{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Regression\n",
"\n",
"Regression modeling is any attempt to predict or explain a continous variable from a collection of input data. This could be student GPA, the position of a planet orbiting a sun, or the color of a pixel in a photo. Values such as whether a student is a STEM student or not, the probability of an event occuring (such as changing a major, an earthquake) are not regression tasks (they are classification).\n",
"\n",
"After completing this tutorial you should be able to:\n",
"\n",
"* use `sci-kit learn` to split data into training and testing sets\n",
"* understand the model, fit, score paradigm in `sci-kit learn` and apply it to a problem\n",
"* understand the most important visualizations of regression analysis: actual vs. predicted, actual vs. residuals, residuals distribution vs. assumed theoretical distribution (in case of OLS models)\n",
"* have a conceptual understanding of the basic goal of any regression task\n",
"* have some understanding that most statistical \"tests\" are typically just specific solutions of a linear regression problem\n",
"* have some understanding of the assumptions of linear models\n",
"\n",
"## Further reading\n",
"\n",
"1. Hands on machine learning, probably the best practical machine learning textbook ever written https://github.com/ageron/handson-ml\n",
"2. Common statistical tests are linear models, stop thinking statistics are something other than y=mx+b, they are not. lol. https://lindeloev.github.io/tests-as-linear/?fbclid=IwAR09Rp4Vv18fOO4lg0ITnCYJICCC1iuzeq-tNYPWsnmK6CrGgdErpvHfyWE\n",
"\n",
"## Data\n",
"\n",
"Here is a file named [`regression_data.csv`](data/regression_data.csv). Import the data like you did in the previous tutorial \"exploring data\". The first step in any regression task is to explore the data the raw data.\n",
"\n",
"# 1. Import the data\n",
"\n",
"We will first need to import the data. To do so, we need to first import the relevant libraries that are necessary to import and visualize the data. Then, we can import the data into a dataframe for analysis. \n",
"\n",
"1. First import the ``pandas``, ``numpy``, and ``matplotlib.pyplot`` libraries\n",
"2. Then, import the data into a data frame using the ``read_csv()`` method."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline\n",
"#comment"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
cGPA
\n",
"
attendance
\n",
"
passed_percent
\n",
"
sex
\n",
"
hsGPA
\n",
"
ethnicity
\n",
"
fci_post
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
2.083621
\n",
"
51.137226
\n",
"
48.929495
\n",
"
1.0
\n",
"
2.642089
\n",
"
2.0
\n",
"
14.618734
\n",
"
\n",
"
\n",
"
1
\n",
"
2.043927
\n",
"
47.819398
\n",
"
31.298644
\n",
"
1.0
\n",
"
1.872415
\n",
"
2.0
\n",
"
10.652295
\n",
"
\n",
"
\n",
"
2
\n",
"
2.147087
\n",
"
34.160519
\n",
"
32.312813
\n",
"
0.0
\n",
"
2.068908
\n",
"
2.0
\n",
"
7.546426
\n",
"
\n",
"
\n",
"
3
\n",
"
2.286152
\n",
"
55.850494
\n",
"
43.496181
\n",
"
0.0
\n",
"
2.058815
\n",
"
2.0
\n",
"
14.041647
\n",
"
\n",
"
\n",
"
4
\n",
"
2.519186
\n",
"
77.501019
\n",
"
39.407797
\n",
"
0.0
\n",
"
1.932764
\n",
"
2.0
\n",
"
15.960932
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" cGPA attendance passed_percent sex hsGPA ethnicity fci_post\n",
"0 2.083621 51.137226 48.929495 1.0 2.642089 2.0 14.618734\n",
"1 2.043927 47.819398 31.298644 1.0 1.872415 2.0 10.652295\n",
"2 2.147087 34.160519 32.312813 0.0 2.068908 2.0 7.546426\n",
"3 2.286152 55.850494 43.496181 0.0 2.058815 2.0 14.041647\n",
"4 2.519186 77.501019 39.407797 0.0 1.932764 2.0 15.960932"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = pd.read_csv('data/regression_data.csv')\n",
"df.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# 2. Investigate the correlations\n",
"\n",
"Now that we have the data imported, you can see there's 7 variables for each student record. We are attempting to see what factors are connected with ``fci_post`` as we want to try to predict a measure of conceptual understanding. To do that it would be useful to see how each variable to correlates with the ``fci_post`` score. \n",
"\n",
"We can do this in a couple ways. \n",
"1. We can use ``pandas`` method ``corr`` to see the correlation coefficients. [[How to use corr]](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.corr.html)\n",
"2. We can use ``pandas`` plotting library to visualize the correlations. [[How to use scatter_matrix]](https://pandas.pydata.org/docs/reference/api/pandas.plotting.scatter_matrix.html)\n",
"\n",
"### Questions\n",
"Once you complete these correlational analysis, answer these questions.\n",
"\n",
"1. Which variables most strongly correlate with ``fci_post``?\n",
"2. Is there any conflict between the information gained from ``corr`` and ``scatter_matrix``? That is, does one provide better information about correlations?\n",
"3. Which variables might you expect to appear predictive in a model?"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"from pandas.plotting._misc import scatter_matrix\n",
"axes=scatter_matrix(df, figsize=(12,12))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# 3. Modeling\n",
"\n",
"Modeling data is as much an art as it is science. There is no \"true\" model, there is only a model that reduces error to an acceptable amount. Most models attempt to do this automatically by minimizing some sort of cost function (or error) using some kind of solver algorithm. These solving methods are beyond the scope of this workshop but are important to know they exist and somewhat how they work. If you are interested in this sort of thing I recommend starting with [this stats exchange thread](https://stats.stackexchange.com/questions/160179/do-we-need-gradient-descent-to-find-the-coefficients-of-a-linear-regression-mode) and googling each solver in the answer that seems interesting. This is only for Linear Least Squares models but its a good place to start. \n",
"\n",
"For this analysis, we will use the paradigm that we discussed where we split the data into a training set the develop the model and then use the model to predict the outcomes of a test set. \n",
"\n",
"\n",
"\n",
"## 3.1 Splitting the data\n",
"\n",
"We first need to split the data into a training set and a test set. To do this, we will also need to know which variable we intend to predict. The library ``sklearn`` has builtin methods for doing this splitting, so we will also need to import it. Notice that you can import a library any time that you need to.\n",
"\n",
"1. Import ``train_test_split`` from ``sklearn.model_selection``. \n",
"2. Look at your data and determine which columns will be the input features of your model and which will be the predicted variable. You might find using ``columns`` useful. [[Return column labels]](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.columns.html)\n",
"3. Split the data into training and testing data sets using the `sklearn.model_selection` method `train_test_split` [[How to use train_test_split]](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html)\n",
"\n",
"### Questions\n",
"\n",
"1. How large is the training data set?\n",
"2. How can you change the amount of data used in the training set? [[How to use train_test_split]](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html)"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"from sklearn.model_selection import train_test_split"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Index(['cGPA', 'attendance', 'passed_percent', 'sex', 'hsGPA', 'ethnicity',\n",
" 'fci_post'],\n",
" dtype='object')"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.columns"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
cGPA
\n",
"
attendance
\n",
"
passed_percent
\n",
"
sex
\n",
"
hsGPA
\n",
"
ethnicity
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
2.083621
\n",
"
51.137226
\n",
"
48.929495
\n",
"
1.0
\n",
"
2.642089
\n",
"
2.0
\n",
"
\n",
"
\n",
"
1
\n",
"
2.043927
\n",
"
47.819398
\n",
"
31.298644
\n",
"
1.0
\n",
"
1.872415
\n",
"
2.0
\n",
"
\n",
"
\n",
"
2
\n",
"
2.147087
\n",
"
34.160519
\n",
"
32.312813
\n",
"
0.0
\n",
"
2.068908
\n",
"
2.0
\n",
"
\n",
"
\n",
"
3
\n",
"
2.286152
\n",
"
55.850494
\n",
"
43.496181
\n",
"
0.0
\n",
"
2.058815
\n",
"
2.0
\n",
"
\n",
"
\n",
"
4
\n",
"
2.519186
\n",
"
77.501019
\n",
"
39.407797
\n",
"
0.0
\n",
"
1.932764
\n",
"
2.0
\n",
"
\n",
"
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
...
