSBGGwlCdFID1UveJkYKCNDTAjIvBdSLgwrp5O6w747Q1IY3ey7eu0uQ1na30gE7wV1fhg9eBbMXw66T4eWfDMXHZozZRhVjibcXHwTCPtfK0GfgTSaTs4Cjo+mPA9f6vn9F1mohsHN/dpxvH7wr4PU13efdOhsq4gG/PyrOoXcEvLJMXbAMBMqiYAku0IpATCHUTfNRWsIYtAX8d20aasug6yrNE+gIXYD0BLruV6QC0ilYm4K1bQIdafzVHiRiIPBqYwISsjG+bhT3XNpd3UIKEPPQICSFx6rqSl7ddRoLp4whng744PMv4UUJlNFOgjQK6JKVxEghKLB5qVhC1xo6/vwrdDy/FCEkzupuwZkn34DpF0Jb1JDrlQVw5BXwyKA0+DPGmKK8xysi3Z7dBapwNb+f6086fR6Z7/snALcBN/m+X9MVdJPJ5L3JZLINeAeoBa7rz44Hqqmpqd/T7WndLOh2+fcSt878DUSnMwp6mVdbXdM9VaRnRqTMB769jG1CQKONw4xEJKM0nL2vLe0nkycIsKGuioVTxgCQjsd4aaeuWxFKgtTGJDw2XTRsiRAiBNGrB23d+4IOX3p34/RAPiObtmmbHp7TQyX0vB5fBXYGcGbG62hgoqre1J9ERLXv27LJZPJGIO37/vlZ82PAfsAJuJLw2v7sfIAGdB/5w38NmDVv801/NBMu2S/OGfemue3V0EWntEKZ54Ji1/mRaLozhHhXFbRCGIK6kiuJGFRElQgd6U0FyjLPbeuJq7ZORaVWVbdeRWJTFXXXenHpfqSpAFo6uwdugDAkESg7rm1kvyBNRRAQitCwrpEvP/w3BKWSlqiEC0pIGa3EaQHSG+dnU2J0MAqAclZuHnxnjIN3Vmx6/82PwdWn95iWMaakDUmd8P/74DM9/jh94x8HFqwOWkS+rKo/6mH+/6rqoN/j7ZHv+wHwTDKZPAz4FXDa1qQ3lP7yYY973laqEjB3Xci98+CTewkf38V1tHjr8XE+MDXkz7MDlrcoC9aFrO1UCKEqpqQ8DxRSKKRxQTEIOGlnj7bA44w9y7j4wTRrWlMg4IVKKEJNQqksU9o6lOZ2wBMOmwaJmHDYlBjr22LsMUH4ha+saVc+taeweF2K617BBWSifYVKTJRANl0MxMKQUZVQXwajl7ZRA6jnEVPFQ0jjSrjNVFJNq/vrOHU/9M9P0RnUAyEejSTY1NmG7jgFXduCfucMElVVEKTR5Wvh+TeQJWuhpR0+MRMu+TBc9zD84RE470j43DF5+iSNMduCsDi7jPwWsFngxbVszk/gzUpnx0FKa0gkYsLHdu76ID0u7uGBqPP28jhvr4FXZXxir95bHfflvP26v//t/8BVj3bw7WeU2jJY/tUKyuNVvW7/xSPfJjWqfuP7ltpKJug1vax9NtqRgngMyWo5KPR2+Xr85rM+d4wFXGPMkCimxlUZHWfEROT9dP+ZnMEQPU60UTKZPBCoBp4GOoFDgIsp4hbNw9UVR5RzxRG5rTtmXRPLGmrR6B5ILNhyIzspT2xt9owxZsgEUvD7uZm6Os6oAP6QMV+B5cDn+5PYQEq8ZcAPgJ2inS4Bfg58bwBpmUFSnkrR6rnW12XpAA371brdGGOKSjGVeLs6zhCRm/szClFvcgq8vu+fkzH9JGA9VxWZlXU1rEsph7wxn6bKclZNGVnoLBljzIAVQQvmzQxG0AXrq7lkNCUSnPXoi5SlXUl38domYO/CZsoYYwZIi6fAu5GI1AFXAjOB0WTc61XVqbmmU3yXFGZAJqzfsDHoAoxoaStgbowxZusEntfjq8B+DewDXA2MxN3bXQj0q+s+K/GWiPEtraytrWBkUzuhQKoyVugsGWPMgBXTPd4MHwJ2VdU1IhKo6j0i4gOz6EfwtcBbIhqaW1g/oo4mLyDwPNaPqi50lowxZsCC4nyO1wM2RNPNIlIPLAN26E8iFnhLRHssTntVGU0jqlGgurGl0FkyxpgB06HpEGtrvYy7v/so8C9c1XMzMKc/iVjgLRFhQzlVja2ueiZUvBy6AjXGmGJVBPdze3IBmxpUXQx8BxgB9Ku1c1Eemem/kx88klCERGeaslSa1hG993JljDHFTkV6fBU0T6rvqOq8aHqlqp6vqqeq6hv9SccCb4momVjLyX87nM7JFfCeci76r3XlaIwZvgJPenwVkjgXiMhjIvJKNO8wETmlP+lY4C0hI3dvYPxXoeEC+1iNMcNbGA1Omv0qsKuBT+K6SO56bncx8JX+JGL3eI0xxhSdQpdue3EO8F5VXS0iv4nmvYsbKCFnFniNMcYUnbA4n+ON4Voxw6YR02sy5uXE6iSNMcYUnWK8xwvcD/xYRMrB3fMFvo3rQCNnFniNMcYUHUV6fBXY/wITcJ1o1ONKutOwe7zbrpmXLmN+x0F4KLH6DRx7WH2hs2SMMQOSLqLneEVkvKouV9VG4CQRGYsLuItUdXl/0yueIzNb5Xd3r+Ulr4oVdRUsra3k3L8Ghc6SMcYMWCg9vwoku2eq36rq8wMJumAl3pLx46c6mdApLK6upCwIqAos8Bpjhq9AiqpcmB3yD9+axCzwloiyzoDXR4wg9IQW4tjYRMaY4ayApdueDGofvBZ4S8TI1nbCkZu+qR2xorpaNMaYfkkXV4k3LiLvZ1PJN/s9qvpYzokNcuZMgXhBSHk6oCMeA1VGt3YUOkvGGDNgQXGVeFcCf8h4vybrvdKPTjQs8JaIdRVljGnuoD0eI64hiSAsdJaMMWbA0oV/ZncjVZ0+mOlZ4C0RVaGSCEO8tHufsMZVxphhLFVcVc2DygJviahOp/jY0kUsqqmmKp0mLQCTCp0tY4wZkCJrXDWoSveSYhtT2Z6iMgzZqbGJya1t1KbS3Zavu3M2i896gPZ31xUoh8YYk7tOkR5fpSCnEm8ymRwJ3AEcCLzt+/6+Q5or028VqjQmEtSlUqRFSMU8/nvYH5BFq6lO1BGbu5YQj8W3vMSktz7NvDMepcVfRWJsOe9ddh5SIl9oY0xpKNJBEgZFrlXNF+JGYBgFzEgmk38GDgLqgIXAT3zfv35osmhy0VmWIEzBhkQCVNll5TLWvdJOu4xibON6hFoEIY3QsvPvaKEBiBOsCHix5tfs0/K5Qh+CMcZs1FHCgTfXquYZwGzf99NAA/BPYD9c4P008KNkMvnRocmiyUVNRycAqspJLzzFvq/NYd6ECSyZMIq3pk0iIEFInBhCSCVj2MBIGvEIibe2gHcypLtXT/PUbNjrS+7179nw01mw/Wfg6Kth5fr8H6QxZpsRiPT4KgWiuuUOOZLJ5Czg6OhtB3Ct7/tXZK1zJ7DS9/0vDEkuuxvUHkRKxSFnz2H/xg48oLKljb3emsP6ypqNy7dfuIa6thQeAQ00bXzqu40EE5nNCNbCvjPA/9GmRKdcAIvXuOlx9bBiw6ZlnzwCrrdSsjFmaIYM2vWiVT3+1s/+5ZhhH337LPH6vn8CcBtwk+/7NT0E3Srcvd+XhyaL3TU1Ndl0D9PlsRhxVWJhyB4L51Ofat24XDSkLOWe6/UIuv2V1LPOBV0gva65W5o0tW1asbmdbpraiubYbdqmbbpw00MlJT2/SkGfJV6AZDJ5I5D2ff/8rPkx4P+AscAHfN9PDUUms1iJtweHXfAuB63YgACxIODUZx5haXkDrfFypqxfRUNTZ1TdLEBIBSEeKcYzhzhtxAhh4XUwZcymRG98DD71Wzd93afh8dfh5sdh/Ah4+ArYc1r+D9QYU2yGJBxO/MLqHn/rl/589LAPvwN+jjeZTCZwJeEJwDF5CrqmFyKy8dsfxGI8tMf+nPfkAzRLJS/VzWAUC/FQEqRYTzlLaKCKJlbyXpQ4+3d+FhJZX4dzPgCnHuKmK8vh3CPgF+dDdTnEbBgGY8zQaS2R+7k9GVDgTSaTFcCfcS2dP+T7fnMfm5ghVtPagbLp0nPOuHE8PWUXghRMa15KAldV4BEyedYpdHzsH3R01BIjYO/V524edLtUlnd/X1c1dAdhjDGRDRZ4N0kmkzXALCCFK+m29bGJyYOYwKKaasa2tZHyPBrLyjlh4XmAa+m84tAbSM9eRdW1/0PD8duTbN++wDk2xpgtKN24O6AS78m4QYDbgFXJZLJr/q2+7184SPky/bS+opzqjpAl9XV46vpt7iIijH/q/C1sbYwxRWZbL/H6vn9OxvRNwE1DlSEzMHWtbXixcka2tJKKx0rmeTdjzDaqhH/DbJCEEqGqbLdkOQ3tHYTAghF1hc6SMcYMnAVeU+wSgTKmsZl4ZwoVoby8vO+NjDGmWJVu3LXAWyrWVJfTVlbG2nFjiKcDaputobkxZhizEq8pdrFYjGXjx4AIHUBTeaLQWTLGmIEr3bhrgbdUVGvY7QpxbY09b2uMGcZKuMSb6+hEpsi95331rIlKuSlPSGvYxxbGGFPEpJdXCbASb4m45pzR7Ppqmp1XNrKhPMGhx40udJaMMWbgSrjEa4G3hMy+djx/vucFEhJy4odPKHR2jDFm4CzwmuGi3LMqZmNMCSjduGuB1xhjTDEq3chrgdcYY0zxKeGmvxZ4jTHGFJ8SvsdbwtcUxhhjTPGxEq8xxpji41mJ1xhjjDGDwEq8xhhjik8J3+O1wGuMMab4lG7ctcBrjDGmCJVw4LV7vMYYY0weWYnXGGNM8bFWzcYYY4wZDFbiNcYYU3ysVbMxxhiTR6Ubdy3wGmOMKUIlHHjtHq8xxphhTUTmi8gehc5HrqzEa4wxpvhYidcYY4zJI5GeXzlvLmeJyKsi8oqI/FVExkbz/yMi+0XTvxaR16PpuIisFpHqITmeDMOuxCsiDwGjC52PYhaPx0en0+nVhc7HcGPnbWDsvA1MCZ23B1X16MFOVC+ND7jMG1U7fw/YV1WXici3gV8ApwKPAkcAzwPvA9pEZAIwHZitqi1bm/e+DLvAOxQfcKlJJpO+7/vJQudjuLHzNjB23gbGztuQej9wv6oui95fB7wcTT8KfF1EbgPWAE/gAvF2wGP5yJxVNRtjjNmWPA3sAxyHC8JdJeAjoukhZ4HXGGNMqfkncKyIjI/eXwD8A0BVO4D/Al8FHgGeAQ4B9oqmh9ywq2o2OfldoTMwTNl5Gxg7bwNj521wPSIi6Yz3XwP+ISIKvAN8OmPZo8B+wPOqGojI28C7qtqZj4yKquZjP8YYY4zBqpqNMcaYvLLAa4wxxuSR3eMdppLJ5E7ATcAoXJP4s3zfn5u1Tgz4OXA0oMD3fN+/Pt95LSY5nrdvAh8HAiAFXO77/kP5zmsxyeW8Zay7M/Ai8Gvf97+cv1wWn1zPWzKZPAX4Jq6/JgWO9H1/RT7zavLHSrzD12+BX/m+vxPwK9xzatk+AewA7AgcBFyZTCan5y2HxSmX8/YcsJ/v+3sB5wF/SiaTlXnMYzHK5bx1XexdB/wtf1kran2et2QymQSuBD7o+/4euE4dNuQzkya/LPAOQ8lkcizuObQ7oll3APskk8kxWaueCvze9/3Q9/1VuB/D/8lbRotMrufN9/2HfN9vjd6+giuFjMpbRotMP75v4B7RuBeYk6fsFa1+nLcvAT/yfX85gO/7G3zfb89fTk2+WeAdnqYAS3zfDwCi/5dG8zNNBRZkvF/YwzrbklzPW6azgHm+7y/OQ/6KVU7nLZlM7g0cBfwk7zksTrl+33YDZiSTySeTyeR/k8nkN5LJZAkPEWAs8BrTi2QyORP4NnBaofNS7JLJZAL3XOqFXYHG5CyG67zhg8BM4BjgzILmyAwpC7zD0yJgUnQ/reu+2sRofqaFwLSM91N7WGdbkut5I5lMHgTcCnzE9/238prL4pPLeZsAbA/cn0wm5wNfBC5IJpPbcicR/fk7/bPv+x2+7zcB9wD75zWnJq8s8A5Dvu+vBF5iU0nsNODF6D5uprtwP35edF/pI8Cf85XPYpPreUsmk/sBfwI+5vv+f/OaySKUy3nzfX+h7/ujfd+f7vv+dOCnuPYFn8pzdotGP/5Obwc+lEwmJao5OIJNHfqbEmSBd/i6EPh8MpmcA3w+ek8ymbw/aiUJcAuuq7S5uD5Ir/Z9/91CZLaI5HLefg1UAtclk8mXoteehclu0cjlvJnN5XLe7gRWAm/gAvXrwA35z6rJF+sy0hhjjMkjK/EaY4wxeWSB1xhjjMkjC7zGGGNMHlngNcYYY/LIAq8xxhiTRxZ4TV6IyHQRURGZPMT7uVBEbsl4/4CIXDaU+zQ9E5G3ReScHNfNy/cjH0SkPDr2XQqdF1OcLPAWGRGZISJ3ichyEWkWkUUi8lcRKYuWnyMib/ewXW/zPxH9oF3Rw7LHRaQj2s8GEXlRRE4emiMbeiJSDVyNG+kFAFU9RlV/ULBM9SH6bN5X6HxsC4biXIvI4SKSzpynqh3AD6OXMZuxwFt87geWATsDtbjh/B7CjZAzEJ8G1gKfFJFYD8u/rao1uNF37gD+JCI7DXBfhXYG8Kqqzit0Rsw27w7gAyKyQ6EzYoqPBd4iIiKjcAH3t6q6QZ3Fqvrb6Cq6v+ntChwKnI3rS/eY3tZV1TSux6YYsFkvTSLyORF5KWvediISiMj06P0foxJ6k4i8ISKnbyFvV4rII1nzHheRb2S830NEHhKRVSKyUES+KyKJLRzyR4B/9JZmRnXm2VH+WkTkfhFpEJHvicjKqKbhcxnbnxNVG35FRJZF61ybmY++jltE9hKRB6PjWNt13CLS1S3gw1Gtw/W9nKsqEflZtI/VIvI3EZmadYzXishfojzME5ETeztJGcf0JRFZHG3zIxEZFaXRKCJvZpYORSQuIt8SkXdEZJ2IPCoie2QsT4jIjzPO4Vd62O+hIvJUdA7micglIpLzBaWInCwiL0e1My+LyEnZx5S1/o1d57S3cy0i86Pjeiqa74vIfj2lkTFvvoicISITgQeAWLRts4icDaCqjcDzwIdzPT6z7bDAW0RUdQ2uu7jrReQsEdmtPz9MPfgU8Iqq3osrSX+6txXFVWV/DkjRcz+xtwO7iMh7MuadAzyuqvOj908B7wFG4Kp8bxSR3QaScREZCzwB3A1MwpX8Pwh8bQub7YPrdq8vJ+MGG58KTAeeBebhOrA/F/hpZmDDDTQxFZgR5eME4NKM5b0et4hMiI7jiWhf44HvAajq3tH2H1LVGlU9v5f8/gQ4MHpNA1YDs6R7DcbZwLVAPfBL4CYRqdrCOZgW5XdGdC4+jwsiPwQacOf9jxnrX4obIvHY6Bj+BfxDROqi5V8FjgcOBraLjnXjAB3R+bg/Sn8McBxwETmOwiMiBwO3RfsZBVwO3CEiB+SyfR/n+kLgYmAkri/z+zOOa0tpLsVdzAZRmjWqelPGKq/ivpPGdGOBt/gcDjyOG93lJWCFiHwzKwBvJyLrM1+40upGIlKB+6Hs+vG8AThGNm+88vVo+8XAicDJqrrZvWJVXYcbNeXcKH3B/dj/IWOdG1R1jaoGqnonbhD5w/t5/F3OAl5W1etUtVNVlwDfjeb3pgFozCHtb6vq2uhC514gpaq/V9W0qj4ArAPem7F+CFyqqm1RNfYPcBcdQJ/HfSbwtqp+V1VbomPpVtLfEhHxcOf5G6q6RFVbcN+NXek+gs2fVPVpVQ1xw/PVAztuIek24KooPy/jLraeV9VnVDXAjcy0g4jUR+ufC3xfVd+Mal+uBgJcAAX3uXxfVd9W1Tbgy0Bmf7SfBe5S1Xui8/Qm7gJhS59npnOAv6jqA9HndB/wV+C8HLffkhtU9QVV7QS+jzs3xw9Cuo24YG5MNxZ4i4yqrlbVy1V1H1yJ5DLgW0QBL/Kuqo7IfOF+2DL9D1CD+wEFV9pYBWSXqq6J0hirqger6qwtZO+PwOlRNesHovzdDS5AiMjVIvJWVBW4HtgbV7oZiO2AQ7IuLv6AK231Zh3QZ0kFdw+9S2vW+655tRnvV6pqa8b7+cBkyOm4pwNzcshTb8YA5cDGwS1UtRnXqX7mgOrLMpa3RJOZx5BtZRSku2Sfh67j7UpjSlYeQtx56MrD5Oh9Zh5WZqS3HXBa1ud5Be4WSC667T8yj80HlR+I+V0T6jqvX0j0+W6lOlz7CmO6scBbxFS1VVVvxJWg3tPPzT+Fu1/7mogsx5VoG+i9kVUu/gF04KpazwHujEo34IY8Ox9XjdsQXQy8TO+NwpqA6qx5EzOmFwCPZF1g1EcNwXrzIjCgqu0+jM2qtp2OO5/Q93HPZ8slz75GKVmFO+fTu2aISA0wlvyOrbwoKw9e9L4rD0uyllfT/aJrAfCHrM+zTlV3H8j+IzMy9t/X9wl6P9eZ+RbcbYWuz7dbuiISx537LpkXL9n2wH0njenGAm8REdfI57viGhUlogYtJ+P+gP/Vj3R2w923OwkXsLte++NKjMcOJH9RFeTNwBeAj5JRzYy7uk/jAoUnIufhSn69eQHYR0T2jY7zIlypqMvNQFJEzhORiqhkOUNEjt5Cmn8Djuz3gfXNA74vIpUiMgNXjdp1L6+v474V2Flc46wqESkTkcw8LmcLgTkqWd4MfFtEJkYXANcCbwLPDdLx5eJG4DIR2SlqD/B1IA7cFy2/BbhURLYXkUpcdXzm78uvgY+LyAkZ3+3dRGRmjvu/CThZRI4SkZiIHIP7DnbdSnkJd4F0fPRdOQk4LCuN3s71eSKyT1STcylQlXFcLwBHiGtIWA5cA2Q28FuOa1yV+d1FRGpxf29/z/H4zDbEAm9x6cRdTd+Nq6JaBXwD+IKq3tWPdD4N/FdVZ6nq8ozXK8BdbKGRVQ7+CMzEVXdn/vDfhGuk9Dau9LMbW7hYUNXHgR8DD+KqOMcB/85Yvhx4P66l8nxcNfJfcaWc3twC7B0Fx8G0AFcCehd3jA/iAgv0cdxRA5zDcQ3DFuN+qDMbZn0duFpcS+Hretn/lwAf10p2Ia569sPRhVC+/BD3iMzDwArcrYYPRa13wd1/fwg37vO7UT4XdG2sqq/h7pt+Efd5r8QF85xuRajqv3H3un+E+y78ADhDVZ+Jls/DNZD6He5v52jgL1nJ9Haufwf8PEr3VOA4Vd0QLbsNFzz/i6vaXoj7nLvyNQf4DfBcVIXe1VjsNOCfqjo3l+Mz2xYbj9eUFBG5EDhEVXNqLZtDeufgGjbZ85glSETm4z7fW/tatx9plgOv4S6OZg9WuqZ0xAudAWMGk6r+FvhtofNhtl1Rq+8t3dc32zirajbGGGPyyKqajTHGmDyyEq8xxhiTRxZ4jTHGmDyywGuMMcbkkQVeY4wxJo8s8BpjjDF5ZIHXGGOMySMLvMYYY0weWeA1xhhj8sgCrzHGGJNHFniNMcaYPLLAa4wxxuSRBV5jjDEmjyzwGmOMMXlkgdcYY4zJIwu8xhhjTB5Z4DXGGGPyaJsOvCLygIicXeh8GGOM2XYMu8ArIs0Zr1BE2jLef6I/aanqMap601DldTgSkZEi8lcRaRGRBSJy+hbWHSEiN4nIyuh1Zdby94jIv0Rkg4gsFpFvZi0/X0Tejj67B0Vk4hAdljHGFI1hF3hVtabrBSwETsiYd1vXeiISz1ee8rmvPPgV0AmMAz4B/EZEdu9l3Z8AVcB0YH/gTBE5N2P57cCTwEhgJvBZEfkwgIgcDnwHODFa/i5wx+AeijHGFJ9hF3h7IyKHR6Wqr4jIcuCPItIgIveKyCoRWRdNT87Y5nEROT+aPkdEnhKRH0Xrvisix2xhf/Ojfb0CtIhIXEQ+LCKvi8j6KO1dM9afIiJ3R3lZIyK/7ON4theRx6J1V4vIbSIyImO5isgOGe9vFJH/l/H+RBF5SUQaRWSeiBydwzmsBk4Gvqmqzar6FPB34MxeNjkB+IGqtqrqfOAG4LyM5dOB21Q1UNV5wFNAVxA/HrhLVV9X1U7g28BhIrJ9X/k0xpjhrGQCb2Q8rvQ0DfgU7vj+GL2fCrQBWwp4BwBvAaOBHwA3iIhsYf3TgOOAEcAMXInti8AY4H5gloiUiUgMuBdYgAtGk4A7+zgWAb4LTAR2BaYAV/axjdtQZH/gZuDSKG+HAfOjZV8VkXt72XQnIK2qczLmvcymYNlbPjOn98h4/1PgLBFJiMjOwEHAI1vYlqztjTGm5JRa4A2BK1S1Q1XbVHWNqv4lKpE1Adfgqjx7s0BVf6+qAXATMAFX5dqbn6vqIlVtA04F7lPVf6hqCvgRUAkcjKuGnQhcqqotqtoelSZ7papvR2l1qOoq4Md95D3TJ4E/RNuHqrpEVd+M0v2eqh7fy3Y1QGPWvA1AbS/rPwh8VURqo9L3ebiq5y73Ah/DXfC8Cdygqs9nbHuKiOwlIpXAtwDN2t4YY0pOqQXeVara3vVGRKpE5LqokVAj7n7jiKgE2pPlXROq2hpN1mxhf4sypifiSrRd24fR8km40uoCVU3neiAiMk5E7hSRJVHeb8WVxHMxBZiX674yNAN1WfPqgKZe1v8CLqjOBe7BlfgXg2ukhQuuVwMVUZ6OEpHPAqjqI8AVwF9wpfH50X4WDyDfxhgzbJRa4NWs95cAOwMHqGodrsoVuldxDtb+luKqtN0OXBX1FGAJLgBP7WcjrO9E6e8Z5f0Muue7le6lw/EZ04uAgdwrnQPERWTHjHl7A6/3tLKqrlXVT6jqeFXdHfd9ei5aPAMIVPVmVU2r6mJc9fqxGdv/SlV3VNVxuAAcB14bQL6NMWbYKLXAm60WVyJbH5XArhjCff0fcJyIHCEiCVzQ7wCexgWjZcD3RKRaRCpE5JAc8t4MbBCRSbj7tZleAk4XkVjUcCqzGvoG4NwoL56ITBKRXfo6AFVtAe4Gro7yeQiu1fEtPa0fNQAbFeXhGNx99a4GXnPcKnJ6lIfxuOr4V6JtK0RkD3GmAr8Dfqaq66Ll54jI/L7ybIwxw02pB96f4u6zrgaewVV9DglVfQtXKv1FtL8TcI86dUb3jE8AdsA9ArUYF4S25CpgH9w91vtwATHTxVGa63GP/fwtIy/PAefiHvfZADxBVBoXkctF5IEt7PezuHO2Eld1/BlVfT3a9lARac5Yd1/gVVwV8XeBT3Stq6qNwEeBLwHrcBcKr7EpMFfgHjdqxl2Y/AfIfM53CvDvLeTTGGOGJVHNrp01pvBE5GHgYlWdXei8GGPMYLLAa4wxxuRRqVc1FzUR+a107wKz6/XbQufNGGPM0LASrzHGGJNHVuI1xhhj8sgCrzHGGJNHwy7wyiAOCxilt3GghG2NuGH7XhCR1uj/9/Sx/sdFZLa4IQPnicihGctOiZY1icgbIvKRjGXZ97I7RKS33rCMMaakDbvAm+uwgEOln71PFS0RKcN183gr0IDrm/qeaH5P638Q+D7u+eBaXC9g70TLJkXp/C+ui8lLgdtFZCyAql6Y9bndAdw1hIdnjDFFa9gF3t5EvSN9NSqJrRGR/4t6q+rqJenWaP56EXk+6gv5GuBQ4JdRSWyzkYtEZLq4Ifg+KSILgceifX0j6gN6pYjcLCL1Gdu8T0Sejva1SETO6SPvx4nIi+KG8FskGQPKSzTcYdb680XkyGg6FnWKMS8qbb4gIlNyOGWH47po/Gk0EMPPcV1SfqCX9a8CrlbVZzIGXlgSLZsMrFfVB9S5D2ihh24rZdPQgzflkEdjjCk5JRN4gc8DH8F1nTgR11vSr6JlZwP1uN6QRgEXAm2q+nXgX8BFUWnsoi2kPxM3PN9RwDnR6/24PolriIYbFJFpwAO4HqzGAO/B9dq0JS3AWbgh/I4DPpNZVduH/8UNT3gsrrR5Hq4fZ8SNP/zVXrbbHXhFuzdrf4UehgAUN6hEEhgjIm+LG/f4l+JGFQLwgdnixiOORXnviNLLdjKwCjdghTHGbHNKoto0ciEugHaNjnMlsFBEzgRSuIC7g6q+ArwwgPSvjPoyJrqX/GNV7apq/RrwmoicC5wOPKKqd0TbrYlevVLVxzPeviIid+AC/d9yyNf5wGVRl5Xgxs/tSre34f/AXSxsyJrX2xCA44AEboi/Q3Hn8x7gG8DXVTUQkZtxXUBWAJ3A/3SdryxnAzdnBXxjjNlmlFKJdxrw16h6dz0wGwhwQeMW4CHgThFZKiI/EDeQQX/0OgRgNB2P9tXvIflE5AAR+aeIrBKRDbiLiGIaArAt+v8XqrpMVVfjxgc+FiCq9v4Brvq6DHfRcH12Yy1xgyEcDtw8gPwaY0xJKKXAuwg4RlVHZLwqonuRKVW9SlV3ww1Mfzyuahc2H0qwN70OAQhMBdLACgY2JN/twN+BKapaD/yWTUMAtpAx/F9U7TsmY9uBDgH4OrCXiGQONbgXPQwBGI0YtJju5yBz+j3Ak6rqR/d/nweeBY7MSupM4N9dNQXGGLMtKqXA+1vgmugeKyIyRkROjKbfLyJ7RkGrEVdVGkbbrcDdp+2PO4Avich2IlKDGzv3T9FA97cBR0aP18TFDZv3nj7SqwXWqmq7iOyPq67uMgeoiBpgJXDVu+UZy68Hvi0iO4qzl4iMyuEYHsfVCHxBRMpFpOv+9mO9rP9H4PMiMlZEGnCjDt0bLXseOLTrOEXkvbgq6ex7vGcBN+aQN2OMKVmlFHh/his1Phw9I/oMcEC0bDzwZ1zQnY0bJu+WjO0+JiLrROTnOe7rD9H2TwLvAu24xl2o6kJcFewlwFpcw6q9+0jvs7gxcJuAb+HG9iVKb0O0/HpgCa4EnNnK+cfR+g9Hx3cDblg/ROQBEbm8px2qaieuMdpZuKEFzwM+Es3vafjAb+MC7BzcOXwRuCZK6wngSuDP0TH8BfiOqj7ctbGIHIRr/bzZY0RbyqcxxpQa66vZGGOMyaNSKvEaY4wxRc8Cb56IyOvS8xCA/e7m0hhjzPBlVc3GGGNMHlmJ1xhjjMkjC7zGGGNMHlngNcYYY/LIAq8xxhiTR/8fnwcQM2uxXgQAAAAASUVORK5CYII=\n" + "image/png": "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\n", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " 2021-02-11T16:04:59.520565\n", + " image/svg+xml\n", + " \n", + " \n", + " Matplotlib v3.3.3, https://matplotlib.org/\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "\n" + ], + "text/plain": [ + "
