PredictingNBASalariesFinal.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Linear Regression to Predict NBA Player Salaries based on Stats"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### by Dotun Opasina"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Introduction\n",
    "The goal of this project is to predict Nba Players salaries from 2017 - 2019 using linear regression model while performing data scraping and feature extractions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Include Libraries\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from sklearn.linear_model import LinearRegression, LassoCV\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plot\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import numpy as np\n",
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Declare split and validate functions to be used later in the code\n",
    "def split_and_validate(X,y,test=.25,random=3):\n",
    "    X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=test, random_state=random)\n",
    "    \n",
    "    # fit linear regression to training data\n",
    "    lr_model = LinearRegression()\n",
    "    lr_model.fit(X_train, y_train)\n",
    "    \n",
    "    # score fit model on validation data\n",
    "    val_score = lr_model.score(X_val, y_val)\n",
    "    \n",
    "    y_pred = lr_model.predict(X_val)\n",
    "\n",
    "    residuals = y_val - y_pred\n",
    "    \n",
    "\n",
    "    # report results\n",
    "    print('\\nValidation R^2 score was:', val_score)\n",
    "    print('Feature coefficient results:')\n",
    "    for feature, coef in zip(X.columns, lr_model.coef_):\n",
    "        print(feature, ':', f'{coef:.2f}')\n",
    "    \n",
    "    return X_train, X_val, y_train, y_val, residuals, lr_model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 1: Data Retrieving and Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "onlyPlayerStats = pd.read_csv(\"NbaPlayerStats20172019.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>MP</th>\n",
       "      <th>ThreePoints</th>\n",
       "      <th>AST</th>\n",
       "      <th>FG</th>\n",
       "      <th>FT</th>\n",
       "      <th>TOV</th>\n",
       "      <th>DRB</th>\n",
       "      <th>FGA</th>\n",
       "      <th>TRB</th>\n",
       "      <th>ORB</th>\n",
       "      <th>FTA</th>\n",
       "      <th>STL</th>\n",
       "      <th>BLK</th>\n",
       "      <th>PF</th>\n",
       "      <th>PointsPerGame</th>\n",
       "      <th>Salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1422.000000</td>\n",
       "      <td>1.422000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>26.274965</td>\n",
       "      <td>20.104993</td>\n",
       "      <td>0.849930</td>\n",
       "      <td>1.938748</td>\n",
       "      <td>3.226231</td>\n",
       "      <td>1.395007</td>\n",
       "      <td>1.119339</td>\n",
       "      <td>2.773980</td>\n",
       "      <td>7.121167</td>\n",
       "      <td>3.603165</td>\n",
       "      <td>0.832911</td>\n",
       "      <td>1.830591</td>\n",
       "      <td>0.642827</td>\n",
       "      <td>0.396343</td>\n",
       "      <td>1.714416</td>\n",
       "      <td>8.695640</td>\n",
       "      <td>6.282043e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.270868</td>\n",
       "      <td>9.080240</td>\n",
       "      <td>0.796339</td>\n",
       "      <td>1.785743</td>\n",
       "      <td>2.174433</td>\n",
       "      <td>1.362123</td>\n",
       "      <td>0.803895</td>\n",
       "      <td>1.857142</td>\n",
       "      <td>4.550271</td>\n",
       "      <td>2.485467</td>\n",
       "      <td>0.779983</td>\n",
       "      <td>1.669720</td>\n",
       "      <td>0.435366</td>\n",
       "      <td>0.426993</td>\n",
       "      <td>0.777079</td>\n",
       "      <td>6.050177</td>\n",
       "      <td>7.368601e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>19.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.608000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>23.000000</td>\n",
       "      <td>13.300000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.525000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>3.700000</td>\n",
       "      <td>1.900000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.725000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1.200000</td>\n",
       "      <td>4.300000</td>\n",
       "      <td>1.312611e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>26.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>1.400000</td>\n",
       "      <td>2.700000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.400000</td>\n",
       "      <td>6.100000</td>\n",
       "      <td>3.100000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>1.300000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>7.300000</td>\n",
       "      <td>2.962860e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>29.000000</td>\n",
       "      <td>27.475000</td>\n",
       "      <td>1.300000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>3.600000</td>\n",
       "      <td>9.800000</td>\n",
       "      <td>4.700000</td>\n",
       "      <td>1.100000</td>\n",
       "      <td>2.400000</td>\n",
       "      <td>0.900000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>2.200000</td>\n",
       "      <td>11.800000</td>\n",
       "      <td>9.000000e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>42.000000</td>\n",
       "      <td>37.800000</td>\n",
       "      <td>5.100000</td>\n",
       "      <td>11.200000</td>\n",
       "      <td>10.800000</td>\n",
       "      <td>9.700000</td>\n",
       "      <td>5.700000</td>\n",
       "      <td>11.100000</td>\n",