\n",
"
\n",
"
\n",
"
3995
\n",
"
2.067698
\n",
"
52.941889
\n",
"
65.051548
\n",
"
1.0
\n",
"
2.107277
\n",
"
2.0
\n",
"
\n",
"
\n",
"
3996
\n",
"
2.449005
\n",
"
63.282221
\n",
"
54.918656
\n",
"
0.0
\n",
"
2.205480
\n",
"
0.0
\n",
"
\n",
"
\n",
"
3997
\n",
"
2.227918
\n",
"
52.155313
\n",
"
52.040404
\n",
"
1.0
\n",
"
1.679202
\n",
"
0.0
\n",
"
\n",
"
\n",
"
3998
\n",
"
2.341053
\n",
"
46.928807
\n",
"
62.977916
\n",
"
0.0
\n",
"
2.468935
\n",
"
1.0
\n",
"
\n",
"
\n",
"
3999
\n",
"
2.582593
\n",
"
50.484197
\n",
"
63.746409
\n",
"
0.0
\n",
"
0.276639
\n",
"
0.0
\n",
"
\n",
" \n",
"
\n",
"
4000 rows × 6 columns
\n",
"
"
],
"text/plain": [
" cGPA attendance passed_percent sex hsGPA ethnicity\n",
"0 2.083621 51.137226 48.929495 1.0 2.642089 2.0\n",
"1 2.043927 47.819398 31.298644 1.0 1.872415 2.0\n",
"2 2.147087 34.160519 32.312813 0.0 2.068908 2.0\n",
"3 2.286152 55.850494 43.496181 0.0 2.058815 2.0\n",
"4 2.519186 77.501019 39.407797 0.0 1.932764 2.0\n",
"... ... ... ... ... ... ...\n",
"3995 2.067698 52.941889 65.051548 1.0 2.107277 2.0\n",
"3996 2.449005 63.282221 54.918656 0.0 2.205480 0.0\n",
"3997 2.227918 52.155313 52.040404 1.0 1.679202 0.0\n",
"3998 2.341053 46.928807 62.977916 0.0 2.468935 1.0\n",
"3999 2.582593 50.484197 63.746409 0.0 0.276639 0.0\n",
"\n",
"[4000 rows x 6 columns]"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"features = ['cGPA','attendance','passed_percent', 'sex', 'hsGPA', 'ethnicity']\n",
"output = ['fci_post']\n",
"df[features]"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"X_train, X_test, y_train, y_test = train_test_split(df[features], df[output])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 3.2 Creating and scoring the model\n",
"\n",
"Now that we have split the data into training and test sets, we can build a model of the training set. We will focus first on a linear model using an ordinary least squares (OLS) fit. This is likely a model that you are familiar with, particularly for lines of best fit between two measurements. The general approach is to construct a linear model for student records that minimizes the error using OLS. to do this we need to import the ``LinearRegression`` method from ``sklearn.linear_model``, then create a model, fit it, and score it. *Notice: this approach to using linear regression with sci-kit learn is quite similar across other regression methods.*\n",
"\n",
"1. Import the ``LinearRegression`` method from ``sklearn.linear_model`` [[How to use LinearRegression]](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html#sklearn.linear_model.LinearRegression)\n",
"2. Create an OLS model and fit it.\n",
"3. Score the model using your model's built in `score` method. \n",
"\n",
"\n",
"### Questions\n",
"\n",
"1. What does score represent? What is it summarizing? [[The score method]](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html#sklearn.linear_model.LinearRegression.score)\n",
"2. Are we justified in using a linear model? [[Read about assumptions of linear models]](https://statisticsbyjim.com/regression/ols-linear-regression-assumptions/)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"from sklearn.linear_model import LinearRegression"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"LinearRegression()"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"LR = LinearRegression()\n",
"LR.fit(X_train, y_train)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"0.6990618724731259"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"LR.score(X_test,y_test)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# 4. Analysing the model output\n",
"\n",
"Now that we have established the goal of the model is to minimize the error, created a model, and found a score for the model, we still must recognize that the model has some error. The error/residual is really just the linear distance from the model \"plane\" to the predicted value as shown below:\n",
"\n",
"\n",
"\n",
"These residuals are data in their own right. But instead of being data about students, courses, etc. they are data about the model and how it is giving predictions. Thus we can use them to describe the model performance.\n",
"\n",
"## 4.1 Predicting from test data\n",
"\n",
"We will start by investigating how well our model, constructed from the training set, predicts the scores from test set.\n",
" \n",
"1. Create predicted data using the model's `predict` method. \n",
"2. Make a scatter plot to compare it to the actual values and draw a diagonal through this plot. \n",
"\n",
"### Questions\n",
"\n",
"1. What \"shape\" does the scatter plot \"blob\" look like? \n",
"2. Does the \"blob\" follow the diagonal line or does it deviate in some way?\n",
"3. Can you tell if the model over or under predicts scores in the test set?"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"predicted = LR.predict(X_test)"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Text(0.5, 0, 'Actual FCI Score')"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots(figsize=(10,10))\n",
"\n",
"ax.scatter(x=y_test, y=predicted, alpha=0.15)\n",
"ax.plot([-0, 30], [-0, 30], color='red')\n",
"ax.set_ylabel('Predicted FCI score', fontsize=15)\n",
"ax.set_xlabel('Actual FCI Score', fontsize=15)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 4.2 Inspecting the residuals of the model\n",
"\n",
"One of the major assumptions of a linear model is that error is normall distributed. Basically, we aim for the error in the model to be distributed equally around zero, so that there's little [heteroscedasticity](https://statisticsbyjim.com/regression/heteroscedasticity-regression/). If a linear model has errors that are not normally dsitributed, we are might be in a little be of trouble with regard to believing the model, and we might have to try another modeling approach. One way to look into this is to compute and plot the residuals. They should be roughly normally distributed if we are justfied in using a linear model. This analysis will tell us if our model tends to overpredict or underpredict scores and for which scores it does so.\n",
"\n",
"\n",
"1. Write a function to calculate the residuals of the model. \n",
"2. Plot the actual values versus the residuals using a scatter plot. (*This is the most common way of seeing a residual analysis in practice.*)\n",