" + ] }, "metadata": { "needs_background": "light" - } + }, + "output_type": "display_data" } ], "source": [ @@ -360,21 +15066,21 @@ "Before using the ShapModelInterpreter consider the following tips:\n", "\n", "- Make sure you do not underfit or overfit the model. Underfitting will cause only the most important relations in the data to be visible, while overfitting will present relationships that do not generalize.\n", - "- Perform feature selection process before fitting the final model. This way, many it will be easier to interpret the explanation. Moreover, highly-correlated features will affect the explanation less.\n", - "- Preferably use model that handles NaNs e.g. LightGBM or impute them before using SHAP. When imputing also extract a MissingIndicator to get insights into when NaNs are meaningful for the model.\n", - "- For categorical features either use a model that handles them e.g. LightGBM, or apply one-hot encoding. Keep in mind that with One-hot encoding the importance of categorical feature might be spread over multiple encoded features." + "- Perform a feature selection process before fitting the final model. This way, it will be easier to interpret the explanation. Moreover, highly-correlated features will affect the explanation less.\n", + "- Preferably use a model that handles NaNs e.g. LightGBM or impute them beforehand using SHAP. When imputing also extract a MissingIndicator to get insights into when NaNs are meaningful for the model.\n", + "- For categorical features either use a model that handles them e.g. LightGBM, or apply One-hot encoding. Keep in mind that with One-hot encoding the importance of a categorical feature might be spread over multiple encoded features." ] } ], "metadata": { "kernelspec": { - "name": "Python 3.6.12 64-bit ('probatus': conda)", "display_name": "Python 3.6.12 64-bit ('probatus': conda)", "metadata": { "interpreter": { "hash": "c0157df04084817c05c209d47d347130bc44c25a3599ff4e17593311298b7c1e" } - } + }, + "name": "Python 3.6.12 64-bit ('probatus': conda)" }, "language_info": { "codemirror_mode": { @@ -391,4 +15097,4 @@ }, "nbformat": 4, "nbformat_minor": 4 -} \ No newline at end of file +} diff --git a/probatus/feature_elimination/feature_elimination.py b/probatus/feature_elimination/feature_elimination.py index 25d25f9d..304cb3af 100644 --- a/probatus/feature_elimination/feature_elimination.py +++ b/probatus/feature_elimination/feature_elimination.py @@ -113,7 +113,7 @@ def __init__( Args: clf (binary classifier, sklearn compatible search CV e.g. GridSearchCV, RandomizedSearchCV or BayesSearchCV): - A model that will be optimized and trained at each round of features elimination. The recommended model + A model that will be optimized and trained at each round of feature elimination. The recommended model is [LGBMClassifier](https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.LGBMClassifier.html), because it by default handles the missing values and categorical variables. This parameter also supports any hyperparameter search schema that is consistent with the sklearn API e.g. @@ -122,10 +122,10 @@ def __init__( or [BayesSearchCV](https://scikit-optimize.github.io/stable/modules/generated/skopt.BayesSearchCV.html#skopt.BayesSearchCV). step (int or float, optional): - Number of lowest importance features removed each round. If it is an int, then each round such number of - features is discarded. If float, such percentage of remaining features (rounded down) is removed each + Number of lowest importance features removed each round. If it is an int, then each round such a number of + features are discarded. If float, such a percentage of remaining features (rounded down) is removed each iteration. It is recommended to use float, since it is faster for a large number of features, and slows - down and becomes more precise towards less features. Note: the last round may remove fewer features in + down and becomes more precise with fewer features. Note: the last round may remove fewer features in order to reach min_features_to_select. If columns_to_keep parameter is specified in the fit method, step is the number of features to remove after keeping those columns. @@ -141,7 +141,7 @@ def __init__( If None, then cv of 5 is used. scoring (string or