       "      <td>24.500000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>5.400000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>2.400000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>36.100000</td>\n",
       "      <td>3.745715e+07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Age           MP  ThreePoints          AST           FG  \\\n",
       "count  1422.000000  1422.000000  1422.000000  1422.000000  1422.000000   \n",
       "mean     26.274965    20.104993     0.849930     1.938748     3.226231   \n",
       "std       4.270868     9.080240     0.796339     1.785743     2.174433   \n",
       "min      19.000000     1.000000     0.000000     0.000000     0.000000   \n",
       "25%      23.000000    13.300000     0.200000     0.700000     1.600000   \n",
       "50%      26.000000    20.000000     0.700000     1.400000     2.700000   \n",
       "75%      29.000000    27.475000     1.300000     2.500000     4.400000   \n",
       "max      42.000000    37.800000     5.100000    11.200000    10.800000   \n",
       "\n",
       "                FT          TOV          DRB          FGA          TRB  \\\n",
       "count  1422.000000  1422.000000  1422.000000  1422.000000  1422.000000   \n",
       "mean      1.395007     1.119339     2.773980     7.121167     3.603165   \n",
       "std       1.362123     0.803895     1.857142     4.550271     2.485467   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.500000     0.525000     1.500000     3.700000     1.900000   \n",
       "50%       1.000000     1.000000     2.400000     6.100000     3.100000   \n",
       "75%       1.800000     1.500000     3.600000     9.800000     4.700000   \n",
       "max       9.700000     5.700000    11.100000    24.500000    16.000000   \n",
       "\n",
       "               ORB          FTA          STL          BLK           PF  \\\n",
       "count  1422.000000  1422.000000  1422.000000  1422.000000  1422.000000   \n",
       "mean      0.832911     1.830591     0.642827     0.396343     1.714416   \n",
       "std       0.779983     1.669720     0.435366     0.426993     0.777079   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.300000     0.725000     0.300000     0.100000     1.200000   \n",
       "50%       0.600000     1.300000     0.600000     0.300000     1.800000   \n",
       "75%       1.100000     2.400000     0.900000     0.500000     2.200000   \n",
       "max       5.400000    11.000000     2.400000     6.000000     4.400000   \n",
       "\n",
       "       PointsPerGame        Salary  \n",
       "count    1422.000000  1.422000e+03  \n",
       "mean        8.695640  6.282043e+06  \n",
       "std         6.050177  7.368601e+06  \n",
       "min         0.000000  4.608000e+03  \n",
       "25%         4.300000  1.312611e+06  \n",
       "50%         7.300000  2.962860e+06  \n",
       "75%        11.800000  9.000000e+06  \n",
       "max        36.100000  3.745715e+07  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Do Some Data Cleaning\n",
    "onlyPlayerStats = onlyPlayerStats.rename(columns={'3P' : 'ThreePoints', 'PTS/G': 'PointsPerGame' })\n",
    "\n",
    "onlyPlayerStats.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Remove Player Salaries that are Zero \n",
    "onlyPlayerStats.drop(onlyPlayerStats[onlyPlayerStats.Salary == 0].index, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Since we are dealing with Salaries data, we perform a logarithmic transformation to make highly skewed distribution salary differinces less skewed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "onlyPlayerStats[\"Salary_Log\"] = np.log(onlyPlayerStats[\"Salary\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2: Perform Initial Validation on initial data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Take out Name, Salary and Salary Logged and Season from our data and use the remaining data for\n",
    "## our X value\n",
    "## Select only the Salary logged as our Y\n",
    "X, y = onlyPlayerStats.drop(['Name','Salary', 'Salary_Log','Season'],axis=1), onlyPlayerStats['Salary_Log']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Validation R^2 score was: 0.42612028609801234\n",
      "Feature coefficient results:\n",
      "Age : 0.08\n",
      "MP : 0.06\n",
      "ThreePoints : 0.08\n",
      "AST : 0.02\n",
      "FG : 0.29\n",
      "FT : 0.19\n",
      "TOV : -0.06\n",
      "DRB : -0.43\n",
      "FGA : -0.06\n",
      "TRB : 0.61\n",
      "ORB : -0.67\n",
      "FTA : -0.07\n",
      "STL : -0.13\n",
      "BLK : -0.17\n",
      "PF : 0.11\n",
      "PointsPerGame : -0.04\n"
     ]
    }
   ],
   "source": [
    "## Do a split and validate of our original data\n",
    "X_train, X_val, y_train, y_val, residuals, model = split_and_validate(X,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What we notice from the result above is there are many features that have negative relationships with the salary column so we need a way to pick what features are most important for our model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 3: Perform Lasso algorithms for Feature selections"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Standard Scale our Inputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "StandardScaler(copy=True, with_mean=True, with_std=True)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## We first use a standard scaler to scale the inputs\n",
    "std = StandardScaler()\n",
    "std.fit(X_train.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Transform our X column\n",
    "X_tr = std.transform(X_train.values)\n",
    "X_te = std.transform(X_val.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Lasso Step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "lassocv = LassoCV(cv=5, random_state=0) ## Initialize Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n",
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n",