"3. Collapse the residual scatter plot into a histogram. (*This is a useful visualization to see the normality of the distribution*) [[How to plot a histogram]](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.hist.html)\n",
"\n",
"### Questions\n",
"\n",
"1. Do we appear to be justified in using a linear model?\n",
"2. Does the model tend to overpredict or underpredict certain groups of scores?"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"def calculate_residuals(actual, predicted):\n",
" return predicted - actual\n",
"\n",
"residuals = calculate_residuals(actual=y_test, predicted=predicted)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Text(0.5, 0, 'Actual')"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots(figsize=(10,8))\n",
"\n",
"residuals['fci_post'].hist()\n",
"\n",
"ax.set_xlabel('Residuals', fontsize=15)\n",
"ax.set_ylabel('Counts', fontsize=15)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# 5. Model Features - training and fitting\n",
"\n",
"All models have some input data X and some output prediction Y. The input data X is of the shape $m \\times n$, so that means there are $m$ columns (or features) and $n$ data \"points\" (or vectors if $m>1$). For many models, you can return values from the model that give some indication as to how \"important\" each particular feature is to the model's training. Typically, the larger the magnitude of this value, the more important the feature is for prediction. This value for linear models is called the model *coefficients*. It may also be called *feature importance*. These values are always calculated from the data that was used to train (fit) the model. Thus, they don't really tell us about how important the features are for new data, rather how important the features were in deciding the \"shape\" of the model itself.\n",
"\n",
"## Finding fit coefficients\n",
"\n",
"For our linear model, the coefficients are related to the correlation between each input varaiable and the output prediction. Earlier you looked at the correlations between each input variable and the output variable. Now, we return the linear fit coefficients and plots them to see which features are most \"important\" to our model. ``LinearRegression`` has a builtin attribute ``coef_`` that returns these fit coefficients.\n",
"\n",
"1. Return the fit coefficients using ``coef_``\n",
"2. Make a bar graph of all the features in the model. [[How to make a horizontal bar plot]](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.plot.barh.html)\n",
"\n",
"\n",
"### Questions \n",
"1. Which is the most important feature for fitting? \n",
"2. Which is least important?"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots(figsize=(10,8))\n",
"coefs_df['coef'].plot.barh()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# 6. Model Features - predicting\n",
"\n",
"The correlary to each feature's coefficient or importance value, is the amount of variance that feature explains in the prediction. Remember, we have split the data into two separate sets, the training data and the testing data. The test data is never shown to the model until after the model is \"fit\" to the training data. This secrecy is why we are able to test the predictive power of each model. This secret or \"hold out\" data can be used to measure the \"explained variance\" of each coefficient/feature. One method of doing this is called [recursive feature elimination](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFE.html). Essentially, the coefficient of the model are ordered by magnitude, and the smallest are then removed one at a time until only one feature is left. Each iteration the model's `score` function is called. This provides a ranking based on the predictive power of the features.\n",
"\n",
"## Finding the explained variance\n",
"\n",
"1. Import ``RFECV`` from ``sklearn.feature_selection``.\n",
"2. Using the `RFE` function, calculate the explained variance of each of the features in your model using ``grid_scores_``. [[How to use RFE]](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFE.html)\n",
"3. Plot the scores returned for each of the combination of features from largest contributions to smallest as a line plot.\n",
"\n",
"### Questions\n",
"\n",
"1. What fraction of the variance is explained by the whole model?\n",
"2. Which input features explain the most variance?\n",
"3. Which explain the least and could be dropped in order to find a parsimonious model?"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"RFECV(estimator=LinearRegression())"
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.feature_selection import RFECV\n",
"rfe = RFECV(estimator=LinearRegression(), )\n",
"rfe.fit(X_train, y_train)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/caballero/opt/anaconda3/envs/JupyterBook/lib/python3.10/site-packages/sklearn/utils/deprecation.py:103: FutureWarning: The `grid_scores_` attribute is deprecated in version 1.0 in favor of `cv_results_` and will be removed in version 1.2.\n",
" warnings.warn(msg, category=FutureWarning)\n"
]
},
{
"data": {
"text/plain": [
"array([[0.19445554, 0.17727154, 0.15683166, 0.1209703 , 0.17859615],\n",
" [0.53720044, 0.5017816 , 0.4552535 , 0.43029317, 0.51731368],\n",
" [0.71183018, 0.6522259 , 0.63278148, 0.59968034, 0.67655925],\n",
" [0.71114926, 0.65166472, 0.63114387, 0.59867254, 0.67635792],\n",
" [0.71133486, 0.65186133, 0.63132631, 0.59807001, 0.6762205 ],\n",
" [0.71130616, 0.651849 , 0.63130423, 0.59809474, 0.67624577]])"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"rfe.grid_scores_"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Text(0, 0.5, 'R-squared')"