probatus.utils.Scorer, optional): - Metric for which the model performance is calculated. It can be either a metric name aligned with predefined + Metric for which the model performance is calculated. It can be either a metric name aligned with predefined [classification scorers names in sklearn](https://scikit-learn.org/stable/modules/model_evaluation.html). Another option is using probatus.utils.Scorer to define a custom metric. @@ -152,7 +152,7 @@ def __init__( verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). @@ -160,7 +160,7 @@ def __init__( random_state (int, optional): Random state set at each round of feature elimination. If it is None, the results will not be reproducible and in random search at each iteration a different hyperparameters might be tested. For - reproducible results set it to integer. + reproducible results set it to an integer. """ # noqa self.clf = clf @@ -203,7 +203,7 @@ def _get_current_features_to_remove(self, shap_importance_df, columns_to_keep=No If step is a positive integer, at each round step lowest SHAP importance features are selected. If it is a float, such percentage of remaining features (rounded up) is removed each iteration. It is recommended to use float, since it is - faster for a large set of features, and slows down and becomes more precise towards less features. + faster for a large set of features, and slows down and becomes more precise with fewer features. Args: shap_importance_df (pd.DataFrame): diff --git a/probatus/interpret/inspector.py b/probatus/interpret/inspector.py index 39273ee4..21af5a6a 100644 --- a/probatus/interpret/inspector.py +++ b/probatus/interpret/inspector.py @@ -31,14 +31,14 @@ def return_confusion_metric(y_true, y_score, normalize=False): """ Computes a confusion metric as absolute difference between the y_true and y_score. - If normalize eis set to tru, it will normalize y_score to the maximum value in the array + If normalize is set to true, it will normalize y_score to the maximum value in the array Args: y_true: (np.ndarray or pd.Series) true targets y_score: (np.ndarray or pd.Series) model output normalize: boolean, normalize or not to the maximum value - Returns: (np.ndarray or pd.Series) conflusion metric + Returns: (np.ndarray or pd.Series) confusion metric """ @@ -140,7 +140,7 @@ class InspectorShap(BaseInspector): """ Class to perform inspection of the model prediction based on Shapley values. - It uses the calculated shapley values for the train model to build clusters in the shap space. + It uses the calculated Shapley values for the train model to build clusters in the shap space. For each cluster, an average confusion, average predicted probability and observed rate of a single class is calculated. Every sub cluster can be retrieved with the function slice_cluster to perform deeper analysis. @@ -323,7 +323,7 @@ def perform_fit_calc(self, X, y, fit_clusters=False, **shap_kwargs): def _compute_report(self): """ - Helper function to compute the report of the ispector. + Helper function to compute the report of the inspector. Performs aggregations per cluster id """ @@ -353,7 +353,7 @@ def compute(self): - average predicted probabilitites - average confusion - If multiple eval_sets were passed in the inspect() functions, the output will contain those aggregations as well + If multiple eval_sets were passed in the inspect() functions, the output will contain those aggregations as well. The output names will use the sample names provided in the inspect function. Otherwise they will be labelled by the suffix sample_{i}, where i is the index of the sample @@ -406,10 +406,10 @@ def slice_cluster( if not passed the slicing is performed on y passed to inspect predicted_proba: Optional parameter - the predicted_proba on which the masking should be performed. if not passed the slicing is performed on predicted_proba generated by inspect method on X and y - complementary: flag that returns the cluster_id if set to False, otherwise the complementary dataframe (ie + complementary: flag that returns the cluster_id if set to False, otherwise the complementary dataframe (i.e. those with ~mask) - Returns: tuple: Dataframe of sliced shapley values, series of sliced targets, sliced probabilities + Returns: tuple: Dataframe of sliced Shapley values, series of sliced targets, sliced probabilities """ if self.cluster_report is None: self.compute() @@ -484,8 +484,8 @@ def create_summary_df(cluster, y, probas, normalize=False): by concatenating the cluster series, the targets, the probabilities and the measured confusion. Args: - cluster: pd.Series od clusters - y: pd.Series od targets + cluster: pd.Series of clusters + y: pd.Series of targets probas: pd.Series of predicted probabilities of the model normalize: boolean (if the predicted probabilities should be normalized to the max value diff --git a/probatus/interpret/model_interpret.py b/probatus/interpret/model_interpret.py index 55b02ef0..cdb73ca7 100644 --- a/probatus/interpret/model_interpret.py +++ b/probatus/interpret/model_interpret.py @@ -36,9 +36,9 @@ class ShapModelInterpreter(BaseFitComputePlotClass): """ - This class is a wrapper that allows to easily analyse model's features. + This class is a wrapper that allows to easily analyse a model's features. - It allows to plot SHAP feature importance, + It allows us to plot SHAP feature importance, SHAP summary plot and SHAP dependence plots. Example: @@ -95,7 +95,7 @@ def __init__(self, clf, scoring="roc_auc", verbose=0): verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - niether prints nor warnings are shown - 1 - 50 - only most important warnings - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). @@ -224,7 +224,7 @@ def _prep_shap_related_variables( def compute(self, return_scores=False): """ - Computes the DataFrame, that presents the importance of each feature. + Computes the DataFrame that presents the importance of each feature. Args: return_scores (bool, optional): diff --git a/probatus/interpret/shap_dependence.py b/probatus/interpret/shap_dependence.py index 2e055744..4e8311ea 100644 --- a/probatus/interpret/shap_dependence.py +++ b/probatus/interpret/shap_dependence.py @@ -68,7 +68,7 @@ def __init__(self, clf, verbose=0): verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings regarding data properties are shown (excluding SHAP warnings) - 51 - 100 - shows most important warnings, prints of the feature removal process - above 100 - presents all prints and all warnings (including SHAP warnings). diff --git a/probatus/metric_volatility/volatility.py b/probatus/metric_volatility/volatility.py index 070033b2..5d8ae228 100644 --- a/probatus/metric_volatility/volatility.py +++ b/probatus/metric_volatility/volatility.py @@ -86,7 +86,7 @@ def __init__( verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings and indication of progress in fitting the object. - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). @@ -462,7 +462,7 @@ def __init__( verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). @@ -618,7 +618,7 @@ def __init__( verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). @@ -720,7 +720,7 @@ def __init__( verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). diff --git a/probatus/sample_similarity/resemblance_model.py b/probatus/sample_similarity/resemblance_model.py index e4fcea41..29cb2be0 100644 --- a/probatus/sample_similarity/resemblance_model.py +++ b/probatus/sample_similarity/resemblance_model.py @@ -36,13 +36,14 @@ class BaseResemblanceModel(BaseFitComputePlotClass): """ - This model checks for similarity of two samples. + This model checks for the similarity of two samples. - A possible use case is analysis whether train sample differs - from test sample, due to e.g. non-stationarity. + A possible use case is analysis of whether th train sample differs + from the test sample, due to e.g. non-stationarity. - This is a base class and needs to be extended by a fit() method, which implements how data is split, how model is - trained and evaluated. Further, inheriting classes need to implement how feature importance should be indicated. + This is a base class and needs to be extended by a fit() method, which implements how the data is split, + how the model is trained and evaluated. + Further, inheriting classes need to implement how feature importance should be indicated. """ def __init__( @@ -62,10 +63,10 @@ def __init__( Binary classification model or pipeline. scoring (string or probatus.utils.Scorer, optional): - Metric for which the model performance is calculated. It can be either a metric name aligned with + Metric for which the model performance is calculated. It can be either a metric name aligned with predefined [classification scorers names in