      "  ConvergenceWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LassoCV(alphas=None, copy_X=True, cv=5, eps=0.001, fit_intercept=True,\n",
       "    max_iter=1000, n_alphas=100, n_jobs=None, normalize=False,\n",
       "    positive=False, precompute='auto', random_state=0, selection='cyclic',\n",
       "    tol=0.0001, verbose=False)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lassocv.fit(X_tr, y_train) ## We fit our data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4331577336227551"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lassocv.score(X_te, y_val) ## Our Lasso Score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Alpha (log10)')"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Plot out the Lasso Alphas\n",
    "plot.figure(figsize=(16, 8))\n",
    "plot.plot(np.log10(lassocv.alphas_),lassocv.mse_path_)\n",
    "plot.ylabel(\"Mean Squared Error\")\n",
    "plot.xlabel(\"Alpha (log10)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Print out Columns to their Lasso Coefficients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('Age', 0.2664831766364393), ('MP', 0.48080219845698413), ('ThreePoints', 0.0), ('AST', 0.0), ('FG', 0.0), ('FT', 0.0), ('TOV', 0.0), ('DRB', 0.26304816412713483), ('FGA', 0.0), ('TRB', 0.0), ('ORB', 0.0), ('FTA', 0.0), ('STL', 0.0), ('BLK', 0.0), ('PF', 0.006711896757269426), ('PointsPerGame', 0.21123285019191615)]\n"
     ]
    }
   ],
   "source": [
    "columnList = list(zip(X.columns,lassocv.coef_))\n",
    "\n",
    "print(columnList)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we see that most of our columns have a lasso coeefficients of 0 so we can remove them from our features list as they do not contribute any significant effect on our data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "usefulColumns = [x[0] for x in columnList if x[1] != 0.0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Age', 'MP', 'DRB', 'PF', 'PointsPerGame']\n"
     ]
    }
   ],
   "source": [
    "print(usefulColumns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The Best Predictors on Salaries are:\n",
    "    1. Age\n",
    "    2. Minutes Played per Game\n",
    "    3. Defensive Rebounds per Game\n",
    "    4. Personal Fouls\n",
    "    5. Average Points Per Game"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Split Data into X and Y to be used for validating Lasso\n",
    "X, y = onlyPlayerStats[usefulColumns], onlyPlayerStats['Salary_Log']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "## View Correlation of Input Data and Output\n",
    "usefulColumns.append('Salary_Log')\n",
    "onlyUsefulColumns = onlyPlayerStats[usefulColumns]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 0, 'Age'),\n",
       " Text(1.5, 0, 'Minutes Played'),\n",
       " Text(2.5, 0, 'Defensive Rebound'),\n",
       " Text(3.5, 0, 'Fouls'),\n",
       " Text(4.5, 0, 'Points'),\n",
       " Text(5.5, 0, 'Log of Salary')]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6IAAAHVCAYAAADrQEbXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3XuYFNWZ+PHvmRvI/SICAgrqoCKKaCReoiCKiNeoqEPcKNFI1g0xWWOSzf6MCe5mExNXg0o2EjVGoyIajQR0JypGRVFxRQQBBYXIKIIKCshlLpzfH9PgzDBcBLq7oL6f5+nHrqpTVe+ZYzH9znuqOsQYkSRJkiQpVwryHYAkSZIkKV1MRCVJkiRJOWUiKkmSJEnKKRNRSZIkSVJOmYhKkiRJknLKRFSSJEmSlFMmopIkSZKknDIRlSRJkiTllImoJEmSJCmnivIdgHbMqBBivmPQzvfTN97IdwjKljZt8h2BsmXPPfMdgbLlo4/yHYGy6Xe/y3cEypbrrgv5DuGLysZn+5/GmMifgxVRSZIkSVJOWRGVJEmSpARIU5UwTX2VJEmSJCWAFVFJkiRJSoA0VQnT1FdJkiRJUgJYEZUkSZKkBEhTldBEVJIkSZISIE2JaJr6KkmSJElKACuikiRJkpQAaaoSpqmvkiRJkqQEsCIqSZIkSQmQpiqhiagkSZIkJUCaEtE09VWSJEmSlABWRCVJkiQpAdJUJUxTXyVJkiRJCWBFVJIkSZISIE1VQhNRSZIkSUqAkO8AcihNSbckSZIkKQGsiEqSJElSAhTmO4AcsiIqSZIkScopK6KSJEmSlABpqhKmqa+SJEmSpASwIipJkiRJCZCmKqGJqCRJkiQlQJoS0TT1VZIkSZKUAFZEJUmSJCkB0lQlTFNfJUmSJEkJYEVUkiRJkhIgTVVCE1FJkiRJSoA0JaJp6qskSZIkKQGsiEqSJElSAoR8B5BDJqJKvLPuuIOeZ5zBZ0uX8j+HHprvcPQFPfvqq/z8jjtYv3495598MiPOO6/e9sqqKn44ejRvvP02bVq25Karr6brXntRVV3NNWPGMPudd6iuqeGrJ57Itxrsq9x79uWX+fmtt7K+pobzTz+dEV/7Wr3tlZWV/PAXv+CNt96iTatW3PTTn9K1U6eN299fsoTThw9n5PDhXHbhhQDc9eCDPDhpEiEEeu63H7/40Y9oUlKS034Jnp0yhZ9ff33t2J57LiO++c162ysrK/nhv/87b8yeTZs2bbjp17+ma5cuLP/kE6686ipmzZrFOWefzbX/7/9t3Oemm2/mLxMmsGLFCqa//HKuu6QMr9uUOOAAOO00CAFefRWee67+9mOPhSOOgPXrYfVqeOQR+PTT2m2tW8PZZ9f+N0b405/gk09y3welilNzsyyEcE4IIYYQDsp3LLuq1+66iz+demq+w9B2qKmp4bqxY7n9Jz9h0s03M3HKFOYvWlSvzYNPPkmr5s154n/+h+FnnskNd98NwP++8AKV1dX8dfRoHv7v/+aB8nIqli7NRzeUUVNTw3WjR3P7L3/JpLvuYuJTTzF/4cJ6bR587DFatWzJE/fey/Dzz+eG226rt/0XY8Zw/Je/vHF5yYcfcvfDD/Pn225j4h/+QE1NDZMmT85Fd1RHTU0N1/3859z+298y6dFHmfj448x/++16bR58+GFatWrFE489xvCvf50bbroJgCYlJXx35Eh+ePXVmxz3xP79efD++3PSBzXO6zYlQoAzzoB77oFbb4VDD4UOHeq3WbwYbrsNfvtbeOMNOOWUz7edey48/zzccguMHQuffZbb+LVRYRZeSWUimn3DgClAWb4D2VW9+9xzrFm2LN9haDu8Pm8e+3buTLdOnSgpLub0r3yFpxpURSa//DLnnHgiAIOPPZapr79OjJEQAmvWrqW6poa169ZRXFREiz32yEc3lPH63Lnsu/fedNt779rxHDiQp55/vl6byc8/zzmDBwMwuH9/pr76KjFGAJ6cMoWue+9Naffu9fapyYzxhrHeq337nPRHn3t95kz23WcfunXrVju2Q4bw1NNP12sz+emnOeesswAYPGgQU196iRgjzZo140tHHNFoNezwPn3Yq+GHYeWU121KdO0Ky5bB8uVQUwMzZ8JBDWogCxZAVVXt+0WLaqufUJuwFhTAhj8+VVZ+3k45V5CFV1IlObZdXgihBXAccBmZRDSEUBBC+G0I4Y0QwsQQwmMhhKGZbUeGEJ4JIfxfCKE8hNA5j+FLO2zJsmV02nPPjcsd27dnyccf12/z8cd0zrQpKiykZbNmLF+5ksHHHMMeTZvylUsv5cQRI7j0q1+lTcuWOY1f9S356CM67bXXxuWOHTqw5KOPNmnTOdOmqLCQli1asHzFClavWcPv77+fkZdcUq99xw4duPSCCzjxwgv5ynnn0aJ5c75y1FHZ74zqWbJ0KZ3qTMXs2LEjS5Ys2aRN50yboqKi2rF16l7ied2mRMuWn0+zBVixAlq12nz7I4+EefNq37dvD2vXQlkZXHFFbaU0pOlOReWLiWh2fRX43xjjW8CyEMIRwLlAd+BQ4JvAMQAhhGLgFmBojPFI4E7g540dNIQwIoTwSgjhlVey3wdpu234i3pdocEvt01b1N6o//q8eRQUFPDcHXfw1O9+x52PPsqiDz7ITqDaJts0no21AW656y4uGTqU5g2q2p+uXMlTL7zAU/ffz3MPPcSatWt59Ikndmrc2rrtHls/rCae121KNHYtNjKuABx2GOy9N0yZUrtcUAD77gvl5bVTd9u2hb59sxertihNFVEfVpRdw4DfZN6PyywXAw/GGNcDH4QQNsx9OhDoDTyR+QVRCCxu7KAxxrHAWIBRIWzmXxkp/zq1b88Hdf7yvuTjj9mrXbtN2iz+6CM67bkn1TU1rFy9mjYtWzLx2Wc5vm9fiouKaN+mDUccdBAz336bbnWqNsqtTh