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEXCAYAAAC6baP3AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABLaklEQVR4nO3ddZwVZRfA8d/ZZYOlO5fuVFLBQEUFCzuwFRDFjhcbUCwMVAQJC1HsAhuwCWFBOpdecukNts/7xzO73C3gwjbny2c/uzPzzMwzl3vvmXlSVBVjjDHmaAUUdgaMMcYULxY4jDHG+MUChzHGGL9Y4DDGGOMXCxzGGGP8YoHDGGOMXwo0cIjIeyKyU0SW5rJdRORNEYkUkcUi0qEg82eMMebICvqJ4wOg12G29waaej8DgLcLIE/GGGP8UKCBQ1X/AvYcJkkf4EN15gAVRaRWweTOGGPM0ShV2BnIog6w2Wc5ylu3LWtCERmAeyqhTJkyHVu0aFEgGTTGmJJi/vz5u1S1mr/7FbXAITmsy3FMFFUdD4wH6NSpk0ZERORnvowxpsQRkY3Hsl9Ra1UVBYT7LNcFthZSXowxxuSgqAWOKcBNXuuqU4D9qpqtmMoYY0zhKdCiKhH5BOgBVBWRKGAIEASgqmOBH4ELgEggHri1IPNnjDHmyAo0cKjqdUfYrsCgAsqOMcaYY1DUiqqMMcYUcRY4jDHG+MUChzHGGL9Y4DDGGOMXCxzGGGP8YoHDGGOMXyxwGGOM8YsFDmOMMX6xwGGMMcYvFjiMMcb4xQKHMcYYv1jgMMYY4xcLHMYYY/xigcMYY4xfLHAYY4zxiwUOY4wxfrHAYYwxxi8WOIwxxvjFAocxxhi/WOAwxhjjFwscxhhj/GKBwxhjjF8scBhjjPGLBQ5jjDF+scBhjDHGLxY4jDHG+MUChzHGGL9Y4DDGGOMXCxzGGGP8YoHDGGOMXyxwGGOM8YsFDmOMMX6xwGGMMcYvFjiMMcb4xQKHMcYYvxR44BCRXiKySkQiReTRHLZXEJGpIrJIRJaJyK0FnUdjjDG5K9DAISKBwGigN9AKuE5EWmVJNghYrqrtgR7AqyISXJD5NMYYk7uCfuLoAkSq6jpVTQI+BfpkSaNAORERoCywB0gp2GwaY4zJTUEHjjrAZp/lKG+dr7eAlsBWYAlwn6qmZT2QiAwQkQgRiYiOjs6v/BpjjMmioAOH5LBOsyyfDywEagMnAW+JSPlsO6mOV9VOqtqpWrVqeZ1PY4wxuSjowBEFhPss18U9Wfi6FfhanUhgPdCigPJnjDHmCAo6cMwDmopIQ6/C+1pgSpY0m4BzAESkBtAcWFeguTTGGJOrUgV5MlVNEZG7gV+AQOA9VV0mIgO97WOBZ4EPRGQJrmhrsKruKsh8GmOMyV2BBg4AVf0R+DHLurE+f28FzivofBljjDk61nPcGGOMXyxwGGOM8YsFDmOMMX6xwGGMMcYvFjiMMcb4xQKHMcYYv1jgMMYY4xcLHMYYY/xigcMYY4xfLHAYY4zxiwUOY4wxfrHAYYwxxi8FPsihMQUpLjGFN2as4Y9VOwEQby4x8ZlSTLwF8Vkvkj2t+CxI1vXecQ79fWh/31+H9jl0bN91kmWqs+zHzD3P+Oyf04xp6bLOnJaxPrcNueyRW/rcj5/LcfzMT67ZPEFd1LYWV3cOP3LCPGSBw5RY05bvYMh3S9m6P4EzmlUjLCgQAPW+elQPfQm5Lyn1+dstpX/ZaZb1kPmL0B0ry/6a/Vyq6QdQ9y+Xc2XNH6pZ8krm/TXnfOVGskao9PW5pvdzfS5Hyj29fyc+XGA80SSkpBb4OS1wmBJn676DDJ2yjF+X76B5jXJ81fdkOtavXNjZMqbEsMBhSozUNGXirA28+usqUlUZ3KsF/U5vSFCgVeUZk5cscJgSYUnUfh7/ZglLtuznzGbVeLZPG+pVCSvsbBlTIlngMMVabGIKr/66iomzNlClbAhv9T2ZC9vWyrUM3xhz/CxwmGLrl2XbGfLdMnbEJHBD1/o8fH5zKpQOKuxsGVPiWeAwxc6WfQcZ8t0ypq/YQYua5RhzQwc61KuUa/rUlDQO7Dro1zn8emLJIWnuu2ff4NfDUa6tm47tCevodjtyoiMeJ48eAO1JMrtSwQEEhxbsV7kFDlNspKSm8cGsDbw2bTWq8PgFLbi1e+6V36rK2gXRzP4mkgO7Ego4t8YUjA7n1+PUy5oU6DktcJhiYdHmfTz29RKWbzvAWc2r8UyfNoRXzr3ye+fGA/zzxRq2Re6ncu0ynHVjC4KCAzOl0dy6kuWw+ii6RhwxcY5rc8uCH53x/MrbUZw787GPv7tdHhziqA6k6ufTWwlRNbxcgZ/TAocp0mISknnll1V8OGcj1cqGMOb6DvRuUzPXIou4fYnM+W4tK+dsp3TZIM7s25xW3WsRYE1yjckzFjhMkaSq/Lx0O0OnLmNnTCI3nVKfh85vTvnQnCu/U5JSWTh9E/N/2URaShon96xHxwsaEFLa3uLG5DX7VJkiJ2pvPEO+W8aMlTtpWas8427sxEnhFXNMq6pERuxk1jeRxO5JpNHJ1eh2eWMqVLM+HMbkFwscpshITk3j/ZnrGTltDQBPXtiSW7o1oFQuxUzb1+9n5hdr2L7uAFXDy9Lz5lbUaZ6lddX+KPjjRVj/16F1GcVckn35cNuOaRk/0+fxcjb+VKoUt7QnqPbXQdcBBXrKIwYOEVmPHwNSqmqj48qROSH9t2kvj3+zlBXbDtCzZXWGXtKaupVyfmqI3ZvA7G/XsvrfHZQuH8xZN7agxam1CAjw+bKM3wP/jIR/xwEKzXtDqdJkvJUzvnj8WT5SWo6wPb+Xczp/1gDiGxwzr85RrrXN/rRBLsC0J6Lggn+6Pponjq/IHDiuBcKAacBOoDpwLhAHfJrXGTQl24GEZF7+eRUf/buRGuVCGXtDR85vXSPHyu/kxFT++3Uj//26CVXo2Ks+HXrVz9yGPSke5o5zQSPhgLsbO+sxqFivAK/KmJLtiIFDVR9O/1tEHgfWAheqapzP+rLA98CB/MikKXlUlR+XbGfY1GXsik3klm4NeOi85pQNyf6W1DRl9dztzP52HXH7EmnSqTqnXtqY8lVLH0qUmgILP4Y/XoCYbdCsF5zzNNRoXYBXZcyJwd86jkHAAN+gAaCqsSLyCjABGJ5XmTMl0+Y98Tz13VL+WBVNmzrleefmTrSrWzHHtNvW7uefz1ezc2MM1euX4/x+ranVxCetKqz8HmY8A7tWQ93OcMW70KC7t1n5cf2P/LPlHwACJCDjtyDZlt3ESW59+t8iQgDesu/29G057Xuk5aM8dk55DJAAEDL2CyDLspc+pzkxcu27koPcJ17KqS+JH2n96T+T2yRSRXA6p9zmIMlv9cvXp2mlpgV6Tn8DRwWgRi7bagJljy87piRLTk3j3X/W8/r01QSK8NRFrbj51Po5Vn4f2H2Q2d+sJTJiJ2UqBNPzlpY061IT8a3H2DATpg+BqHlQtRlc8zG0uDCjTHxzzGaenf0ss7fNpmrpqoQEhqDqJlBK0zRvIqRDy0DG+jRNA4U00jLSqOa8X1H8EjMnjtva3MYDHR8o0HP6GzimAC+LyAFgqqomikgIcAnwEjA1rzNoSob5G/fyxDdLWLk9hvNa1WDoJa2pXbF0tnRJCSks+HkjC6dvRgQ6X9iAk8+rT1CIT6/vHctg+jBY8wuUqw2XjIL2fSHQvZ1T0lKYtHwSYxaOITAgkMe7Ps7Vza4mMCAw2/nyyuECUqbAkx6IjjGApeEt+273SZ/1uDndBec+O9/Rp82xDju34+Y4PtfRp81NURq3Ki962B+ryqEFP0mZv4HjTuAD4HNARSQGKId7G03xthuTYf/BZEb8vJLJczdRs3wo42/syHmta2ZLl5amrJy9jX+/W0f8gSSadanBKZc2plzl0EOJ9m2C35+HRZ9CaHnoORS63JGpVcmyXcsYOnsoK/espEd4D57o+gQ1y2Q/X17LWqxkTEnmV+BQ1f3AZSLSGuiMK7baDsxT1eX5kD9TTKkqUxdv45mpy9kTl8ht3RvywLnNcqz83rJ6L/98sYZdm2Op0bA8ve9sS82GFQ4liNsNf78K8yYAAt3ugdMegLBDd1rxyfGM+m8Uk1dOpkpoFUb2GMk59c4pUnelxpQUx9QBUFWXAcuOZV8R6QW8AQQC76jqizmk6QG8DgQBu1T1zGM5lykcm3bH8+R3S/lrdTRt61Tgg1s706ZOhWzp9kfHM+vrtaz7L5qylUI47/bWNOlU/dCXfVIczBkDM9+EpFg4qS/0eAwq1M10nL+i/mL4nOFsi9vG1c2u5v6O91MuuOAHfjPmROF34BCR6sBDQCegLnC5qi4TkfuAuao6+zD7BgKjcf0+ooB5IjLF92lFRCoCY4BeqrrJO58pBpJT05jw9zremL6GoMAAhl7cihtPbUBgQOa7/sSDKcz/cQOLft9MQGAAXS9pyEk961EqffTa1GRY8CH8+RLE7oDmF7qmtdVbZDrOroO7GDF3BD9t+IlGFRoxsddEOtToUFCXa8wJy6/AISJdcB3/ooE/gR5AiLe5Fi6gXHmYQ3QBIlV1nXe8T4E+gG8xV1/ga1XdBKCqO/3JoykcERv28Pg3S1i9I5ZerWsy5JJW1KqQufI7LU1Z/s9W5k5dx8GYZFqcWpNT+jSmTEXvLaQKy7+FGc/CnrVQ71S4+kOod0qm46gq30R+wysRr5CQksBdJ93F7W1uJzgwOMe8JaxaTdzsWYBXoSpCRg9qEVdD5zWHzTb8SLZtmffLtE+m7T7bsqxPT599e5ZzZd0vx31yOFf6tozLyDIcSdb8Zlrnc5709Vn3I4e0uZ3jqPOROT+HdvczH4WkMCvHA8LCCCxbsA1a/X3iGAn8DlwOBAC3+mybi/vSP5w6wGaf5Siga5Y0zYAgEfkDV/H+hqp+mPVAIjIAGABQr571Ci4s++OTefHnFXwydzN1KpbmnZs60bNV9hbbm1fuYeYXa9i9JY5aTSpw0d1NqV6//KEE6/+CaUNg6wKo1hKu+9R14svyZbBh/waGzR5GxI4IOlTvwJBuQ2hUIedRblJjYoh+cxR7P/4Y0tLy9LqNKSqq9O9H9YceKtBz+hs4OgB9VDVNsof33bjhRw4np1uCrKG6FNAROAcoDcwWkTmqujrTTqrjgfEAnTp1sob0BUxVmbJoK89+v5y98ckMOKMR953TlDJZKr/37Yhn5leRbFi8i/JVQzm/fxsad6h26O5w22KYPhTWzoDydaHPGGh/LWRpOpucmsx7S99j/OLxhASGMOTUIVze9PIcWzGpKge+/54dI0aQums3Fa+9hqoD7ySgdKh7qvGaw3qJD43t5LtNAfTQ9oxtPut99z/MfpnW57ZP+lhYOe53mGNm2S/TPjmNr+X91kzX7XOOLGlVs64j83LW1yL92NnSHiYf6WmznotjyEehNoYonHOHtmxx5ER5zN/AsR+olsu2RsCOI+wfBYT7LNcFtuaQZpfXOz1ORP4C2gOrMUXChl1xPPntUv6J3EX78IpMvK0NrWtnrvxOiEsm4scNLPk9isDgAE69rDHtzq5LqSAvIOxZD78/B0u+gNKV4Lzh0Lk/BIVmO9/CnQsZNnsYkfsiOb/B+QzuPJhqYTm/DRMjI9n+zLPEz51LaLt2hL89ltJtbNgRY/KSv4HjO2CYiMwGNnrrVESqAg8DXx9h/3lAUxFpCGzBDZiYtXjrO+AtESkFBOOKskb6mU+TD5JS0hj/11re/C2SkMAAnunTmuu71s9U+Z2Wmsayv7cyd+p6EuKTadW9Nl0vaURYea/+ITYa/noZIt6DgFJw2oPQ/T4oXTHb+WKTYnl9wet8vupzapSpwaizR9EjvEeOeUuLiyN6zBj2TPyQgDJlqDlsGBWvuhIJsH4VxuQ1fwPHo8AMXGX2fG/dWKAJsB54+nA7q2qKiNwN/IJrjvue1yJroLd9rKquEJGfgcVAGq7J7lI/82ny2Nz1rvI7cmcsF7atxdMXt6JG+cxPBxuX7Wbml5Hs3RZHneYVOe2qplSt6zWLTYyB2aNh1ihIPggdboQzH4XytXI834xNM3j+3+eJjo+mb8u+3HPyPZQJKpMtnaoS88uv7HjxRVK2b6fClVdQ/aGHKFWpUp6/BsYYR/xtDSAiwcCNuDqIqsAeXDD5UFUT8zyHR6FTp04aERFRGKcu8fbFJ/HCjyv5LMJVfg+/tA1ntchclbVnWxwzv4xk07LdVKhWmm5XNKFh+6quHiMlCRZMdE1r46Kh5SWuaW3VnAdl2xG3gxfmvsCMTTNoVqkZQ08dSttqbXNMm7h+PTuGP0fczJmEtGxJzaefIuzkk/P8NTCmpBKR+arayd/9jvqJQ0RCccOKPK+q7wLv+nsyU3yoKt8u3MLw71ew72Ayd5zpKr/Dgg+9ZRJik5n7/XqW/rWFoJBAul/ZhLY96hJYKsC1Ylr6Ffz2LOzdAPVPg2s/gfDOOZ4vTdP4YtUXvL7gdZLTkrm/w/3c1PomggKyzzGedvAgu8aNY8+77yEhIdR44gkqXXctUsomtDSmIBz1J01VE0SkM66IyZRg66Jjeeq7pcyM3M1J4RX56PK2tKx1qOlsakoaS//cwrwf1pN0MIXWZ9Shy0UNKV3Oq8dY+5trWrt9MdRoA9d/CU165triJXJvJMNmD2Nh9EK61urK06c8Tb3yOTexjvntN3YMf47krVspf8nF1HjkEUpVy629hjEmPxzL6LiX4oqmTAmTmJLKuD/X8dbvkYSUCmD4pW3o26VexpSsqsrGJbuZ+VUk+3bEE96qMt2vaEKVOl7noy0LXNPa9X9ChXpw2XhoexXkUkGdmJrIhMUTeHfpu5QJKsPw7sO5pPElOXbkStq8mR3PPU/sH38Q0rQJ9T6cSJkuXfLrpTDGHIa/geMX3LDqtYAfcc1vM1WSqOqPeZQ3U4DmrNvN498sYV10HBe1q8XTF7Wiuk/l9+4tsfzzxRqiVu6lYo0wLhzUjvptqrgv+d1r4bfhsOxrKF0Zzn8BOt8OpUJyPd+87fN4ZvYzbDiwgYsaXcQjnR/JcXjotMREdr/zDrvHT4DAQKo/8giVb7oRCcpehGWMKRj+Bo6PvN+Xez9ZKVaUVazsiUvi+R9X8OX8KMIrl+aDWzvTo/mhyu+DMUn8O3U9y//eQnDpUpx+TVNan1GHwMAAiNkBf42A+R9AYDCc8YgbuTY0+4CG6fYn7mfk/JF8teYr6pStw9ieY+lep3uOaWP//pvtw4eTvHET5Xr3osbgwQTVzP8h0o0xh+dv4GiYL7kwBU5V+WrBFp77YTkxCSnc1aMx95zdlNLeQIOpyWks/j2KiB/Xk5KURtsedel8UUNCywRBwgH4c5RrXpuaCB1uhjMHQ7ncJod05/tl4y+8+O+L7Evcx62tb2Vg+4GEBYVlS5u8bRs7nn+BmGnTCG7QgPB336Fs95yDizGm4Pk7H8fGI6cyRd3a6Fie+GYJc9btoWP9Sjx/WVua13T9LVSV9Qt3MfPrSA5EH6R+2yp0v6IJlWqWgZREmPO268AXvxtaXwZnPwVVGh/2fNtitzH83+H8FfUXraq04u2eb9OySsts6TQpid0TJ7JrzNugSrX776fybbcSEJzz4IXGmMJxTO0XvV7d9YBs40PYhE5F25RFW3n480WEBgXw/GVtubZzeEbld/TmGGZ+sYYtq/dRuXYZLr63PfVaVXFNaxd9Br8Pd7PwNTwDeg6DOocfwjw1LZVPVn7Cm/+9CcDDnR7m+pbXUyog+9subs6/bH/2WZLWrqVsz3Oo8ehjBNetk/cvgDHmuPk7rHoQ8CZwM4eGU8/K6jiKqIWb9/HwF4toH16BMdd3pFo5918Ytz+Rf6esY8WsbYSWCeLMvs1p1b2WCyhrprn5vXcsgZrt4IbXofHZRxxMbtWeVQydNZSlu5fSvU53njrlKeqUzR4IknfuZOdLIzjwww8EhYdTd+zblOvRIx+u3hiTV/x94ngauAi4HfgYGATEATcAjYF78jR3Js/sPJDAHZMiqF4uhHE3dqJymWBSklNZNGMz83/aSGpKGiedE06nCxoQEhYEUfNh+hDY8DdUrA9XvAutL8+1aW26gykHGbtoLBOXTaRCSAVGnDGCXg16ZWtiqykp7P34Y6LfHIUmJ1N10CCq9O9HQGj2QQ6NMUWLv4HjamAo8DkucMxV1fnAhyIyETcpkzXHLWISklMZMGk+MQkpfH1XNyqFBbEmYgezv15LzJ4EGravSrcrmlCxehjsWgNTn4EVUyCsKvR+GTreAqWOXM8we+tsnpn9DFGxUVzW5DIe6vQQFUKyt7CKX7CA7cOeIXHVKsqccTo1n3iC4Pr18+HKjTH5wd/AEQ6sVtVUEUkAfEeS+xiYDNyRV5kzx09VeeKbpSzcvI+xN3SkVmApvnllAdvW7qdK3bL0uflk6javBAe2wdTHYMEkCCrt5vY+dRCEHHnu7r0Je3kl4hWmrJ1CvXL1ePe8d+lSK3vnvJTdu9n58ivs//ZbStWqRZ1Rb1KuZ89CnbnNGOM/fwPHNqCi9/d64Axgurd8+KY1plC8+896vloQxf09m9KjYRW+fDGChLhkzrqhBS261SIg6QDMeAZmj4G0FOjcz/XHKHvkYTxUle/Xfc/L814mJimG/m37M6DdAEJLZS5u0tRU9n72GdGvv0HawYNU6d+fqncOJCAse1NcY0zR52/g+AM4HZgKTABeEZEmQCJwDfBJnubOHJc/V0fz/I8r6N2mJnef2ZgfRi8mZk8Clz5wMrXqh8Kct+DvV+HgXmhzJZz9BFTOeRrWrDbHbObZ2c8ye9ts2lVtx5BuQ2hWqVm2dAcXL2b7sGdIWLaMsFNOoebTTxHS6OjOYYwpmvwNHE/ghlJHVV/3po+9EjfF6yjgmbzNnjlW63fFcc/kBTSrUY5XrmrPrK/XErVyL2ff2JxasT/CqOfgQJRrIdVzKNRqf1THTUlLYdLySYxZOIbAgEAe7/o4Vze7msAsU72m7N1L9MjX2ffFF5SqWpXar75C+QsusGIpY0oAfzsAbge2+yyPxGbnK3IOJCTTb+I8AgOECTd1YsPcHSz5PYr254TTMvp5mPYR1DoJLh0NjXoc9XGX7VrG0NlDWblnJT3Ce/BE1yeoWSbzECCalsa+r74i+tXXSI2JofJNN1H1nrsJLFs2by/SGFNobAKDEiY1Tbn/04Vs3B3PpNu7ErArkb8+WU291pXpVvVb+PMjOP0hOOvJIzatTRefHM9bC9/i4xUfUyW0CiN7jOSceudke3pIWL6c7cOe4eCiRZTu2JGaTz9NaPPsxVfGmOLN3w6A0WQZDTcrVa1+uO0mf73y6yp+W7mTZ/u0plX5ML58MYIK1Utz3imRBPwwHNpd44YJOcoio7+i/uK5Oc+xNW4rVze7mvs73k+54MwtrVIPHCD6jTfZ+8knBFaqRK0XX6BCnz5WLGVMCeXvE8dosgeOysDZQHlsVsBC9d3CLbz9x1qu61KPq9vX4etXFqCqXNAnhZDv74T63eGSUUcVNHYd3MWIuSP4acNPNKrQiIm9JtKhRuYhRlSVA1OmsOPlV0jds4dK115LtfvvI7B8+VyOaowpCfyt4xia03qvkvxzICUP8mSOwZKo/fzvy8V0aVCZIRe1YvqEpezdHs/FN1ej4q8Xud7f13x02DkywAWDbyK/4ZWIV0hISeCuk+7i9ja3ExyYuQNgwurV7HjmWeIjIght147wcWMp3bp1fl6iMaaIyJM6DlVVEXkHeB94Pi+OaY7ezpgEBkyKoGrZEMbc0IH/fljPhiW7OeOyOoT/ey1IAFz/OYRlnyjJ14b9Gxg2exgROyLoUL0DQ7oNoVGFzE1nU2Pj2DV6NHs+/JDAsmWp+cwwKl55JXKU9SXGmOIvLyvHGwE2/nUBS0xJ5c6PFrAvPpkv7zyV3cv2suCXTbQ+rQZtNt8D+7fALd8ftn9Gcmoy7y19j/GLxxMSGMKQU4dwedPLCZBDwUBVifn5Z3a88CIpO3dS8aqrqPbgA5SqVCnX4xpjSiZ/K8fvymF1MNASuB74Ii8yZY6OqvLUt0uZv3Evo/t2oEoifDtpJXWaVuT0oFeRyH/hqg8gPPe5uRfuXMiw2cOI3BfJ+Q3OZ3DnwVQLy9xrPHHdenYMf5a4WbMJadWSuqPepHT7o+v3YYwpefx94ngrh3WJQBQwBhh23DkyR+2DWRv4PCKKe85uwpnhlfjihQjKVAymV4ufCJz3levY1/qyHPeNTYrl9QWv8/mqz6keVp1RZ4+iR3iPTGnSDh5k19tj2f3++wSEhlLjqSepdO21SKCNnG/MiczfynEryC4iZkbuYvgPKzi3VQ3uPqMx3732H8lJqVxy7kZCZ45w07l2vz/HfWdsmsHz/z5PdHw0fVv25Z6T76FMUJmM7apK7IwZbH/+eVK2bqNCnz5Uf+RhSlWtWkBXZ4wpyqwDYDG0cXccd328gMbVyvDa1e35Y9JKojfHcOFlUGXWPW4YkQtfzdbsVlUZOX8k7y97n2aVmvF6j9dpW61tpjRJmzax/bnniPvzL0KaNqXOR5MI69SpIC/PGFPE+VvHcZM/6VX1Q/+yY44kNjGF/h9GIAITburEyhlRRM7fyannlqPB/MuhWgu4aiIEBmXbd8KSCby/7H2uanYVj3V9jKCAQ2nSEhPZPeEddo8fj5QqRfXBg6l8w/VIUPbjGGNObP4+cXzAoQ6Avrezua2zwJGH0tKUBz5byNroOD68rQspG+OYO3U9zTtU4OSNN0NQmGt2G5q9A97kFZMZ9d8oLmp0EU+e8mSmFlOxf/7J9uHPkbx5M+Uv6E31wYMJqlGjIC/NGFOM+Bs4OuM6+r0LfA3sBKoDVwC34WYIXJmXGTSHjJy+mmnLdzDk4lY0Cwnh6w/mU6NBWXqkPoIc3A23/gQV6mbbb8raKbww9wXOCj+LZ7s/mxE0krdsYfsLLxA7fQbBDRtS7/33KHPqqQV9WcaYYsbfwPEqMFpVX/NZtwd4zpsR8DVVPTPPcmcy/LB4G6N+i+SaTuFc3aY2X74YQWhYEL1rjaXUhv/g2slQ+6Rs+83YNIOnZz5N15pdefnMlykVUApNSmL3+x+w6+23QYRqDz5IlVtuRoKtG44x5sj8DRxdgBdy2bYUePb4smNysmzrfh7+YhEd61diyIUt+WnUIhJik7m820zKrPoKeo+A5r2z7Tdn2xwe+fMRWldpzRtnv0FIYAgJK1aw5cGHSFq/nnLn9qTGY48RVLt2IVyVMaa48jdwbAZuBX7JYdvtuP4cJg/tik1kwIfzqRgWxNvXd2D252vYvu4A5/fYRrWVI6DrndA1+zTvC3cu5N7f7qV++fqM6TmGMkFlSNq4kU2390OCgggfP46yZ5xRCFdkjCnu/A0cjwOfishSYAqH6jguAVrgpo81eSQpJY27PlrArthEvhzYjS3/7mTl7O10PiWFJqvuhuYXwvnPZdtv1Z5V3DXjLqqWrsr4c8dTIaQCKdHRbOrXH9LSqPf++4Q0algIV2SMKQn87QD4lYh0BR4FrgNq4mYEnAfcrKrz8z6LJyZVZciUZczdsIc3rj2JcnuT+fHrSBq3DKJz1A1uqtcrJkCWKVs3HtjIHdPuoHSp0kw4bwLVwqqRGhvLpgF3kLJrF/U/sKBhjDk+fncAVNUFuNZTJh99NGcjn8zdxJ09GnN6jYp89VIEVWqFcE7iQKRMZbjuMwguk2mf7XHb6f9rf9I0jffOfY86ZeuQlpRE1N33kLhmDeFvj7Expowxx+24hxARkRYicqmIHFUNq4j0EpFVIhIpIo8eJl1nEUkVkSuPN4/Fzey1uxk2dTlnt6jO3d0a8cOYxQQGCReUH05Q6j7o+zmUy9zPYk/CHvr/2p+YpBjGnjuWRhUboampbP3fYOLnzKH2c8Mpe/rphXNBxpgSxa/AISLjRGSsz/I1uNZUXwMrRaTbEfYPxM0i2BtoBVwnIq1ySfcSOVfCl2ib98Rz18fzqV8ljNeuase0d5cRuzeBCxp8QrkDc+HqiVAj80sWkxTDwGkD2Ra3jbfOeYtWVVqhqux47nlifv6Z6o88QoU+fQrpiowxJY2/Txy9gL98lp8FJgO1cV/yR2qO2wWIVNV1qpoEfArk9I12D/AVrvL9hBHnDSeSmqa8c3NnFk3ZwJZVezmrVQQ1oyfDRSPdOFQ+DqYc5O4Zd7Nm7xpG9hhJxxodAdg9bhx7J0+m8q23UuX22wrjcowxJZS/gaM6rkkuItIUaAKMUNXtwHjg5CPsXyd9f0+Uty6DiNQBLgPGchgiMkBEIkQkIjo62q+LKIrS0pSHPl/E6h0xvNW3A3HL9rL0zy2c3GonLXYOh9MehA6ZhwpLTk3mgT8e4L+d//HCGS9wel1XFLX3iy+Ifv0Nyl9yMdUfebgwLscYU4L5Gzj2AOmF6z2B7aq61FsW4EgTNUgO6zTL8uvAYFVNPdyBVHW8qnZS1U7VqlU7XNJi4c3f1vDzsu08fkFLGmkp/vpsDfXrJ3PK7juh9eVw9lOZ0qempTL478HM3DKTIacOoVeDXgDE/PYb24cMpcxpp1H7uedsSldjTJ7zt1XVT8AzIlID+B9u3Kp0bYANR9g/Cgj3Wa4LbM2SphOurwhAVeACEUlR1W/9zGux8fPSbbw+fQ1XdKjLlc1r8uVLEVSsLJyXNICAel3g0rchIPM0rsNmD2Paxmk83Olhrmh2BQDx8+ez5YEHCW3dmrpvvG4j2xpj8oW/t6MPAXOAgbi6jiE+2y4Dfj7C/vOApiLSUESCgWtxHQkzqGpDVW2gqg2AL4G7SnLQWLn9AA9+voiTwisypFcLfhyzGNLSuLD0YIIrVXZjUAWFZqRXVV6OeJlvIr/hjnZ3cHPrmwFIWL2azXfeRVCtWoSPG0tAmTK5ndIYY46Lvx0A9+NGwc1p2xHbeqpqiojcjatIDwTeU9VlIjLQ237Yeo2SZk9cEv0mRlAutBTjru/AXx+uZP/OeC6uN44KsgOunwFlqmTaZ9zicUxaPom+Lfoy6KRBACRv3crm/gMICAkh/J13KFW5cmFcjjHmBHHMMwCKK0t6ChjvVY4fFVX9Efgxy7ocA4aq3nKs+SvqklPTuOvj+eyMSeSLO05l7YwtbFy6mzMbTKNu8m9w03dQpXGmfT5e8TGjF47mksaXMLjLYESElL172dSvP2nx8dT/aBLBdevkckZjjMkbx1NzGoArqrKhVY/BM1OXM2fdHl66oi0hUQf5b9om2oSvoU3CGOgzBupn7hLzXeR3vDj3Rc4OP5th3YYRIAGkxccTNfBOkqOiqDv6LUKbNy+kqzHGnEiOt8lNTq2kzBFM/ncTk+ZsZMAZjTi1Qjl+/3gldarv57