sklearn](https://scikit-learn.org/stable/modules/model_evaluation.html). - Another option is using probatus.utils.Scorer to define a custom metric. Recommended option for this + Another option is using probatus.utils.Scorer to define a custom metric. The recommended option for this class is 'roc_auc'. test_prc (float, optional): @@ -77,7 +78,7 @@ class is 'roc_auc'. verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). @@ -85,7 +86,7 @@ class is 'roc_auc'. random_state (int, optional): Random state set at each round of feature elimination. If it is None, the results will not be reproducible and in random search at each iteration a different hyperparameters might be tested. For - reproducible results set it to integer. + reproducible results set it to an integer. """ # noqa self.clf = clf self.test_prc = test_prc @@ -253,13 +254,12 @@ def fit_compute( List of class names assigned, in this case provided samples e.g. ['sample1', 'sample2']. If none, the default ['First Sample', 'Second Sample'] are used. - return_scores (bool, optional): Flag indicating whether the method should return a tuple (feature importances, train score, test score), or feature importances. By default the second option is selected. **fit_kwargs: - In case any other arguments are accepted by fit() method, they can be passed as keyword arguments + In case any other arguments are accepted by fit() method, they can be passed as keyword arguments. Returns: (tuple of (pd.DataFrame, float, float) or pd.DataFrame): @@ -278,19 +278,20 @@ def plot(self): class PermutationImportanceResemblance(BaseResemblanceModel): """ - This model checks for similarity of two samples. + This model checks the similarity of two samples. - A possible use case is analysis whether train sample differs - from test sample, due to e.g. non-stationarity. + A possible use case is analysis of whether the train sample differs + from the test sample, due to e.g. non-stationarity. - It assigns to labels to each sample, 0 to first sample, 1 to the second. Then, It randomly selects a portion of - data to train on. The resulting model tries to distinguish which sample does a given test row comes from. This - provides insights on how distinguishable these samples are and which features contribute to that. The feature - importance is calculated using permutation importance. + It assigns labels to each sample, 0 to the first sample, 1 to the second. Then, it randomly selects a portion of + data to train on. The resulting model tries to distinguish which sample a given test row comes from. This + provides insights on how distinguishable these samples are and which features contribute to that. The feature + importance is calculated using permutation importance. - If the model achieves test AUC significantly different than 0.5, it indicates that it is possible to distinguish - the samples, and therefore, the samples differ. Features with high permutation importance contribute to that - effect the most. Thus, their distribution might differ between two samples. + If the model achieves a test AUC significantly different than 0.5, it indicates that it is possible to distinguish + between the samples, and therefore, the samples differ. + Features with a high permutation importance contribute to that effect the most. + Thus, their distribution might differ between two samples. Examples: ```python @@ -329,7 +330,7 @@ def __init__( feature are done. scoring (string or probatus.utils.Scorer, optional): - Metric for which the model performance is calculated. It can be either a metric name aligned with + Metric for which the model performance is calculated. It can be either a metric name aligned with predefined [classification scorers names in sklearn](https://scikit-learn.org/stable/modules/model_evaluation.html). Another option is using probatus.utils.Scorer to define a custom metric. Recommended option for this @@ -344,7 +345,7 @@ class is 'roc_auc'. verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). @@ -374,10 +375,10 @@ class is 'roc_auc'. def fit(self, X1, X2, column_names=None, class_names=None): """ - This function assigns to labels to each sample, 0 to first sample, 1 to the second. + This function assigns labels to each sample, 0 to the first sample, 1 to the second. Then, it randomly selects a - portion of data to train on. The resulting model tries to distinguish which sample