068EGd+3SXfPjhJtPxOnXowOKlS+nUoUPteK5aRZtWrZgxZw7lzzzDDbfdxopVqygoKKBJSQnt27ala6dOtGvTBoBTjj+e6bNmcfagQTntW9p16tiRD+r8oWfJkiXsVaeKtqHN4g8+oFOnTlRXV9eO7YapfUosr9uUWLHi86m2UFsNXbly03b77Qf9+8Odd9ZO4d2w7+LFtdN6AebOrZ3qK2VZkpPkXVoIoT0wELg9hLAQ+AFwIZt/KnMA3ogxHp55HRpjPGUzbaVdwqGlpSxcvJhFS5ZQWVXFpClTGNhg+tbAo47ikcy9aOUvvMDRhx5KCIHOHTrw0syZxBhZvXYtM956i/26dMlHN5Rx6EEHsfC991i0eHHteE6ezMBjj63XZuCxx/JIeTkA5c88w9F9+xJC4L6bb2byuHFMHjeOS4YO5VsXXcQ/nXMOe++1FzNmz2bN2rXEGJn66qvsv++++eheqh3auzcL//EPFlVU1I7t448zcMCAem0GDhjAIxMmAFD+xBMc3a+fFdFdgNdtSrz3HrRrB23aQGFh7cOK5s6t36ZTJzjrLLj33voPI3rvPdhjD2jWrHa5Rw/48MPcxa56rIhqZxgK3B1j/NaGFSGEZ4CPgPNCCH8EOgADgPuAN4EOIYRjYoxTM1N1e8YY38h96Mly7n330X3AAJrtuSf/umgRf//pT5l+5535DkvboKiwkGsvv5xvjhpFzfr1nHfSSZTusw+j77uP3gccwEn9+jH05JP5wW9+w6ArrqB1ixbc9P3vA3DRkCH8+JZbOOO73yXGyLkDB3JQg4dlKLeKCgu59sor+eYPf1g7nkOGUNqjB6PvvJPeBx7ISccdx9DTT+cH//VfDLroIlq3asVNP/nJFo/Zp1cvBvfvzzkjRlBUWMjBpaVceMYZOeqRNigqKuLaf/93vvnP/0xNTQ3nnXMOpQccwOhbb6X3IYdw0oknMvTcc/nBj3/MoNNOo3Xr1tz0q19t3H/g4MGsWrWKqqoqnpw8mTvHjuWA/ffnVzfeyMRJk1izdi0nnHQS5593Ht/5l3/JY0/Tx+s2Jdavh0mT4OKLa6favvpqbTI5cGBtovnmmzB4MJSUQOYrePj0U7jvvtopvOXlMHx47RTf99+H//u/vHYnzZKcOO5sobH7ArTjQgh/B34ZY/zfOuuuBA6mtvp5AvAW0AS4Mcb4RAjhcOBmoDW1fyT4TYzx91s6j1Nzd08/fSP1f3/YfWWmsmk3VOfBXNrNNHi4j3Yzv/tdviNQtlx33S43beORLHy2PyfGrf4cQginAqOpvT3w9hjjLxts3wf4I9Am0+bfYoyP7UhcVkSzJMY4oJF1N0Pt03RjjKsy03dfBmZmtr9GbYIqSZIkKWXyURENIRQCY4BBQAUwLYQwIcY4u06za4DxMcb/CSH0Ah6j9gGs281END8mhhDaACXAf8QYfRSoJEmSpHzoB8yPMb4DEEIYB5wN1E1EI7DhO4FaA+/v6ElNRPOgsWqpJEmSpHTLRkU0hDACGFFn1djMt3Bs0AVYVGe5Avhyg8P8DPhbCOE7QHPg5B2Ny0RUkiRJknZTdb/6cTMau4e04b2qw4C7Yoz/HUI4BrgnhNA785WU28VEVJIkSZISIE9PV6oAutVZ7sqmU28vA04FyHzDR1NgT2Ap2ylNTwiWJEmSpMQqzMJrG0wDSkMIPUIIJUAZMKFBm3eBkwBCCAcDTYEd+sJZE1FJkiRJSqkYYzUwEigH5lD7dNw3QgjXhRDOyjT7PnB5CGEGcD8wPO7g94A6NVeSJEmSEiBfVcLMd4I+1mDdtXXezwaO25nntCIqSZIkScopK6KSJEmSlABpqhKaiEqSJElSAqQpEU1TXyVJkiRJCWBFVJIkSZISIE1VwjT1VZIkSZKUAFZEJUmSJCkB0lQlNBGVJEmSpARIUyKapr5KkiRJkhLAiqgkSZIkJUDIdwA5ZEVUkiRJkpRTVkQlSZIkKQEK8x1ADlkRlSRJkiTllBVRSZIkSUqANFUJTUQlSZIkKQHSlIimqa+SJEmSpASwIipJkiRJCZCmKmGa+ipJkiRJSgAropIkSZKUAGmqEpqISpIkSVICpCkRTVNfJUmSJEkJYEVUkiRJkhIgTVXCNPVVkiRJkpQAIcaY7xi0I2bPdgB3Q6MOOSTfIShLzsh3AMqa6nwHoKyZm+8AlFWn5TsAZU2HGEO+Y/iiZoWw0z/b907oz8GpuZIkSZKUAIX5DiCHnJorSZIkScopK6KSJEmSlABpqhKmqa+SJEmSpASwIipJkiRJCZCmKmGa+ipJkiRJSgAropIkSZKUAGmqEpqISpIkSVICpCkRTVNfJUmSJEkJYEVUkiRJkhIgTVXCNPVVkiRJkpQAVkQlSZIkKQHSVCU0EZUkSZKkBAj5DiCH0pR0S5IkSZISwIqoJEmSJCVAYb4DyCEropIkSZKknLIiKkmSJEkJkKYqoYmoJEmSJCVAmhLRNPVVkiRJkpQAVkQlSZIkKQHSVCVMU18lSZIkSQlgRVSSJEmSEiBNVcI09VWSJEmSlABWRCVJkiQpAdJUJTQRlSRJkqQESFMimqa+SpIkSZISwIqoJEmSJCVAmqqEaeqrJEmSJCkBrIhKkiRJUgKkqUpoIipJkiRJCZCmRDRNfZUkSZIkJYCJqBLh2VdfZfC3v82gK65g7J//vMn2yqoqvnfDDQy64grO/+EPqVi6FICq6mp+NHo0Z373uwwZOZLbGtlXyXXWHXdw9ZIlXDFzZr5D0Q5qNXgwh8ydyyHz5tHxRz/aZHvXG2/k4OnTOXj6dA558036LF+ehyi1rVoPHsxhc+fSZ948OjcynvvceCO9p0+n9/TpHPbmmxyZGc9mffrQ64UXOHTWLA6dMYN2F1yQ69C1FXsPHsxX587lnHnz6N3I2Dbv1o1TJk/mjFdf5cwZM+gyZAgAPb72Nc6cPn3j6+KaGtr26ZPr8LUFxYMH03buXNrNm8cejYxtk0suof3SpbSdPp2206fT9LLLNm5rfv31tJ01i7azZ9N89Ohchq0GCrLwSqq8xRZCiCGEe+osF4UQPgwhTMwsnxVC+LcdOP73QgjNdkKcwzNxvRZCmB1CuLzO+lt39PhbOffCEMKe2TxHEtTU1HDd2LHc/pOfMOnmm5k4ZQrzFy2q1+bBJ5+kVfPmPPE//8PwM8/khrvvBuB/X3iByupq/jp6NA//93/zQHn5xiRVyffaXXfxp1NPzXcY2lEFBewzZgzzhgxhdq9etBs2jKYHH1yvScVVVzGnb1/m9O3L0ltu4ZOHH85TsNqqggK6jxnDm0OG8HqvXrQfNow9Goznu1ddxay+fZnVty9LbrmFZZnxXL96NW9ffDEze/dm7qmnsu9vfkNh69b56IUaEQoKOHrMGJ4cMoRHe/Wix7BhtG4wtoddcw3/GD+eiUccwbNlZRz9298CsOC++/hr3778tW9fnvv611m1cCHLZ8zIRzfUmIICWo4Zw6dDhrCsVy+aDhtGYYOxBVj3wAMs79uX5X37svaOOwAoOuYYio87juWHHcby3r0pPuooivv3z3UPlEL5TJI/A3qHEPbILA8C3tuwMcY4Icb4yx04/veAHU5EMx6IMR4ODAD+K4TQcScdV8Dr8+axb+fOdOvUiZLiYk7/yld46uWX67WZ/PLLnHPiiQAMPvZYpr7+OjFGQgisWbuW6poa1q5bR3FRES322KOx0yiB3n3uOdYsW5bvMLSDmvfrx9r586lcsIBYVcXyceNoc/bZm23fbtgwlt1/fw4j1BfRIjOe6zLjuWzcONpuYTzbDxvGx5nxXDtvHuvmzwegavFiqpYupahDh5zEra3bs18/Vsyfz6oFC1hfVcWCcePo1mBsY4wUt2oFQEnr1qx+//1NjtNj2DAWeA0nSlG/ftTMn8/6BQugqoq148ZRsoXrtp4YoWlTKCmBJk2guJj1S5ZkN2BtlhXR3HkcOD3zfhiw8V+1uhXHEMJdIYSbQwgvhBDeCSEMzawfsKGCmlm+NbPflcDewNMhhKcz204JIUwNIbwaQngwhNAis/6XmUrn6yGEG7YUbIxxKfA2sG/d9SGEM0MIL4UQpocQngwhdAwhFIQQ5oUQOmTaFIQQ5ocQ9gwhdAgh/DmEMC3zOi7Tpn0I4W+Z49wGhO3/0e46lixbRqc9Py/8dmzfniUff1y/zccf0znTpqiwkJbNmrF85UoGH3MMezRtylcuvZQTR4zg0q9+lTYtW+Y0fintirt0oarOLIbKigqKu3RptG3JPvvQpEcPVk6enKvw9AWVdOlC5RcczxWNjGfzo46ioKSEdW+/nbVY9cU069KFz+qM7eqKCpo3GNsZP/sZ+/3TPzF00SJOeuwxXvrOdzY5To8LLzQRTZiCLl2oqTO26ysqKGzkui057zzazphBqwcfpKBrVwCqX3yRqqefpv3ixbRfvJjK8nJq5s7NWeyqz0Q0d8YBZSGEpsBhwEtbaNsZ+ApwBrDFSmmM8WbgfeDEGOOJmemt1wAnxxiPAF4BrgohtAPOAQ6JMR4G/OeWjhtC2A/YD5jfYNMU4OgYY99Mn34YY1wP/Am4KNPmZGBGjPEjYDRwU4zxKOA84PZMm58CUzLHmQDss6V4dhcxxk3WhVA/B9+0RW2W/vq8eRQUFPDcHXfw1O9+x52PPsqiDz7ITqCSGhca+ZtZI9c1QNuyMpY/9BCsX5/loLTdvsB4ti8rY1kj41ncqRP733MP73zjG5vdV3nQyNg2/B3cY9gw5t91Fw9168ZTp53G8ffcU2+/Pfv1o3r1aj55442sh6svYBuu28q//pVl3buzvE8fKp98kpZ//CMABfvvT+HBB/Nx16583KULJQMHUnz88bmIWimX10Q0xvg60J3aauhjW2n+lxjj+hjjbOCLTo09GugFPB9CeA24hNqq5gpgLXB7COFcYPVm9r8ws9/9wLdijA3nEnYFykMIM4EfAIdk1t8JXJx5fynwh8z7k4FbM8ecALQKIbQETqA2eSXGOAlo9GkeIYQRIYRXQgivjB0/flt/BonVqX17Pvjoo43LSz7+mL3atdukzeJMm+qaGlauXk2bli2Z+OyzHN+3L8VFRbRv04YjDjqImf71XcqpqooKirt127hc0rUrVY1M5wNoV1bmtNyEq6yooGQbx7N9WdnGabkbFLZsyYGTJlFxzTWsemlLf19Wrq2uqKB5nbFt1rXrJlNvSy+7jIWZzxYfvvgihU2b0rTOrKUeZWVWQxNofUUFhXXGtqBrV2oajG1ctgwqKwFY+/vfU3TkkQA0Oeccql98ET77DD77jMrHH6fo6KNzF7zqCSHs9FdS5bsiCrWJ2A3UmZa7GevqvN/wE62mfh+abmbfADwRYzw88+oVY7wsxlgN9AP+DHwV+N/N7P9AZr8vxxgfaWT7LcCtMcZDgW9tiCPGuAhYEkIYCHyZ2qnIZGI+pk48XWKMKzPbtvqn4xjj2Bjjl2KMXxqxGzyR8NDSUhYuXsyiJUuorKpi0pQpDDzqqHptBh51FI88/TQA5S+8wNGHHkoIgc4dOvDSzJnEGFm9di0z3nqL/TYzhUxSdnw2bRpNS0sp6d6dUFxM27IyPpkwYZN2TXr2pLBtWz6bOjUPUWpbrcqMZ5PMeLYrK2N5I+PZtGdPitq2ZVWd8QzFxZQ+8ggf3X13baVUifLRtGm0Ki2lRffuFBQX06OsjIoGY7vq3XfpfNJJALQ+6CAKmzZl7Ycf1m4MgX3PP58F48blOnRtRfW0aRSWllLQvTsUF9O0rIzKBmNb0KnTxvclZ51FzZw5AKx/993ahxMVFkJREcX9+2/cJmVTEhLRO4HrYozb8/0N/wB6hRCahBBaAyfV2bYS2HCz4IvAcSGEAwBCCM1CCD0z94m2jjE+Ru3DjQ7fzj605vMHLV3SYNvt1FY5x8cYazLr/gaM3NAghLDhvM+SmcobQhgCtN3OeHYpRYWFXHv55Xxz1ChO+853GHLssZTusw+j77tv40OLhp58Mp+sXMmgK67gDxMmcPXXvw7ARUOG8NmaNZzx3e8y9Ac/4NyBAzmoe/c89kZfxLn33cdlU6fS/sAD+ddFi+h76aX5Dknbo6aGd0eOpLS8nEPmzGH5+PGsnT2bzqNG0frMMzc2azdsGMv9AJt8NTUsHDmSA8vLOWzOHJaNH8+a2bPpMmoUbeqMZ/thw/i4wXi2u+ACWp5wAnsOH77x612a+RUfiRFranhp5EhOLi/nq3PmsHD8eD6ZPZvDR42iW2ZsX/n+9+l5+eWc+dprnHD//Tw/fPjG/TuecAKrKypYtWBBnnqgzaqpYdXIkbQuL6fdnDmsGz+emtmzaTZqFCWZsd3jyitrv6LltdfY48orWZkZ23UPPUTN22/TduZM2s6YQfWMGVROnLiFkymriop2/iuhQmP35+XkxCGsijG2aLBuAHB1jPGMEMJw4EsxxpEhhLuAiTHGhxruG0L4FXA2MA+oBCbEGO8KIXwH+DawOHOf6EDgeqBJ5nTXANOAR6mtYAbghhjjHxvEtDGOza0PIZwN3ERtMvoicFSMcUCmXTHwMdAvxjg3s25PYAxwMFAEPBtj/OcQQntqK8N7As8A5wJHZu4rbdzs2d58sxsadcghW2+kXdIZ+Q5AWVOd7wCUNT62Zfd2Wr4DUNZ0iDG581I3p6Rk53+2r6xM5M8hb4loWoQQvkTtg4myc9e3iehuyUR092UiuvsyEd19mYju3kxEd1+7ZCLarNnO/2y/enUifw7JrdXuBkII/wZcwedPzpUkSZKkxiV4Ku3OloR7RHdbMcZfxhj3jTFOyXcskiRJkpQU6Um5JUmSJCnJrIhKkiRJkpQd6Um5JUmSJCnJUlQRTU9PJUmSJCnJUpSIOjVXkiRJkpRT6Um5JUmSJCnJrIhKkiRJkpQd6Um5JUmSJCnJUlQRTU9PJUmSJCnJUpSIOjVXkiRJkpRT6Um5JUmSJCnJrIhKkiRJktIghHBqCOHNEML8EMK/baHd0BBCDCF8aUfPmZ6UW5IkSZKSLA8V0RBCITAGGARUANNCCBNijLMbtGsJXAm8tDPOa0VUkiRJktKrHzA/xvhOjLESGAec3Ui7/wB+BazdGSc1EZUkSZKkJCgq2umvEMKIEMIrdV4jGpy1C7CoznJFZt1GIYS+QLcY48Sd1tWddSBJkiRJ0g7IwtTcGONYYOwWmoTGdtu4MYQC4CZg+M6My4qoJEmSJKVXBdCtznJX4P06yy2B3sDfQwgLgaOBCTv6wCIropIkSZKUBPn5+pZpQGkIoQfwHlAGfG3Dxhjjp8CeG5ZDCH8Hro4xvrIjJ7UiKkmSJEkpFWOsBkYC5cAcYHyM8Y0QwnUhhLOydV4ropIkSZKUBPmpiBJjfAx4rMG6azfTdsDOOKeJqCRJkiQlQZ4S0Xxwaq4kSZIkKafSk3JLkiRJUpJZEZUkSZIkKTvSk3JLkiRJUpKlqCKanp5KkiRJUpKlKBF1aq4kSZIkKafSk3JLkiRJUpKlqCKanp7urtq0yXcEyoIz8h2AsmZivgNQ1pyb7wCUNUfkOwBllZ+kpPwwEZUkSZKkJEhRRdR7RCVJkiRJOZWelFuSJEmSkixFFdH09FSSJEmSkixFiahTcyVJkiRJOZWelFuSJEmSksyKqCRJkiRJ2ZGelFuSJEmSkixFFdH09FSSJEmSkixFiahTcyVJkiRJOZWelFuSJEmSksyKqCRJkiRJ2ZGelFuSJEmSkixFFdH09FSSJEmSkixFiahTcyVJkiRJOZWelFuSJEmSksyKqCRJkiRJ2ZGelFuSJEmSkixFFdH09FSSJEmSkixFiahTcyVJkiRJOZWelFuSJEmSksyKqCRJkiRJ2ZGelFuSJEmSksyKqCRJkiRJ2ZGelFuSJEmSkixFFdH09FSSJEmSkixFiahTcyVJkiRJOZWelFuSJEmSksyKqCRJkiRJ2ZGelFuSJEmSkixFFdH09FSSJEmSkixFiahTcyVJkiRJOWUiqrx59uWXGXzxxQy66CLG3nffJtsrKyv53qhRDLroIs6/4goqPvig3vb3lyyh75Ah3PHAAxvX3fXgg5w+fDhnfOMbXPUf/8G6ysqs90PbrtXgwRwydy6HzJtHxx/9aJPtXW+8kYOhCHDtAAAgAElEQVSnT+fg6dM55M036bN8eR6i1M5w1h13cPWSJVwxc2a+Q9F2aDF4MD3nzqXnvHl0aORaBWh9/vmUvvEGpbNm0e3eezeu73T99ZTOmkXp7Nl0Hj06VyFrGzm2u68weDBFc+dSNG8eBY2MbbjkEoqWLqVo+nSKpk8nXHZZ7YY+fSh84QWKZs2iaMYMwgUX5Dhy1VNUtPNfCbXVRDSEUBNCeC2E8EYIYUYI4aoQwrbs9+vMPr/eOaFCCOH2EEKvnXCcv4cQ3sz0Z1oI4fBt2GdhCGHPHT33tgohDAghTMzV+XKtpqaG60aP5vZf/pJJd93FxKeeYv7ChfXaPPjYY7Rq2ZIn7r2X4eefzw233VZv+y/GjOH4L3954/KSDz/k7ocf5s+33cbEP/yBmpoaJk2enIvuaFsUFLDPmDHMGzKE2b160W7YMJoefHC9JhVXXcWcvn2Z07cvS2+5hU8efjhPwWpHvXbXXfzp1FPzHYa2R0EBe48Zw4IhQ5jXqxethw2jSYNrteSAA+jw4x/z9nHHMa93b97/3vcAaHbMMTQ77jjmHXYY83r3ptlRR9G8f/989EKNcWx3XwUFFI4ZQ/WQIVT36kXBsGHQYGwB1j/wANV9+1Ldty/xjjtqV65eTc3FF1PduzfVp55K4W9+A61b57gDSqNtqYiuiTEeHmM8BBgEnAb8dBv2+xZwRIzxBzsSYF0xxm/GGGfvpMNdFGPsA/wW2GnJsrbN63Pnsu/ee9Nt770pKS7m9IEDeer55+u1mfz885wzeDAAg/v3Z+qrrxJjBODJKVPouvfelHbvXm+fmpoa1q5bR3Xmv3u1b5+T/mjrmvfrx9r586lcsIBYVcXyceNoc/bZm23fbtgwlt1/fw4j1M707nPPsWbZsnyHoe3QrF8/KufPpypzrX46bhytGlyr7S6/nI/HjGH9J58AUPPhh7UbYqSgaVNCSQmhSRMoLqZ6yZJcd0Gb4djuvkK/fsT582HBAqiqYv24cRRs4XdsPfPmwfz5te8XL4alS6FDh+wFqy2zItq4GONSYAQwMtQqzFQ+p4UQXg8hfAsghDABaA68FEK4MITQIYTw50y7aSGE4zLtfhZCuDNToXwnhHBlZn3zEMKkTMVyVgjhwsz6v4cQvhRCuCKE8KsNcYUQhocQbsm8/6cQwsuZKu5tIYTCrXRrKtClzrFOCSFMDSG8GkJ4MITQok7bH2SO/XII4YBM+31DCE9l+v9UCGGfzPq7QghD6xx3Vea/AzL9eCiEMDeEcG8IIWS2nZpZNwU494uMza5myUcf0WmvvTYud+zQgSUffbRJm86ZNkWFhbRs0YLlK1awes0afn///Yy85JJ67Tt26MClF1zAiRdeyFfOO48WzZvzlaOOyn5ntE2Ku3ShatGijcuVFRUUd+nSaNuSffahSY8erLSiLeVcUYNrtaqRa7WkZ0+a9OzJflOmsP/UqbTI/NFw9Ysvsurppzl48WIOXryYVeXlrJs7N6fxa/Mc291Yly5QZ2xjRUXtugYKzjuPohkzKHzwQejadZPt4aijoKQE3n47q+FqC0xENy/G+E5mv72Ay4BPY4xHAUcBl4cQesQYz+LzSuoDwGjgpky784Db6xzyIGAw0A/4aQihGDgVeD/G2CfG2Bv43wZhPET9RO1C4IEQwsGZ98fFGA8HaoCLttKlU4G/AGSm3l4DnBxjPAJ4BbiqTtsVMcZ+wK3AbzLrbgXujjEeBtwL3LyV8wH0Bb4H9AL2A44LITQFfg+cCRwPdNrcziGEESGEV0IIr4z905+24XTJs6GyWVcmH99yG+CWu+7ikqFDab7HHvW2fbpyJU+98AJP3X8/zz30EGvWruXRJ57YqXFrBzQYXwAaGWOAtmVlLH/oIVi/PstBSdpEI9dqw3+PQ1ERJaWlvDNgAO8OG0bX22+noHVrSvbfnyYHH8zcrl2Z26ULLQYOpNnxx+cqcm2NY7v72obfsfGvf6W6e3eq+/QhPvkkhX/8Y/32nTpReM891HzjG5v9/SztTNubIm/4v/0U4LA6lb/WQCmwoEH7k4FedRKNViGElpn3k2KM64B1IYSlQEdgJnBDCOF6YGKM8bm6B4sxfpipoB4NzAMOBJ4Hvg0cCUzLnGsPYOlm+nBvCKE5UAgckVl3NLXJ4fOZ/UuorZhucH+d/96UeX8MnyfF9wC/YutejjFWAIQQXgO6A6uABTHGeZn1f6K2+ryJGONYYCwA77+/S/5L0alDBz5Y+vnQLPnww02m0Xbq0IHFS5fSqUMHqmtqWLlqFW1atWLGnDmUP/MMN9x2GytWraKgoIAmJSW0b9uWrp060a5NGwBOOf54ps+axdmDBuW0b2pcVUUFxd26bVwu6dqVqvffb7Rtu7Iy3v32t3MVmqQ6qhtcq8Vdu1Ld4Fqtqqhg9YsvQnU1VQsXsu7NN2lSWkrzAQNY8+KLrP/sMwBWPv44zY4+mtXP1fs1rjxxbHdjFRVQZ2xD167Q8Hdsndsl1v/+9xRdf/3n21q2pGjSJGquuYb40kvZjlZbkuAK5s72hSuiIYT9qK00LqU2If1OpvJ5eIyxR4zxb5s5zzF12nWJMa7MbFtXp10NUBRjfIvahHIm8IsQwrWNHPMB4AJqK6yPxNo/6QXgj3XOc2CM8Web6cpFQA/gPmDMhu4BT9TZv1eM8bI6+8TNvKeR9dWZfpOZeltSp80mfd7KMXc7hx50EAvfe49FixdTWVXFpMmTGXjssfXaDDz2WB4pLweg/JlnOLpvX0II3HfzzUweN47J48ZxydChfOuii/inc85h7732Ysbs2axZu5YYI1NffZX99903H91TIz6bNo2mpaWUdO9OKC6mbVkZn0yYsEm7Jj17Uti2LZ9NndrIUSRl2+pp02hSWkpx5lptXVbGigbX6oq//IUWJ54IQGH79pT07EnlO+9Q9e67tQ+wKSyEoiKa9+/Pujlz8tENNcKx3X3FadMIpaXQvTsUF1NQVsb6hr9jO30+2S6cdRZxw/gVF1P4yCOsv/tu4kMP5S5opd4XSkRDCB2A3wG3ZhK/cuCKzHRaQgg9M1XGhv4GjKxznC0+pTaEsDewOsb4J+AGPq9Y1vUw8FVgGLVJKcBTwNAQwl6Z47QLIWw2E4kxVlE7FffozLTeF6mdJrvh/s9mIYSedXa5sM5/N3xKfgEoy7y/CJiSeb+Q2mQa4GygeEt9BuYCPUII+2eWh22l/S6tqLCQa6+8km/+8IecNnw4Q048kdIePRh9550bH1o09PTT+WTFCgZddBF/ePBBrh7RaIF4oz69ejG4f3/OGTGCMy+9lPUxcuEZZ+SiO9oWNTW8O3IkpeXlHDJnDsvHj2ft7Nl0HjWK1meeubFZu2HDWD5uXB4D1c5w7n33cdnUqbQ/8ED+ddEi+l56ab5D0raqqeH9kSPpUV5O6Zw5fDp+POtmz2avUaNomblWV5WXU/Pxx5S+8Qb7Pf00H/zgB9QsW8anDz3EurffpnTmTEpnzGDtjBmsnLjbPgB+1+PY7r5qaqgZOZKi8nKK5sxh/fjxMHs2BaNGETJjW3DllbVf0fLaaxRceSU1w4cDEC64gHDCCRQMH77xq13o0yePnUm5FN0jGhq7D69egxBqqK1MFlNb5bsHuDHGuD7zNS7/Se19jQH4EPhqjPHTEMKqGGOLzDH2pLbqeDC11b9nY4z/HEL4GbAqxnhDpt0s4Axqp9r+GlgPVAFXxBhfCSH8Hbg6xvhKpv1EoFeMcb868V4I/JjaJLsK+HaM8cUGfWp4nO9njnNZCGEgcD3QJNP8mhjjhBDCQuAP1D41uAAYFmOcH0LoDtwJ7Jnp/zdijO+GEDoCj2baPkVt5bhFCGFA5txnZM59K/BKjPGuEMKp1N57+hG1CW3vDe02axedmqst+7/NPMRHuz4/tu2+dusnzEm7sYPyHYCypjjGRm6eTbgJE3b+Z/uzzkrkz2GriagSzkR0t2QiuvsyEd19mYhKuyYT0d3XLpmIPvbYzv9sf9ppifw5JLdWK0mSJElpkuCptDvbF35YkSRJkiRJOyI9KbckSZIkJZkVUUmSJEmSsiM9KbckSZIkJVmKKqLp6akkSZIkJVmKElGn5kqSJEmScio9KbckSZIkJZkVUUmSJEmSsiM9KbckSZIkJVmKKqLp6akkSZIkJVmKElGn5kqSJEmScio9KbckSZIkJZkVUUmSJEmSsiM9KbckSZIkJZkVUUmSJEmSsiM9KbckSZIkJVmKKqLp6akkSZIkJVmKElGn5kqSJEmScio9KbckSZIkJZkVUUmSJEmSsiM9KbckSZIkJVmKKqLp6akkSZIkJVmKElGn5kqSJEmScio9KbckSZIkJZkVUUmSJEmSsiM9KbckSZIkJZkVUUmSJElSThUV7fzXNgghnBpCeDOEMD+E8G+NbG8SQnggs/2lEEL3He2qiagkSZIkpVQIoRAYAwwBegHDQgi9GjS7DFgeYzwAuAm4fkfPm57aryRJkiQlWX6m5vYD5scY3wEIIYwDzgZm12lzNvCzzPuHgFtDCCHGGLf3pCaiu7o998x3BMqC6nwHoKw5N98BKGsezncAypqR+Q5AWVXcrVu+Q5CyKoQwAhhRZ9XYGOPYOstdgEV1liuALzc4zMY2McbqEMKnQHvgo+2Ny0RUkiRJkhJgfRbunMwknWO30CQ0ttt2tPlCvEdUkiRJktKrAqg7NaAr8P7m2oQQioDWwLIdOakVUUmSJElKgOos3J9VUrLVJtOA0hBCD+A9oAz4WoM2E4BLgKnAUGDyjtwfCiaikiRJkpQI+UhEM/d8jgTKgULgzhjjGyGE64BXYowTgDuAe0II86mthJbtaFwmopIkSZKUYjHGx4DHGqy7ts77tcD5O/OcJqKSJEmSlADZqIgmlQ8rkiRJkiTllBVRSZIkSUqANFVETUQlSZIkKQHSlIg6NVeSJEmSlFNWRCVJkiQpAayISpIkSZKUJVZEJUmSJCkB0lQRNRGVJEmSpARIUyLq1FxJkiRJUk5ZEZUkSZKkBLAiKkmSJElSllgRlSRJkqQEsCIqSZIkSVKWWBGVJEmSpARIU0XURFSSJEmSEiBNiahTcyVJkiRJOWVFVJIkSZISwIqoJEmSJElZYkVUkiRJkhIgTRVRE1FJkiRJSoA0JaJOzZUkSZIk5ZQVUUmSJElKACuikiRJkiRliRVRSZIkSUqANFVETUQlSZIkKQHSlIg6NVd58+yUKQw+80wGnXYaY2+/fZPtlZWVfO/qqxl02mmc/7WvUfHeewAs/+QTvn7ppfTt14/rfv7zevvcdPPN9D/5ZPr265eTPmjrWg8ezGFz59Jn3jw6/+hHm2zf58Yb6T19Or2nT+ewN9/kyOXLAWjWpw+9XniBQ2fN4tAZM2h3wQW5Dl1b0WLwYHrOnUvPefPo0MjYArQ+/3xK33iD0lmz6HbvvRvXd7r+ekpnzaJ09mw6jx6dq5C1E5x1xx1cvWQJV8ycme9QtB2KBw+mzdy5tJk3j6aNXLdNLrmEtkuX0nr6dFpPn06Tyy4DoGjAgI3rWk+fTrs1ayg+++xch68t6d8fnn4ann0W/uVfNt/utNPg3XfhsMPqr997b5gzB0aMyG6cUoYV0R0UQqgB6v42/mqMceF2HGch8KUY40c7KbREq6mp4bqf/5w/jB1Lx06dGFpWxsATT+SA/fff2ObBhx+mVatWPPHYY0x6/HFuuOkmfnPDDTQpKeG7I0cyb/585s2bV++4J/bvz0XDhjH49NNz3SU1pqCA7mPGMHfQICorKjhk2jQ+mTCBNXPmbGzy7lVXbXzfceRImvXtC8D61at5++KLWTd/PsWdO9P7//6PT8vLqfn005x3Q40oKGDvMWNYMGgQ1RUV7D9tGismTGBdnbEtOeAAOvz4x7x93HGs/+QTCjt0AKDZMcfQ7LjjmJf5ELT/lCk079+fz555Ji9d0Rfz2l138fKtt3LO3XfnOxR9UQUFNB8zhhWDBrG+ooLW06ZRNWECNXWuW4DKBx7gs+98p9666r//nU8z/z6Htm1pM38+VX/7W85C11YUFMB//idcdBEsXgx//Ss88QQ0+JxE8+bwjW/Aq69ueoxrr4W//z0n4WrzrIjqi1gTYzy8zmthvgPaFbw+cyb77rMP3bp1o6S4mNOHDOGpp5+u12by009zzllnATB40CCmvvQSMUaaNWvGl444giYlJZsc9/A+fdgr82FX+deiXz/Wzp/PugULiFVVLBs3jrZb+At6+2HD+Pj++wFYO28e6+bPB6Bq8WKqli6lyLFNjGb9+lE5fz5VmbH9dNw4WjUY23aXX87HY8aw/pNPAKj58MPaDTFS0LQpoaSE0KQJFBdTvWRJrrug7fTuc8+xZtmyfIeh7VDUrx818+ezfsECqKpi3bhx21XVLBk6lKrHH4c1a7IQpbbL4YfDwoW1lc6qqtpE9JRTNm139dXwu9/BunX1159ySu2+b72Vk3AlMBHNihBC0xDCH0IIM0MI00MIJ2bWDw8h3Fqn3cQQwoAG+zYPIUwKIcwIIcwKIVyY4/BzYsnSpXTq1GnjcseOHVnS4IPokqVL6ZxpU1RURMsWLVie+UCrXUNJly5ULlq0cbmyooLiLl0ab7vPPjTp0YMVkydvsq35UUdRUFLCurffzlqs+mKKunShqs7YVjUytiU9e9KkZ0/2mzKF/adOpcXgwQCsfvFFVj39NAcvXszBixezqrycdXPn5jR+KY0KunRhfZ3rdn1FBYWN/Jtcct55tJ4xgxYPPkhB166bbG9SVsa6zB8NlRCdOsH773++vHgxdOxYv80hh0DnzvDUU/XX77EHXHEF/OY32Y9TW1VdvfNfSeXU3B23Rwjhtcz7BTHGc4BvA8QYDw0hHAT8LYTQcxuPdyrwfozxdIAQQuudHnECxBg3WRdC+MJtlHCNjVcj4wrQvqyMZQ89BOvX11tf3KkT+99zD+9ccslm91UeNDK2Da/ZUFRESWkp7wwYQHHXruz/3HO81bs3RXvuSZODD2Zu5gNujyeeYOXxx7P6uedyErqUWttw3Vb+9a+1SWZlJU2+9S1a/PGPrDjppM8P0akThYceSlV5edbD1Rewtd+3IdROvf3+9zdtd9VVcMcdsHp19uKTGmFFdMfVnZp7TmbdV4B7AGKMc4F/ANuaiM4ETg4hXB9COD7GuMkNcSGEESGEV0IIrzT2kJ9dQaeOHfnggw82Li9ZsoS99tprkzaLM22qq6tZuWoVbVrvlnn5bquyooKSbt02Lpd07UpV3b/Y1tG+rGzjtNwNClu25MBJk6i45hpWvfRSVmPVF1NdUUFxnbEt7tqV6gZjW1VRwYpHH4XqaqoWLmTdm2/SpLSUVuecw5oXX2T9Z5+x/rPPWPn44zQ7+uhcd0FKnfUVFRTUuW4LunZlfYPrNi5bBpWVAKz7/e8pPPLIetubXHABlY88kuwySxotXlz7sKENOneGpUs/X27RAg48EB54AJ5/Hvr2rU0+Dzus9v2Pf1y7/tJLYeRIuOSS3PdBQLoqoiai2bG5sl019X/mTRs2iDG+BRxJbUL6ixDCtY20GRtj/FKM8UsjvvnNnRFvzh3auzcL//EPFlVUUFlVxaTHH2fggAH12gwcMIBHJkwAoPyJJzi6Xz8roruYVdOm0bS0lCbduxOKi2lXVsbyzJjW1bRnT4ratmXV1Kkb14XiYkofeYSP7r67tlKqRFk9bRpNSkspzoxt67IyVjQY2xV/+QstTjwRgML27Snp2ZPKd96h6t13ad6/PxQWQlERzfv3r/eQI0nZUT1tGoWlpRR07w7FxTQpK6OqwXUb6tw2U3LWWZs8yKhk2DCn5SbRjBnQowd06wbFxXDmmbUPK9pg5cra+0iPO672NX06XHYZvP46DB36+fo774Rbb4U//jF/fUk5E1HtqGeBiwAyU3L3Ad4EFgKHhxAKQgjdgE2+YySEsDewOsb4J+AG4IhcBZ1LRUVFXPvv/843//mfOe2ssxgyeDClBxzA6Ftv3fjQoqHnnssnn3zCoNNO4w93383V3/vexv0HDh7ML3/9ax559FFOOOkk5mfuHfzVjTdywkknsWbtWk446SRu+e1v89I/ZdTUsHDkSA4sL+ewOXNYNn48a2bPpsuoUbQ588yNzdoPG8bH48bV27XdBRfQ8oQT2HP48I1f79KsT59c90CbU1PD+yNH0qO8nNI5c/h0/HjWzZ7NXqNG0TIztqvKy6n5+GNK33iD/Z5+mg9+8ANqli3j04ceYt3bb1M6cyalM2awdsYMVk6cmOcOaVude999XDZ1Ku0PPJB/XbSIvpdemu+QtK1qavhs5EhalZfTZs4cKsePp2b2bPYYNYrizHW7x5VX0nrWLFq/9hpNr7ySVcOHb9y9YN99KezWjWqfcJ08NTXwk5/APffA5MkwcWLtg4euugoGDcp3dFKjQmP34WnbhRBWxRhbNFjXFPgdtZXNauCqGOPTobac9yfgcGAW0BH4WYzx7xu+viWzz6+B9UAVcEWM8ZXNBlBZ6QDuhl5q0iTfIShLmuU7AGXNw/kOQFkzMt8BKKva15murN3Mu+/uclPp7r6bnf7Z/uKLNztbM698WNEOapiEZtatBYY3sj6SqZQ2sq175m155iVJkiRJuyUTUUmSJElKgCTf07mzmYhKkiRJUgKkKRH1YUWSJEmSpJyyIipJkiRJCWBFVJIkSZKkLLEiKkmSJEkJkKaKqImoJEmSJCVAmhJRp+ZKkiRJknLKiqgkSZIkJYAVUUmSJEmSssSKqCRJkiQlgBVRSZIkSZKyxIqoJEmSJCVAmiqiJqKSJEmSlABpSkSdmitJkiRJyikropIkSZKUAFZEJUmSJEnKEiuikiRJkpQAaaqImohKkiRJUgKkKRF1aq4kSZIkKaesiEqSJElSAlgRlSRJkiQpS6yISpIkSVICpKkiaiIqSZIkSQmQpkTUqbmSJEmSpJyyIipJkiRJCWBFVJIkSZKkLLEiKkmSJEkJYEVUkiRJkqQssSIqSZIkSQmQpoqoieiu7qOP8h2BsmBuvgNQ1hyR7wCUNSPzHYCy6tZ8B6Cs+d6iRfkOQVnSOt8BbIc0JaJOzZUkSdoCk1BJ2vmsiEqSJElSAlgRlSRJkiQpS6yISpIkSVICpKkiaiIqSZIkSQmQpkTUqbmSJEmSpJyyIipJkiRJCWBFVJIkSZKkLLEiKkmSJEkJkKaKqImoJEmSJCVAmhJRp+ZKkiRJknLKiqgkSZIkJUASK6IhhHbAA0B3YCH8//buPT6K+t7/+OuzuYAhgggo18PFhgpHRbxQFClIixTbYi2ocLCCtbW1Bym19Ng+jq3S9vTYy2mrlf5UBPGGULyViDYVFO8oSEQgYqHFQoQCCirhlmTz+f0xE9hsdpOAZHeTvJ+Pxz