Skx+CsJ6HdVZnST984nadnPc0ptU45NKdGcjJbHniQg0uWUPuVlynTpUshXY0x5kRjbTUL2Nz1e3j6u6Wc2awag7o04MexSygblkwv7ibw5GvhjMz9LmZtmcUjfz1C26pteeOsNwgODEZV2fbU08T++Sc1n36a8uedV0hXY4w5ER1P4EgDhuE1pxWRs0TkpzzJVQkVtTeeOz+aT73KYbx2eTt+HreE1IQkLgx5hNDGHeCi10EOPcQt3LmQ+/+4n0YVGjH6nNGEBYUBEP3aa+z/9luq3n03la61keyNMQXrqAKHiFQUkWtF5BERuUJEgtQZBpwuIhHADMDG685FfFIKAz6cT1JKGuNv7Mi8z9awKyqWcyu+RuUaIXD1JCh1aOrWVXtWcdf0u6hWuhrjzh1HhZAKAOyZOJHdE96h4rXXUHXQXYV1OcaYE9gRW1WJSFvgVw71GAdYICJX4MapOgVYDlwPfJYfmSzuVJVHvljMiu0HeO+WzuyZG83a/6LpVu1bGpRb6Zrdlq6YkX7D/g0MmDaAsKAwJpw3gaqlqwKw//sf2PHCi5Q791xqPvUUIlbFZIwpeEfzxPE8cAA4FQgDWuKGHpmH6y1+s6q2VdVPVDUt33JajI3+PZIflmzj0V4tCI+FeT9soEWVRZwU/LmbV6NS/Yy022K30X9afwDGnzee2mVdo7XYmTPZ+thjhHXqRO1XXkYCjzS6izHG5I+jCRydgKdU9V9VTVDVVcCduOFAHlLVj/I1h8XctOU7eOXX1Vx2ch0ua1CNGR8sp2b5HfQo9Rxy5QSo2zEj7e6DuxkwbQBxSXGM7TmWRhUaAXBwyVK23HMvIY0aUXfMaAJCQgrrcowx5qgCRw2yj0GVvrwoLzNT0qzeEcP9n/5Hu7oVeKpnc34au4TQUvH0DhlM4PlDoeXFGWkPJB1g4PSBbI/bzuieo2lZpSUASRs2sPmOOwisVInw8eMJLF++kK7GGGOco21VlXUE23QpeZWRkmavN5xIWEgp3r62A7+9s4yEmINcEPYYYadcBacOykgbnxzPoOmDiNwXyetnvc7J1d3o9CnR0Wzq1x9UCX9nAkE1qhfW5RhjTIajHXLkFxHJKUjMyLpeVU/4b7eU1DTu/mQB2/cn8MmArqz4fgM71h+gV8VXqdaqGfR6KaPZbVJqEg/88QCLdy3m5TNepnud7gCkxsSwqf8AUvbsof4H7xPS0BqsGWOKhqMJHMPyPRclzPAfVjAzcjcvX9kOWRXDqjnb6VLhKxo3iIErv4BA97KnpKXw6N+PMmvrLJ7p9gznNXAd+dISE4kadDeJkZGEv/02pdu1K8zLMcaYTI4YOLy+GuYofT5vMx/M2sBt3RvSObg0P3yzmCZl59Op2m/QdwaElAUgTdMYOmso0zZO43+d/8dlTS8DQFNT2fq/wcTPnUvtl0dQ9vTTCvNyjDEmGxtyJA/N37iHJ75dwulNqzLwpHB+fXcp1UK3cnbFMcgNn0N517RWVXl53st8t/Y77mx/Jze2ujFj/Y7nnifml1+oPngwFS6++HCnM8aYQmGBI49s3XeQOyYtoHbF0rxySRt+GbeEoLQYLig3lKBrxkPNthlpxy4ay0crPuKGljdwZ/s7M9bvHjuWvZMnU/n226hy6y2FcBXGGHNkFjjyQEJyKndMmk9Ccirjr+/A7EmriNsTR+9ywyh78WPQ9NyMtJOWT2LMojFc2uRSHun8SEbv772ff070G29Soc8lVH/oocK6FGOMOSILHMdJVfnfl4tZunU/r1/Tnm2/b2PL6n2cVW4UNXv0gs63Z6T9Zs03jJg3gnPrn8uQU4cQIO7lj5k+ne1Dh1HmjNOpNXw4EmD/LcaYosu+oY7T2D/XMWXRVh4+rzk1dqaw7O+tdCjzNc07VICez2Sk+3XDrwydPZRutbvx4ukvUirAtUuIj4hgy0MPE9qmDXVffx0JCiqsSzHGmKNigeM4/LZyByN+WcnF7WtzSc1K/P3ZahqELqBr85Vw+XjwnhxmbpnJ4L8H065qO0b2GElwoBsFN2HVajbfNYig2rUJHzeWgLCwwrwcY4w5KhY4jlHkzhju+2QhrWuX54nTm/DL+CVUKrWFc8M/JaDvZAgqDcCCHQu4//f7aVKxCaN7HppTI3nLFjb3709AaCj13plAqUqVCvNyjDHmqB1tz3HjY398Mv0mRhASFMDoq07itzGLkaQYLqj5OsE3TYayrvP8it0rGDRjEDXL1GRsz7GUD3bjTKXs3cumfv1JO3iQ+h99RFCdOoV5OcYY4xcLHH5KH05ky76DTL69K4s/j2R/dDyXVB5BhRtGQrXmAKzfv56B0wdSLrgcE86bQJXSVQBIi49n88CBJG/ZQr333iW0ebPCvBxjjPGbBQ4/vfjTSv5es4sXL29L0oI9bFq+lx7lx1Hnqjuh4RmAm1NjwLQBAIw/dzw1y9QEQJOTibr/fhKWLKXum28Q1qlToV2HMcYcK6vj8MNX86N455/13HxqfdqnlGLh9M20DfuB1r07wEl9Adh1cBf9p/UnLimO8eeOp0GFBoBrtrvtyaeI++tvag4ZQrmePQvxSowx5thZ4DhK/23ay2PfLOHURlXo17IOf3y0grrBCzmtWyz0eAyA/Yn7uWPaHeyM38mYnmNoXrl5xv7Rr77K/u++o+q991DpmqsL6zKMMea4WVHVUdhxIIE7Js2nRvkQXu7dil9HRlBOtnN+238IuPQTEHFzaswYxPr963nrnLc4qfpJGfvv/uADdr/zLpX6XkfVO+/M/UTGGFMMWOA4goTkVAZMmk9cYgrv33AK/7y7iNSEeC5sNInQ6ydDqRCSUpO47/f7WLJrCa+e+SrdanfL2H//1KnsfPElyp13HjWeeCJjiBFjjCmurKjqMFSVx79ewqLN+3jt6vZs+n4te7Yncl71cVS6bQyEVSYlLYX//fU/5mybwzPdnqFn/UN1F7H/zGTrY48T1qULtV8egQQGFuLVGGNM3rDAcRjv/L2er//bwoPnNqNiZAzrluynW4WPqH/bk1C5EWmaxpBZQ5ixaQaPdnmUPk36ZOx7cMkSou69l5AmTag7+i0CQkIK8UqMMSbvWODIxZ+ro3nhpxVc0LYmvcqVJeLnKFqWnk77G/tAva6oKi/NfYkpa6cw6KRBXN/y+ox9E9evZ/OAOyhVqRLh48cRWK5cIV6JMcbkLQscOVgXHcvdkxfQvGZ5Hu3aiN8+WEqtoBWceWl1pI2bqW/0wtFMXjmZm1rdxB3t7sjYN3nnTjb36w8i1Hv3HYKqn/BTsBtjShirHM/iQEIy/T6MICgwgFGXtuW3N+dRmt30OmM9gWe8DMDEZRMZt3gclze9nIc7PZxR4Z0aE8PmAXeQsncv9SdOJLhBg0K8EmOMyR8WOHykpin3fvIfm3bHM+nmzvz3wQIS45O5ov0Mwi4fAyJ8tforXol4hfPqn8fTpzydETTSEhOJGnQ3iZGRhI8dS+m2bQr5aowxJn9Y4PAx4peV/LEqmmf7tCZ2xhp2boPe9b+i6i2vQWAQP2/4mWGzh9G9TndePP1FAgNcKylNTWXrI/8jfu5car/8MmVP617IV2KMMfmnwOs4RKSXiKwSkUgReTSH7deLyGLvZ5aItC+IfH373xbG/bmO67vWo3V0LKsXJ9C18nc0unMohFbg76i/eezvxzi5+smM7DGSoEA34ZKqsv3ZZ4n59VdqPPYoFS6+qCCya4wxhaZAA4eIBAKjgd5AK+A6EWmVJdl64ExVbQc8C4zP73wtjtrH4K8W07VhZW6pU4HZP+6gadgsOg66GSqGM3/HfB7840GaVmzKW+e8RelSpTP23TVmDPs+/Ywq/W6n8s0353dWjTGm0BX0E0cXIFJV16lqEvAp0Mc3garOUtW93uIcoG5+ZmjngQQGfDifqmVDeL5HU36fuJzqQWs5+/aOSJ0OLN+9nLtn3E2tsrUYe+5YygUfalq797PP2TXqLSpceinVHnooP7NpjDFFRkEHjjrAZp/lKG9dbm4Hfsppg4gMEJEIEYmIjo4+pswkpqQy8KP57D+YzOgr2jNz3L8EaSy9LwugVOverNu/joHTBlI+uDzjzx1P5dDKGfsemDaN7cOGUebMM6j17DM2lIgx5oRR0IEjp29XzTGhyFm4wDE4p+2qOl5VO6lqp2rVqh1TZr5ZsIUFm/bx6hXtWP/RHOLjA7ngtNWUPbsfW2K30P/X/gRIAOPPOzSnBkD8vHlsfehhQtu2oe7IkUhQ0DGd3xhjiqOCblUVBYT7LNcFtmZNJCLtgHeA3qq6O78yc03ncBpXK0PsDwtYv6M057b4jRrXDiM6Ppr+v/YnISWB93u9T/3y9TP2SVi1ms13DSKobl3Cx44lICwsv7JnjDFFUkE/ccwDmopIQxEJBq4FpvgmEJF6wNfAjaq6Oj8zIyKELFrD8qWBdKjxN83ueoz9ybEMmDaAXQd3MabnGJpVOjS1a1LUFjb360dAWBj13plAqUqV8jN7xhhTJBXoE4eqpojI3cAvQCDwnqouE5GB3vaxwNNAFWCMV2+Qoqr5Msfq5nmr+OfnWBqWXcopDw4gXoS7pt3FxgMbGdNzDO2rHWoJnLJ3L5v79SMtMZH6H00iqHbt/MiSMcYUeQXeAVBVfwR+zLJurM/f/YB+BZGXssnraBi2lHPu7U1S2crcO2MQy3Yv49Uer3JKrVMy0qXFx7P5joEkb9tGvffeJbRZs8Mc1RhjSrYTuud4pW696d3xdFKCQnnwjwf5d9u/PH/a85xT75yMNJqcTNR995OwdCl1R71JWMeOhZhjY4wpfCd04ABICw7j6X+e5PfNv/N418e5uPHFGds0LY1tTz5J3N9/U/PZZyh3zjmHOZIxxpwYTuhh1VWVF/59ganrpnLvyfdyXYvrMm3f+eqr7P9uCtXuu5dKV11VSLk0xpii5YQOHN9Gfsunqz7l1ta30q9t5mqV3e+9z55336NS375UGTiwkHJojDFFzwldVHVBowtITkvmqmZXZer5vX/KFHaOGEG5Xr2o8cTj1ivcGGN8nNCBIyQwhKubX51pXezf/7D18ScI69qV2iNeQgIDCyl3xhhTNJ3QRVVZHVy8mKj77iOkaVPqvjWKgODgws6SMcYUORY4PInr17P5joGUqlyZeuPHEViu3JF3MsaYE5AFDiB5x042394PRKj37juUOsZBE40x5kRwQtdxAKQeOMDmAQNI3bePeh9+SHD9+kfeyRhjTmAndOBIS0wk6q5BJK5bR/jYtyndpnVhZ8kYY4q8EzpwHJg6lfiICGq/+gplu3cv7OwYY0yxcEIHjgpXXEFIs2aUbteusLNijDHFxgldOS4iFjSMMcZPJ3TgMMYY4z8LHMYYY/xigcMYY4xfLHAYY4zxiwUOY4wxfrHAYYwxxi8WOIwxxvjFAocxxhi/WOAwxhjjFwscxhhj/GKBwxhjjF8scBhjjPGLBQ5jjDF+scBhjDHGLxY4jDHG+MUChzHGGL9Y4DDGGOMXCxzGGGP8YoHDGGOMXyxwGGOM8YsFDmOMMX4p8MAhIr1EZJWIRIrIozlsFxF509u+WEQ6FHQejTHG5K5AA4eIBAKjgd5AK+A6EWmVJVlvoKn3MwB4uyDzaIwx5vAK+omjCxCpqutUNQn4FOiTJU0f4EN15gAVRaRWAefTGGNMLkoV8PnqAJt9lqOArkeRpg6wzTeRiAzAPZEAxIrIqmPMU1Vg1zHuW1zZNZ8Y7JpPDMdzzfWPZaeCDhySwzo9hjSo6nhg/HFnSCRCVTsd73GKE7vmE4Nd84mhMK65oIuqooBwn+W6wNZjSGOMMaaQFHTgmAc0FZGGIhIMXAtMyZJmCnCT17rqFGC/qm7LeiBjjDGFo0CLqlQ1RUTuBn4BAoH3VHWZiAz0to8FfgQuACKBeODWfM7WcRd3FUN2zScGu+YTQ4Ffs6hmqz4wxhhjcmU9x40xxvjFAocxxhi/WOA4CiKSUxNhY4wp9kTE7zhggSMHIlJaRB4VkZEAahVBRyQigRZgD09EQgo7Dyc6EQkXkR9EZFBh56WoUNU08O8G2QKHR0Sq+iwmArcDN4rI2SJS2ktjX4y5UNVUVVURCbPXKTMRKSUi3XFD7hRmPkJF5D4Rubgw81HQRKSxz2IN4Hyg+4n0OohIkIgE5bKtjYg8jXttjsoJHTi8D9JtIvIjMMN7ymjsReDfcM2GzwVuSd+lkLJaZHj9awJ9lgO8dX1E5C9gInBz4eWwaPBelwBwzdCBesCtIvK2iDwgImUKKB+hIvKMd2OUAlwHLC+IcxcmEWkpIm+JyEJgrIi85G1aDSwEfgfuKMmjb3ufy/Rx/q4BTvHWlxWRU0QkvaN1IHCdqm4/2mMX9JAjRYKIhKpqAvAsbjTe+4B1wBPA3SLyMG6Yk7FAKjBKRL5U1ejCynNR4RXbpQKISAtVXSkiTXB3cf2Ag0CEiESp6vRCzGqhEJEAVU3zefwPUdVE3AgI1wE/Ac+qalw+5qEacDHwi6puEZFSuIC+HvgbWCcipbyAVmKISCVV3esVCb6AC5QXAjHAUhGZBqwE/gO+A8rhgsdwVd0sIlISiqV9rqOft2oC7iFhvIj8BsQB5YEUEXkE2AQsE5E6qrrlaM5xwgQOEekC3AQ0Az4H3gGm4+4EZ6pqgoh8Azyoqqki0g54WFVjvbvD50VkhKquKSlvsCPxuWNOS79mr9juaeA8YL2IXA90AoKBO3B3Nf8CR/UGLO68YrkA7z0T4BMw2uO+vNJE5DlcJ60ywFZVzfMhdLx8CPAYcDkwG2gsIitV9XER6Yqb0qCcqj6C+1It9kQkDLgBuAxQERmtqj+IyN+492Syqh4QkalAL1zwrK2qO0XkP2AQkCwi9xT3z7T3edX061DVCSJSU0RCgTNwRVHrVfUV7/3yBa40JQo3sOxR3xiX+KIq73HtMdy8HpuA/wE7RaQ8MA03smRrL3lnXPEUuKi8SESWARtwH7RXoORXlqfXUaTfOYtIWZ9rPhX3ml2uqld6d9NhwEm4EYz7qOolqrrCt0irpPKG/0/1/k4TkYoisgK4EfgAd5PyJC5oLABae0ElT4jIGSKSfiPUAyijqh1xXwpn474sUdV/cSOoLheRL0XkzNzKvIsLESmHe417Ai8B93JolNi/cXP6VBSRikAo8DPuS/Ic7//oTW//DsBg78ms2BCREBHpmb7sfV5VRBqIq1MDd803qmo/4DO8Gwbv8/wCUB0XeDt4U10cHVUtcT+4gBjo/d0AmAWUzSXtcFx9xmzvRT4Z9xj3gbetok/aJUD3wr6+fHi90u+afdcF4p4qpnqvzcXe+reBV72/S3u/2wKf4IIJQAvgdaBOYV9bHr9OgTm8Tg2BYcDH3ocP4Btgovd3GPAa8ABQGXgLuMjbVuE483OS996+BijrfQnsxwWoH3BP2KW8tDVxxVX1ORTUzirs1/QY3qeBPsvXAt8fJv1Y7/VZBLzhfa6b4IL5DT7pegDvAS0L+xr9fD1ae/+X6d91zXE3DCuBB7x1twA/eX9fgitd8T1GVe81+hmoetTnLuyLL4AX9zZgBBDCoSFWKnovYjVc0VUEcH6W/RYBbby/Q73fOQafkvYDVAFGATNxd2N9cWXC7YBLgU9z2OcC7037O67y8SV8gm5x/Ul/z2RZF+x9iVUGJgEDcUUlk3B3vxcBv3lpA73lqd7yzd5rtNrbJ+A48tYTNxlaL6CN9zMVuMknTVmgEW6CtJneuowbq+Lyk8v/w6PAPVnWdfO+LMOAK4CvgKY+29sDc31eh4Ccjl0Uf3J6r+Dqznp6f/cHRmXZHoqrv23g/b0YF2AEqOul6Qd8nNs5cvopVo9mR0tE2uKKB0KBICBJVRPF9TXoAnyIe2Qboarvi0gUkN4KJkBdpeEi3ARSS9VVpKOqsYVxPXnJqzj8BHhdVf/y1rXHvXkScLMvLhGRbbhAuUBEluKaknYB/gQaiUhvVf3Jq+PYo6o/ei1YqqvqwoK/srzlU8mtPut6A/fgii6fxE1ClgLMAZ7i0IfzV+ANcS301nqvS1UROVtVJ4rIVlxZc+RxZnMmLihdhQtmobhy7Ou89/TpuEYLr3HoqRpcSUXacZ4734hIsKompf8Gl2ERqYG7w+6ICw6NgWU++w3CBZPFuPL6WbhGArWANV4R7GIgWkQqqOr+Arys46aH6s8a4d6D4K7vYlx97Vm4m7f0z3mqurrbf4FrVfVFEZmMeyLeBwzBFd2twN1Ic9Tvi8KOovkQlcvgWhHc5/1dznuRqnvbK3q/r8d9oIKAR3BfpiX2iQKfOwngebw7E1xx1J+4fitnAKtwd8ltccUdTb10dwMv4u5grwPex31of8S74ynuP+R8V3uq9xrVB/7A3cWWwd2pnoP7gvoCuDTLfpOBF7y/Q3Bf8PWOdL7jzP81uKfrK3HFMa/iFZ8Vlx/cHfQSn+Ug7/Wri/tyfAk4z9t2s/ceregth3m/B/i89u8Cg9O3FfUfIMT7Hez9DvB+V/febzO9z91zQGlc0dtUXHB8G7gvh2NeAGzyeS929NnWAlcEf5M/78dCf6Hy4YWvhWunXtZn3bfAUN8PLq4I6y3v79rpb8aS8kMOj+C4u9KLcA0E5nvryuLuVC8GvgbS8Mp/gTHA/d7fHbwvyAu85XJAtcK+zrx6rXJ47bp51/8FrgjgceDpLOk64uoKzvBZdzmuCOts4I3DnDNPggaujPoKXDHZSqBHYb+eeXBNPwAP457i1uGe7vrjGq6Eep/XQO/9/BOuUry8z//dKOBub7kDrhVVoV/XUVx3T2CGz3KIz98/4+Yq6u4FjKnAQG/bG7gWje281+h2XN3bcA7Vp/XN5TshBK8ezJ+fktiqKr31RDOfdQ/jyjwnisiLIjIHd1f2A4CqblXVXws8p/kgvSWTZi9mKYNrRXYj7rWoJCIXqit+uxb3hDYY9wZ80NttDu5uBVVdgPtymuMtx2gx7deS3mosnbrWUOVF5BYRaa3ucb0hrg5iuKpOwDXb3uvtnz50yEJc652XxHXsW4CbP6a8qv6mqvfldl7f/5vjlIgrs/4TOElV/8ij4xamj4G7cEEj/WmvEi5QfIgLKGtwX6K34opo3haRKbj6ylRcQEFVF2g+NH/OC+JGFDjdZ9VSXJHmRSIyDvhPRNL7YkzGq69Q1YO4z+IN3rbfcU/9i3HB4nRvezCuaTyqOjnrd4K3PlGPpT9PYUfZfIjaobhWLmO95XK4opUw4DTc3Uuxenw/imvOqYilNa68t6W3fDXwgR66yxgCfOYtvwHc5v19C+6powbuA/sw3t1ccf4hh5ZjPtvuxBUBvIarbL4ed1f7MYeevq7A3c2195Zr4T3y48raBwGNcjhnsah4LUo/3msbD7T1lq/DNR31LWJ5ElcfB6611Bm4L9KQgs7vMVxfRus84Hvve6kerthoCDAfVzd1Lq4IPb0hzx9Ac2+/MFzJSmNci7nfgXPSX4/8voYSOZGTiNTEtWtPxd2NTQOe1GJWGXYkvh3OfNadi3ssLYX7QKX3kK8IjFHVdt4dcyNcu+5OuDqeBrg342bcm/hZVV1RMFdSsMRNSXwQV1asuKetX4AduDqbbbhAcDWubuxRb78ncM21w3Bly08CX6nXj8NLkx4simzlc3EgIh8Bi1T1ZRFpjbuhWY6r3O6BayU2WlU/K7RM5gER+Rh3U/IHrn6qNK6FU0Vv+z24Zu2Pisj7uAYO76mbTfVrYJWqPiaHRnHw7YQaCGR7ysgLJbJVlapuF5G+uAi+WL1WUcWdiASpanL6sroilmBcpdkmVV2GCwDX4u46kkXkGVyRy0hcy7GTVHWhiNTGlcVfr6rPicgtuLvssaq6p2CvLP+JSAXcHVx34Ezc1MQbcU+nV+OK8Pbi+jp8oKpxXiexFiLSSlWXe69TZ1z/i2zDqYhkjChQ8u7GCt4XuL4vL+MabOzDleFfiavLeVBVIwotd8fI63Q5AFcHtg3XqilJVXt528vgWoB1Vddpcw3QVESa455OBuCaGO/GtfA7AKCqK73fGTcsvjc0ea0k1nEAoKoHVHVuSQga4gbMOxf3+Oq7/nZcme6VuJ6vp+H6W2zDPVGAK4evDlTAfRifEpH/eft8gmu1gqp+oKrPF9egISKNRKS/uHGzchrJuAmu57+oagfcE8V1uMYBccBIVT1bVUfjAmx93N1tGq4HMgCqOi89aGTtGZ8fd3YnsN+BWiJyiroy+F9xLQEvVNWHimPQ8FyN+2wOxr3vtgFxInITgLoxzKZxqP5iOa7IszcucHwJxHppt6hqTIHm3lNiA0dxJyK1ROR8yLiLSABeEZGh4kY7LY17A/bGfQl2xb3ZknDFLn29Qy3FVaCep6rDcOMVNQc+V9VHVLXAJ7rPSz7DRJTBvQZtIPuXuKrOB+YBW0WktLqK/Zm41280cIm4IcfvBmYA/dQN+PY/Vf0uyznTh2TJtzu6E52qHsD1sk8fzmWeqq4r3Fzlie7ATlxH2qtwN26jcE8P6b7BtXJEVTfhnjC+U1eRPUHdMD+FqkQWVRVXXgdE8b6QonHDQXfDVfydjqunqI8LFPVxY2tNx5XNv48rZtrnlX0Ow1W0rcHd1QR5Hap+w3UGKxH0UIuQINxwG818O45BprqgObiivAq4Oo7puObFV4nIBlyTT8U1QZ7lHf9ADue0J4sCoKqjCjsPeckrphJc44u3ga6qGu0VN/9PRK7C1aFNAD4Qb7RazdJSzqdItNBY4ChCfCq1SuHa5wfjerCPxN0V78W1GokXkWRcOeeXvk8NIlIP1wSvooh0VNX5IjJK/RnArAjKrdJZRJoB43CBIBD3NPUNXk/hLHUOX+FaTd0vIj/giutGQkZz4zsL4lrMicmrc1wKpKjq2wBec9wgXOvFS3Atqrao6lDffX2DRWEHDaBktqoqbtLfFCLSA9dDey+ubDMEuErdaKd4FbPjcXcqSSIyENe65HXc8CgDgHdU9R0RqaGqO4rC3cmxyqXVWDnc0CdLvesbCJyqqjeLGzr8Qdxj/eRcjvk+7q7vAK6fwNu+j/7iM5R8vlyUOaF5ld9jcN0G6uC6C7wGTPZt+OKlDSyqxaH2xFEEeEGjHa6ccwJuHJofcK1/HhORuqoahausjcEVXU3H9R5dj3v0DQGGqOov3jF3pB+7YK/m+KV/YHy/vL0P3IO4uptZuArFUbixovZ6yebiGgM0FzcUfGzWY+Jet17AIzmVFVvAMPnJa613B25k42D1xotL5xssimrQAKscLzLU9fq8Bdf56U1cUdWZuArdG700ibixd94S1/v9TC9Q3KKq16QHjeLIt4VS+gdGRLqKyF3e6qq4SYha4u7QrsIFgFBgr7gBBRX32H820Mo7hqTXHXnH+Qo3QmrrrK2ijCkIqpqgqnP00CCj2d77RZ09cRQtPXB9LvriWl/cjGsy+6D35VdRVR8RkS3AcvWGUijOd8npRWmauRNdfVznxDjgL6/ysClwtrhZ23bjhgIZLSJNcWMVPSoib+IaDawBqmSp40jzKifrA/29Og1jCk1O7/3iwp44ipZwIFZVdwF7cONEbcU1SyyDG6cHVZ2uRXT8nSPx+qRkGrNJRMp6fTDGekVSZ+AC5rmqOsyr2E/Gddgboqo9vaARhCumegFXhDcKN3bUXar6U3oxnbhZ8t7GFWP1wE10ZEyhKo7FyOmscrwI8ZrePoF7EozD1Wm849VvFGs5VXR761vgZmpbCfyqql+LyHV4M5vhWpVVxnXeu5JDQ6Rc7C3/T1X/kSy96r1j18e1morGVUguKc5PZ8YUFRY4ihiveV5d4Oui0NHneEgOY+V4LZ96Ar+r6iwRuQY37Mkl4jO5jog0xo3bE4ILGpNx9RM3efvvxI3Z809u5/SebAL1WEb/NMbkygKHyXNZny68IqUw3N1/OG74iNa4eZ7n44aX2AJsx40UeocXVEJxAy6OAL5Q1Y9ze3IxxhQcqxw3eU7d4IuhuMEX78M1GX4TN3vbetxkUvfiOu0tBk7BzdAW6zWxvUpE1uFGOK7AoacN306SgRTxKVCNKansicPkORGpg+sv8TfwEW5U2vq4yv2rcTcsc3HDK3yjqj97nRtb4cbbGquqX6V3YiyMazDG5M6eOEyeU9UtInIQ2K+qf4nISlwT4+eBParayxtWZQWwRUSm4cbW2gY8r6q/e8fZ4dVTBBTHJovGlFQWOEx+mYDrkwJueI8duKeMvl6rqS64zo3RuArsC3I6iFexbkHDmCLEiqpMvhCRWrjAcKmqRojIGFzxVSpuHox/gQlZhwXB6i2MKfIscJh8IyI/4iq3l+L6XwxU1XlZ0qTPbWFvRGOKCQscJt+ISC9gKHCNqm7Msq3IjvxpjDk8G3LE5KdZQBWgPGT05wCKz2BuxpjsLHCYfKNu9rw3cCPYknVIEGNM8WRFVcYYY/xiTxzGGGP8YoHDGGOMXyxwGGOM8YsFDmOMMX6xwGGMMcYv/wdOyWZr7xgchQAAAABJRU5ErkJggg==",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots()\n",
"\n",
"ax.plot(rfe.grid_scores_)\n",
"ax.set_ylim(0, 1)\n",
"\n",
"ax.set_xticks(np.arange(len(df.columns[:-1])))\n",