does a given test row + portion of data to train on. The resulting model tries to distinguish which sample a given test row comes from. This provides insights on how distinguishable these samples are and which features contribute to that. The feature importance is calculated using permutation importance. @@ -448,14 +449,14 @@ def plot(self, ax=None, top_n=None, show=True, **plot_kwargs): Args: ax (matplotlib.axes, optional): - Axes to which the output should be plotted. If not provided a new axes are created. + Axes to which the output should be plotted. If not provided new axes are created. top_n (int, optional): - Number of the most important features to be plotted. By default are features are included into the plot. + Number of the most important features to be plotted. By default features are included in the plot. show (bool, optional): - If True, the plots are showed to the user, otherwise they are not shown. Not showing plot can be useful, - when you want to edit the returned axis, before showing it. + If True, the plots are shown to the user, otherwise they are not shown. Not showing a plot can be useful + when you want to edit the returned axis before showing it. **plot_kwargs: Keyword arguments passed to the matplotlib.plotly.subplots method. @@ -510,16 +511,16 @@ class SHAPImportanceResemblance(BaseResemblanceModel): """ This model checks for similarity of two samples. - A possible use case is analysis whether train sample differs - from test sample, due to e.g. non-stationarity. + A possible use case is analysis of whether the train sample differs + from the test sample, due to e.g. non-stationarity. - It assigns to labels to each sample, 0 to first sample, 1 to the second. Then, It randomly selects a portion of data - to train on. The resulting model tries to distinguish which sample does a given test row comes from. This + It assigns labels to each sample, 0 to the first sample, 1 to the second. Then, it randomly selects a portion of data + to train on. The resulting model tries to distinguish which sample a given test row comes from. This provides insights on how distinguishable these samples are and which features contribute to that. The feature importance is calculated using SHAP feature importance. If the model achieves test AUC significantly different than 0.5, it indicates that it is possible to distinguish - the samples, and therefore, the samples differ. Features with high permutation importance contribute to that + between the samples, and therefore, the samples differ. Features with a high permutation importance contribute to that effect the most. Thus, their distribution might differ between two samples. This class currently works only with the Tree based models. @@ -558,7 +559,7 @@ def __init__( Binary classification model or pipeline. scoring (string or probatus.utils.Scorer, optional): - Metric for which the model performance is calculated. It can be either a metric name aligned with + Metric for which the model performance is calculated. It can be either a metric name aligned with predefined [classification scorers names in sklearn](https://scikit-learn.org/stable/modules/model_evaluation.html). Another option is using probatus.utils.Scorer to define a custom metric. Recommended option for this @@ -573,7 +574,7 @@ class is 'roc_auc'. verbose (int, optional): Controls verbosity of the output: - - 0 - nether prints nor warnings are shown + - 0 - neither prints nor warnings are shown - 1 - 50 - only most important warnings - 51 - 100 - shows other warnings and prints - above 100 - presents all prints and all warnings (including SHAP warnings). @@ -596,10 +597,10 @@ class is 'roc_auc'. def fit(self, X1, X2, column_names=None, class_names=None, **shap_kwargs): """ - This function assigns to labels to each sample, 0 to first sample, 1 to the second. + This function assigns labels to each sample, 0 to the first sample, 1 to the second. - Then, It randomly selects a - portion of data to train on. The resulting model tries to distinguish which sample does a given test row + Then, it randomly selects a + portion of data to train on. The resulting model tries to distinguish which sample a given test row comes from. This provides insights on how distinguishable these samples are and which features contribute to that. The feature importance is calculated using SHAP feature importance.