"ax.set_xticklabels(df.columns[:-1], rotation=20)\n",
"ax.set_ylabel('R-squared', fontsize=15)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 7. Other regressors\n",
"\n",
"We used a linear model, but we could easily subsitute another regressor. Let try an algorithm called \"Random Forest\". This algorithm has the ability to weight the features relative to each other. Let's explore the residuals and the feature importances of the Random Forest algorithm.\n",
"\n",
"1. Import the ``RandomForestRegressor`` method from ``sklearn.ensemble`` [[How to use RandomForestRegressor]](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html?highlight=random%20forest#sklearn.ensemble.RandomForestRegressor)\n",
"2. Using your train dataset, create your Random Forest model and fit it. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 7.1 Creating the model"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [],
"source": [
"from sklearn.ensemble import RandomForestRegressor"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/var/folders/5_/9z7lhk0s2y95hvkzs6lzdvvc0000gn/T/ipykernel_85267/3759250952.py:2: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
" RF.fit(X_train, y_train)\n"
]
},
{
"data": {
"text/plain": [
"RandomForestRegressor(n_estimators=1000, random_state=42)"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"RF = RandomForestRegressor(n_estimators = 1000, random_state = 42)\n",
"RF.fit(X_train, y_train)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 7.2 Predicting from test data\n",
"\n",
"Just like for the OLS model above, we will start by investigating how well our model, constructed from the training set, predicts the scores from test set.\n",
" \n",
"1. Create predicted data using the model's `predict` method. \n",
"3. Score the accuracy of your model by calculating the `mean_squared_error` between the predictions from your fit model and your test dataset outcomes.\n",
"2. Make a scatter plot to compare it to the actual values and draw a diagonal through this plot. \n",
"\n",
"### Questions\n",
"\n",
"1. What does the root mean squared error tell us? How does it compare to the `score` from the OLS model above? \n",
"2. What \"shape\" does the scatter plot \"blob\" look like? How does it compare to the Linear Regression plot we made above?\n",
"3. Does the \"blob\" follow the diagonal line or does it deviate in some way?\n",
"4. Can you tell if the model over or under predicts scores in the test set?"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [],
"source": [
"predicted = RF.predict(X_test)\n",
"## Unfortunately the 'predict' function for RandomForestRegressor is in a different shape than the predict function for LinearRegression\n",
"## We need to reshape this array in order to calculate the errors\n",
"predicted = np.reshape(predicted,(1000,1)) "
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Mean Absolute Error: 1.5800581835981227\n",
"Mean Squared Error: 4.173471812775219\n",
"Root Mean Squared Error: 2.0429076858182356\n",
"Average fci_post score: 17.347833832521747\n"
]
}
],
"source": [
"from sklearn import metrics\n",
"print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, predicted))\n",
"print('Mean Squared Error:', metrics.mean_squared_error(y_test, predicted))\n",
"print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, predicted)))\n",
"\n",
"print('Average fci_post score:', df['fci_post'].mean())"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"With 1000 trees, the root mean squared error is 2.01 which is greater than 10 percent of the average fci_post score i.e. 1.73. This may indicate, among other things, that we have not used enough estimators (trees). Computational power issue??"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Text(0.5, 0, 'Actual FCI Score')"
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots(figsize=(10,10))\n",
"\n",
"ax.scatter(x=y_test, y=predicted, alpha=0.15)\n",
"ax.plot([-0, 30], [-0, 30], color='red')\n",
"ax.set_ylabel('Predicted FCI score', fontsize=15)\n",
"ax.set_xlabel('Actual FCI Score', fontsize=15)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 7.3 Inspecting the residuals of the model\n",
"\n",
"1. Write a function to calculate the residuals of the model. \n",
"2. Plot the actual values versus the residuals using a scatter plot. (*This is the most common way of seeing a residual analysis in practice.*)\n",
"3. Collapse the residual scatter plot into a histogram. (*This is a useful visualization to see the normality of the distribution*) [[How to plot a histogram]](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.hist.html)\n",
"\n",
"### Questions\n",
"\n",
"1. How does this plot compare to the plot we produced for Linear Regression?\n",
"2. Does the model tend to overpredict or underpredict certain groups of scores?"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [],
"source": [
"residuals = calculate_residuals(actual=y_test, predicted=predicted)"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Text(0.5, 0, 'Actual')"
]
},
"execution_count": 28,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots(figsize=(10,8))\n",
"\n",
"residuals['fci_post'].hist()\n",
"\n",
"ax.set_xlabel('Residuals', fontsize=15)\n",
"ax.set_ylabel('Counts', fontsize=15)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 7.4 Finding feature importances\n",
"\n",
"For our linear model above, the coefficients are related to the correlation between each input varaiable and the output prediction. Earlier you looked at the correlations between each input variable and the output variable. The Random Forest algorithm feature importances says that in a given model these features are most important in explaining the target variable. These importances are relative to each of the other features in your model.\n",
"\n",
"1. Return the fit importances using ``feature_importances_``\n",
"2. Make a bar graph of all the features in the model. [[How to make a horizontal bar plot]](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.plot.barh.html)\n",
"\n",
"\n",
"### Questions \n",
"1. Which is the most important feature for fitting? \n",
"2. Which is least important?\n",
"3. How do this compare to the coefficients we analyzed in OLS model above?"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"