52duY7s9/Z78zsfubznVkud/fdScq2Bd4GHnf3KXUtVxlRERERERERSeaHwFJ3LwCWhq+T+RnwfEMWqoyoiIiIiIhIBsjEjChwCTA8HL4PWAbcGF/IzM4GTgb+ApxT30IViIqIiIiIiGSAxghEzexa4NqYUXe7+91HsIiT3X0bgLtvM7OTErxHBPg/4GvA5xqyUAWiIiIiIiIizVQYdNYZeJrZEqBzgkn/3cC3+Q7wlLtvMbMGzaBAVEREREREJAOkq2uuu38+2TQz225mXcJsaBdgR4Ji5wFDzew7QD6Qa2Zl7p70elIFoiIiIiIiIpLMImAScGv4/Of4Au4+sXrYzCYD59QVhIICURERERERkYyQoTcruhX4k5ldA2wGLgMws3OAb7v7N45moQpERUREREREJCF3/4AENyBy95VArSDU3ecCc+tbrgJRERERERGRDJChGdFGoUBUREREREQkA7SkQDSS7gqIiIiIiIhIy6KMqIiIiIiISAZQRlRERERERESkkSgjKiIiIiIikgFaUkZUgaiIiIiIiEgGaEmBqLrmioiIiIiISEopIyoiIiIiIpIBlBEVERERERERaSTKiIqIiIiIiGSAlpQRVSAqIiIiIiKSAVpSIKquuZI2L7z+OqOuuoqREydy97x5taaXl5czbcYMRk6cyGXXXUfpv/5VY/rW7dsZOHo0sxcsODRu7sKFfHHyZL509dXc8LOfcbC8vNHXQ+rWddQovrJ+PZdu2MBpN95Ya3qbHj246Nln+dKqVXx59Wq6jR4NQO//+A++XFx86HFVNEr7AQNSXX2pQ/6oUfRdv56+GzbQKUHbArS77DIK1q2jYO1aejz00KHxnX/5SwrWrqWgpIQut92WqipLA+WMGsUJ69dzwoYNtE7Qtq0mTaL9jh20Ky6mXXExra65BoDs4cMPjWtXXMyJ+/eTc8klqa6+HKUxs2czfft2rluzJt1VkaOQPWoU+evXk79hA60S7Lc5kyZx/I4d5BcXk19cTE643wJYjx7kFRWRX1JC/rp1WM+eqay6tFAKRI+AmUXN7E0zW2tmC80sr57yrzRgmdPqW05zFI1G+eltt3HPrbeyeO5cnly6lI3vvlujzMKnnqLt8cfzzEMPMfmyy/jNXXfVmP6/M2cy9DOfOfR6+86d3P/YYzx61108ee+9RKNRFj/7bCpWR5KwSITBM2eyZPRo/ty/P70nTKBdv341ypxx0038809/4smzzuKF8eMZ/Mc/ArBp3jwKBw6kcOBAXvza1yh79112r16djtWQRCIRus6cyabRo9nQvz/tJkygVVzb5n7qU3T60Y/4+5AhbDjtNLZOmwZA3nnnkTdkCBvOOIMNp51G3rnn0mbYsHSshSQSidBm5kw+Hj2aD/v3p9WECWTFtS1A+YIFfDRwIB8NHMjB2bMBqFy27NC4j0eMwPfto+Kvf031GshRenPuXB78whfSXQ05GpEIrWfOZO/o0ZT170/OhAlEEuy3FQsWUDZwIGUDB2GP3CQAACAASURBVFIR7rcAefffT/mvf01Z//6UDRqE79iRytpLjMrKY//IVApEj8x+dz/T3U8DyoFv11XY3c9vwDKnAS0uEH1r/Xp6du1Kj65dyc3J4YsjRrD05ZdrlHn25Ze5dNQoAEYNG8arq1bh7gAseeklunftSkGvXjXmiUajHDh4kMrw+aQOHVKyPpJYx0GD+HjjRso2baKqooJN8+fTIy474u7ktG0LQG67duzburXWcnpPmMCmhx9OSZ2lYfIGDaJ840YqNm3CKyr4aP582sa17Ynf/CYfzJxJ1YcfAhDduTOY4E6kdWssNxdr1Qpycqjcvj3VqyBJZA8aRHTjRqo2bYKKCg7On39UWc3cceOoePpp2L+/EWopjWHziy+yf9eudFdDjkLWoEFUbdyIh/ttxRHst5F+/SA7m8olS4IRe/dqv5WUUCB69F4EPgVgZjeEWdK1ZjatuoCZlYXPw81smZk9YmbrzewhC0wFugLPmdlzZpZlZnPD5awxs++lZc1SYPv779P5pJMOvT65Uye2v/9+rTJdwjLZWVkcn5/P7o8/Zt/+/cx6+GGmTJpUo/zJnTrx9csv58IrruCCsWPJb9OGC849t/FXRpLK69aNvVu2HHq9r7SUNt261Siz+pZb6HPllYzbsoXPPfUUr11/fa3l9L7iCgWiGSa7WzcqYtq2orSUnLi2ze3bl1Z9+9LnpZc45dVXyQ9PLO1bvpyy556j37Zt9Nu2jbKiIg6uX5/S+ktykW7dqIpp26rSUrLi2hYgd+xY2q1eTf7ChUS6d681vdX48RzUfiuSEtatGx6331qC/TZn7FjyV68mb+FCLNxvI3374h9+SN6jj5K/ahWtf/UriChESBdlRKVOZpYNjAbWmNnZwNXAZ4DBwDfNbGCC2QYSZD/7A32AIe5+O7AVuNDdLwTOBLq5+2nufjpwb5L3v9bMVprZyrsffPBYr15KVGc2Y5lZ/WWAP8ydy6Rx42hz3HE1pn20Zw9LX3mFpQ8/zIuPPML+Awf48zPPHNN6yxGKa1Oo3a69J0xg49y5PNKjB0svvpihDzxQY76OgwZRuW8fH65b1+jVlSPQgLa17GxyCwr4x/DhbJ4wge733EOkXTtyTzmFVv36sb57d9Z360b+iBHkDR2aqppLfRrQtuWFhezu1YuPBgygYskS8u+7r+YiOncm6/TTqSgqatSqikgowX5L3H5bWVjInl69KBswgMolS8ir3m+zs8keOpT906dTdu65RPr0IWfy5Mavs7R4umvukTnOzN4Mh18EZgPXAY+7+14AM3sMGAoUx837uruXhmXeBHoBL8WV+QfQx8z+ACwGEl5Y4+53A3cDsHVr7WitCejcqRP/irn+YPvOnbW60Xbu1IltO3bQuVMnKqNR9pSVcULbtqx++22Knn+e39x1Fx+XlRGJRGiVm0uH9u3p3rkzJ55wAgAXDR1K8dq1XDJyZErXTQ7bV1pKmx49Dr3O6969Vtfbgmuu4ZnwmqSdy5eT1bo1rTt25EDYjbP3+PHKhmagytJScmLaNqd7dyrj2raitJR9y5dDZSUV777LwXfeoVVBAW2GD2f/8uVU7d0LwJ6nnyZv8GD2vfhiStdBEqsqLSUS07aR7t2pimtbj+m+eXDWLPJ++csa01tdfjnljz+e2afiRZoRLy3F4vZbr2O/LZ81i9bhfuulpUSLi4NuvUDFE0+QNXgwFXPmpKDmEs+9qhGWmpm5x8ysVeaqvkb0THe/3t3LCZJ0DXEwZjhKgpMA7r4bGAAsA/4TuOcT1jdjnX7qqbz73nts2baN8ooKFj/7LCPOr3lJ7Yjzz+fx8Gx60fPPM3jgQMyMebffzrPz5/Ps/PlMGjeOb02cyJWXXkrXk05idUkJ+w8cwN15ddUqTtFd39Lq/RUraFtQQH6vXkRycug9fjylixbVKFO2eTNdPvc5ANqdeipZrVsfCkIxo+dll7Fp/vxUV13qsW/FCloVFJDTqxeWk0O78eP5OK5tP37iCfIvvBCArA4dyO3bl/J//IOKzZuDmxNlZUF2Nm2GDePg22+nYzUkgcoVK8gqKCDSqxfk5NBq/Hgq4trWOnc+NJw7ZgzRuPbLnTBB3XJFUiga7rcW7rc59ey32TH7bXTFCqx9e6xjx2DaiBFUlZSkqupSS7QRHplJGdFP7gVgrpndShCUXgp87Qjm3wMcD7xvZh2Bcnd/1Mz+Dsw91pXNFNlZWfxk6lS+8V//RbSqirGjR1PQuze3zZnDaZ/+NJ8bMoRxX/wiP/jFLxg5cSLt2rbldz/+cZ3LHNC/P6OGDePSa68lOyuLfgUFXPGlL6VojSQRj0Z5bcoUPl9URCQriw1z5vBhSQlnzpjBBytXsqWwkJXf/z7nz5pF/+99D9x5OaY70Mmf/Sz7SkspC8/SSgaJRtk6ZQq9i4ogK4vdc+ZwsKSEk2bMYP/KlewpLKSsqIjjL7qIgnXrIBrlXz/4AdFdu/jokUdoM2IEBWvWgDtlf/kLe558Mt1rJNWiUfZOmULbsG0PzplDtKSE42bMoHLlSioKCzlu6lRyxoyBykp81y7KYvbbSM+eZPXoQeXzz6dvHeSofHXePHoNH05ex458b8sWlt18M8XKijUN0Sj7p0yhTbjfVsyZQ1VJCa1mzCC6ciWVhYXkxu23+6v326oqDkyfTpulS8GM6BtvUD5rVlpXR1oGS3QdniRmZmXunp9g/A3A18OX97j772PLm9lwYLq7fykcfwew0t3nmtn1BNnPbQTXkN7L4Uz1j9z96Tor1US75krd7ktwgwFpHs5KdwWk0XRNdwWk0dyR7gpIo5pWfxFpotq5N7TnYsYwO3DMf9u7t87Iz0GBaFOnQLRZUiDafCkQbb4UiDZfCkSbNwWizZcC0UCmBqLqmisiIiIiIpIRMveazmNNgaiIiIiIiEhGaIy75mYm3TVXREREREREUkoZURERERERkYzQcrrmKiMqIiIiIiIiKaWMqIiIiIiISEZoORlRBaIiIiIiIiIZoeUEouqaKyIiIiIiIimljKiIiIiIiEhGUEZUREREREREpFEoIyoiIiIiIpIRqtJdgZRRRlRERERERERSShlRERERERGRjNByrhFVICoiIiIiIpIRWk4gqq65IiIiIiIiklLKiIqIiIiIiGQEZURFREREREREGoUyoiIiIiIiIhmh5WREFYiKiIiIiIhkBP2PqIiIiIiIiEijUEZUREREREQkI7ScrrnKiIqIiIiIiEhKKSMqIiIiIiKSEVpORlSBqIiIiIiISEZoOYGouuaKiIiIiIhISikjKiIiIiIikhGUERURERERERFpFMqIioiIiIiIZISqdFcgZZQRFRERERERkZRSRlRERERERCQjtJxrRBWINnV33pnuGkgjuDjdFZBGc0K6KyCNJqdHj3RXQRrJtC1b0l0FaUS/T3cFpNHcnO4KHJWWE4iqa66IiIiIiIiklDKiIiIiIiIiGUEZUREREREREZFGoYyoiIiIiIhIRmg5GVEFoiIiIiIiIhlB/yMqIiIiIiIi0iiUERUREREREckILadrrjKiIiIiIiIiklLKiIqIiIiIiGSElpMRVSAqIiIiIiKSEVpOIKquuSIiIiIiIpJSyoiKiIiIiIhkBP19i4iIiIiIiEijUEZUREREREQkI+gaUREREREREZFGoYyoiIiIiIhIRmg5GVEFoiIiIiIiIhmh5QSi6porIiIiIiIiKaWMqIiIiIiISEZQRlRERERERESkUSgjKiIiIiIikhGq0l2BlFEgKiIiIiIikhHUNVdERERERESkUSgjKiIiIiIikhGUERURERERERFpFApERUREREREMkK0ER6fjJmdaGbPmNmG8Ll9knK/MrN1Zva2md1uZlbXchWIioiIiIiIZITMC0SBHwJL3b0AWBq+rsHMzgeGAGcApwHnAsPqWqgCUREREREREUnmEuC+cPg+4CsJyjjQGsgFWgE5wPa6FqpAVEREREREJCNUHfOHmV1rZitjHtceYaVOdvdtAOHzSfEF3P1V4DlgW/gocve361qo7porIiIiIiLSTLn73cDddZUxsyVA5wST/rsh72FmnwL6Ad3DUc+Y2Wfd/YVk8ygQFRERERERyQjp+fsWd/98smlmtt3Murj7NjPrAuxIUOxSYLm7l4XzPA0MBpIGouqaK5nhU5+CqVPhu9+FoUNrTz//fJgyBb7zHZg8Gdq1OzytXTu46iq4/vqgzAknpKzaUr+cUaNov349J27YwHE33lhreqtJk+iwYwfti4tpX1xM62uuOTStzS9/Sfu1a2lfUkKb225LZbWlAWzUKLLXryd7wwYiCdrWJk0ie8cOsouLyS4uxqrbdsAAsl55hey1a8levRq7/PIU11zqNWwYPPccvPBCcNxN5uKLYfNmOOOMmuO7doW334Zrj7T3lzS27FGjyF+/nvwNG2iVYL/NmTSJ43fsIL+4mPziYnJijsnWowd5RUXkl5SQv24d1rNnKqsun8CY2bOZvn07161Zk+6qSNO0CJgUDk8C/pygzGZgmJllm1kOwY2Kmn7XXDMrc/f8FL3XVOA6YJW7T4wZnwfMIrgTlAEfAl+ojvqTLOtd4Bx3f79RK93UmcGXvgT33Qcffwzf+hasXw87dx4us20b3HUXVFTAuefCRRfBwoXBtK9+Nfix9Pe/Q24uuKdnPaS2SITjZ87kw5EjqSotpf2KFZQvWkT07ZrHpYMLFlB2/fU1xmWfdx45Q4awO/yBe8JLL5EzbBgVzz+fsupLHSIRsmbOpHLkSCgtJXvFCqoWLQqCjxhVCxZQFde27NtH9KqrYONG6NKF7DfeoLKoCD76KIUrIElFIvDzn8PEicGxt7AQnnkGNmyoWa5NG7j6ali1qvYyfvITWLYsJdWVIxCJ0HrmTPaOHImXlpK/YgUVixZRFbffVixYwIH4/RbIu/9+Dv7P/1C5ZEnQ/lVVqaq5fEJvzp3L63fcwaX335/uqki90pMRrcetwJ/M7BqCgPMyADM7B/i2u38DeAQYAawhuHHRX9y9sK6FNolANMW+A4x2901x478LbHf30wHM7NNAxbF8YzPLcveM3PoaVffusGsX7N4dvF6zBk49tWYguimmObZsgQEDguFOnYIfTX//e/C6vDw1dZYGyR40iOjGjVSF7Xdg/nxyL7mE/W/XeYIs4A6tWwcnF8wgJ4eq7XXefE1SyAYNwjduPLRvVs2fT+SSS2r9oE0oNqDZtg127Aj2ZQWimeHMM+Hdd4NMJwSB6EUX1Q5Ep0+HO++snfW86KJg3v37U1JdabisQYOo2rgRD/fbivnzybnkEg42YL+N9OsH2dlBEAqwd29jVlWOsc0vvkg7ZbCbiMwLBdz9A+BzCcavBL4RDkeBbx3Jcpts11wz62lmS83srfD538Lxp5jZcjNbYWY/NbOEGUszu8HM1oaPaeG4O4E+wCIz+17cLF2A96pfuPs77n4wnO8JM3sj/APXhP2QkpUxs7Kwnq8BN5nZ4zHTRprZY0f1ATUlxx9f8wfoxx9D27bJy5999uEfRB06wIEDMH48XHdd8AOo7v/OlRSKdOtGdMuWQ6+rSkvJ6tatVrncsWNpv3o1bRcuJNI9uMa9cvlyKp57jg7bttFh2zbKi4qIrl+fsrpLPbp1C04Khby0NBgXJzJ2LNmrV5O1cGFw0imOnXtucLKh+mSSpF/nzrB16+HX27bBySfXLPPv/w5dusDSpTXHH3dccCz+/e8bv55yxKxbNzzumGwJ9tucsWPJX72avIULsXC/jfTti3/4IXmPPkr+qlW0/tWvghPBIiJHqSkfQe4A7nf3M4CHgNvD8bcBt7n7ucDWRDOa2dnA1cBnCC6i/aaZDXT3b4fzXOjuv4ubbQ5wo5m9amY/N7OCmGlfd/ezgXOAqWbWIcHbJivTBljr7p8Bfgr0M7NO4bSrgXsb9nE0YYkCx2Tda884I7j26KWXgteRCPTsCUVFQdfd9u1h4MDGq6scmQa0bXlhIbt69WL3gAGUL1nC8fcFf1MVOeUUsvr144Pu3fmgWzdyR4wgJ9H1w5IeDWhbLyykslcvKgcMwJcsIeu++2qW79yZrAceIHr11epSn0nqa1uzoOvtz39eu9wNN8Ds2bBvX+PVT45eA/bbysJC9vTqRdmAAVQuWUJe9X6bnU320KHsnz6dsnPPJdKnDzmTJzd+nUVanGgjPDJTUw5EzwPmhcMPABfEjA8vHjw0Pd4FwOPuvje8xvMxoM5fuO7+JkG29NfAicAKM+sXTp5qZquB5UAPoCDBIpKViQKPhu/h4bpcaWYnhOvydPyCYv8L6O5E1+Y0NR9/XPPmQ23bwp49tcv16RPcQGPePIhGD8+7bVvQrbeqKri2tEuX1NRb6lVVWkpWjx6HXke6dye6teb5Id+161CX6gOzZpF99tkAtLr0UiqXLw+6f+3dS/nTT5M9eHDqKi91Ky2FmLa17t1rZtEg6HIftm3VrFlY2LYAHH882YsXE73pJvy111JRY2mobduCE37VunQJuk9Xy8+HT38aFiyAl18OTv7Nnh2cKBw4EH70o2D8178e3EBu0qTa7yFp4aWlWNwx2es4JpfPmkVWuN96aSnR4uKgW280SsUTT5B11lmpq7yINDtNORCNdySn04+q76a7l7n7Y+7+HeBB4GIzGw58HjjP3QcAxUDrGm9Wd5kDcdeF3gtcCUwAFrp7ZYJ63O3u57j7Odc2hy+B996DE08M7nablQWnnx4ElLE6d4YxY+Chh2pel/Lee0FXsLy84HXv3jWvLZW0qlyxgqyCAiK9ekFODq3Hj6d80aIaZSKdD/9lVe6YMYduZFS1eTM5w4YF20R2NjnDhtW6yZGkj69YgRUUQNi2kfHjg5sVxYppWxszBq9uv5wcsh5/nKr778cfeSR1lZaGWb06OJb26AE5OfDlLwc3K6q2Z09wHemQIcGjuBiuuQbeegvGjTs8fs4cuOOO4EZ0khGi4THZwv02Z/x4KuL2W4vZb7NjjsnRFSuw9u2xjh2DaSNGUFVSkqqqi7QgLScj2pRvVvQKMJ4ggzgRCPtqshwYCywIpyfyAjDXzG4lCEovBb5W15uZ2RCgxN13m1ku0B9YBrQDdrv7PjM7laCrb7yGlAHA3bea2VbgJmBkXXVqNqqqYPHi4C9YIpHgDow7d8KIEUGg+c47MGpUcB3ZFVcE83z0UZAZdQ+65U6eHHQ52roV3ngjrasjMaJRyqZMoV1REZaVxYE5c4iWlJA3YwaVK1dSXljIcVOnkjtmDFRWUrVrF3vCrl4HH3mEnBEjaL9mDbhT/pe/UP7kk+ldHzksGiU6ZQrZRUWQlUXVnDlQUkJkxgx85Uq8sJDI1KlEwrb1XbuIhm1rl1+OffazWIcORMJxlZMnBwGQpF80Cj/+MTzwQHAiaMEC+Nvfgm63a9bUDEqlaYlG2T9lCm3C/bZizhyqSkpoNWMG0ZUrqSwsJHfqVHJi9tv91d1vq6o4MH06bZYuBTOib7xB+axZaV0dabivzptHr+HDyevYke9t2cKym2+meM6cdFdLEmo5d6M2bwLX5ZhZFTWv9/wtQXfaOUBHYCdwtbtvDq/dfJAgwFwMXOvuta7EN7MbgK+HL+9x99+H498lwV+umNlVwPRwuZFw2TcCucATQDfgHaATcIu7L6teFrCnjjK1/prGzMYD09y9/n6IP/lJ5jegHLGdP/tZuqsgjUT/ctt85cR0eZTm5aOYG/xI86NbazVfN7s3uTtYmv3smP+2d/9xRn4OTSIj6u7JuhCPSDDuPWCwu3sY0K1MsszfEgS08eN7JSl/P5Doz5cOAqOTzBO7rGRlEv0/6gUE/1kqIiIiIiItRuZ2pT3WmkQgeoTOBu4wMwM+5HDWs0kwszeAvcD3010XERERERGRxtDsAlF3fxEYkO56HK3wL15ERERERKTFUUZUREREREREUqrlBKLN6e9bREREREREpAlQRlRERERERCQjtJy/b1FGVERERERERFJKGVEREREREZGMoGtERURERERERBqFMqIiIiIiIiIZoeVkRBWIioiIiIiIZISWE4iqa66IiIiIiIiklDKiIiIiIiIiGUEZUREREREREZFGoYyoiIiIiIhIRqhKdwVSRoGoiIiIiIhIRlDXXBEREREREZFGoYyoiIiIiIhIRlBGVERERERERKRRKCMqIiIiIiKSEVpORlSBqIiIiIiISEZoOYGouuaKiIiIiIhISikjKiIiIiIikhGUERURERERERFpFMqIioiIiIiIZISqdFcgZZQRFRERERERkZRSRlRERERERCQjtJxrRBWIioiIiIiIZISWE4iqa66IiIiIiIiklDKiIiIiIiIiGUEZUREREREREZFGoYyoiIiIiIhIRmg5f99i7p7uOog0iJld6+53p7secuypbZsvtW3zpbZt3tS+zZfaVjKFuuZKU3JtuisgjUZt23ypbZsvtW3zpvZtvtS2khEUiIqIiIiIiEhKKRAVERERERGRlFIgKk2JrmdovtS2zZfatvlS2zZvat/mS20rGUE3KxIREREREZGUUkZUREREREREUkqBqIiIiIiIiKSUAlHJKGZ2qZm5mZ2a7ro0V+Hn+0DM62wz22lmT4avx5jZDz/B8qeZWd4xqOfksF5vmlmJmX0zZvwdn3T59bz3u2bWsTHf40iYWTT8HNaZ2Wozu8HM6j1+m9mvw3l+fQzrco+Z9T8Gy1lmZu+E67PCzM5swDwpbRczG169X7QEMdtZ9aPXUS4no/afliymTdea2cL6js1m9koDlnlMjvEtmZmVpfC9pprZ22b2UNz4PDN7yMzWhNvHS2aWX8+ytG/LMZWd7gqIxJkAvASMB25Jb1Warb3AaWZ2nLvvB0YC71VPdPdFwKJPsPxpwIPAvk9Uy8ACd59iZicB68zsk9SrKdvv7mcChJ/FPKAdcHM9830L6OTuB49VRdz9G8dqWcBEd19pZlcDvybYFiV9Dm1n0mzEHjseAr4N/DZZYXc/vwHLPJbHeGl83wFGu/umuPHfBba7++kAZvZpoOJYvrGZZbl79FguU5oXZUQlY4Rn4oYA1xAEophZxMz+GGZ1njSzp8xsXDjtbDN73szeMLMiM+uSxuo3NU8DXwyHJwAPV0+IzTia2Vwzu93MXjGzf8R89jUyRWZ2RzjfVKAr8JyZPRdOu8jMXjWzVeEZ+fxw/K1hpvMtM/tNXZV19x3A34GesePN7Mtm9pqZFZvZEjM7OdxmNphZp7BMxMw2mllHM+tkZo+GGbgVZjYkLNPBzP4aLucuwI7+o21c4WdxLTDFAllh5nNF+Fl+CyAM2tsAr5nZFXWs+y1mNifMUP4jbEPMrI2ZLQ4zlmvN7Ipw/DIzO8fMrjOzX1XXK2z/P4TDV5rZ62Em5i4zy6pntV4FusUsK+E2E/pBuOzXzexTYfmeZrY0XP+lZvZv4fi51dts+LosfB4erscjZrY+zApYOO0L4biXgK8eeQs1L2bW2szutSBrUmxmF4bja/RMCI/Pw+PmTbgNSdq8CFTvMzeEbbLWzKZVF6hvH4k/xofHn7nhctaY2ffSsmbNQB3HsVPMbHl43P6pJcmmJmpTM7sT6AMsStA2Xah5Evqd6pOWZvaEBb+t1pnZtUneL2EZMysL6/kacJOZPR4zbaSZPXZUH5A0T+6uhx4Z8QCuBGaHw68AZwHjgKcITpp0BnaH43LCMp3C8lcAc9K9Dk3hAZQBZwCPAK2BN4HhwJPh9MnAHeHwXGBh+Pn3BzaG4w+VD1/fAUwOh98FOobDHYEXgDbh6xuBnwAnAu9w+M7dJySoZ2w9+gA7wvlix7ePWcY3gP8Lh28GpoXDFwGPhsPzgAvC4X8D3g6Hbwd+Eg5/EfDqdciEB1CWYNxu4GSCoPSmcFwrYCXQO36+Otb9lnBfahW21wfh/jUWmBUzf7vweRlwDtCpensIxz8NXAD0AwqBnHD8H4GrEtR/GXBOODwN+EVd20zMtvXf4fBVHN5mC4FJ4fDXgSditt9x8Z8jwfb7EdCdYNt+Nax7a2ALUEBwMuJPxGznzf0BRAmOB28Cj4fjvg/cGw6fCmwOP6fJhPthOO1JYHhMO3VMtg3pkdI2rd7ms4E/A9cBZwNrCE5U5QPrgIFx5RPuI7HtGw6fDTwT8361juV6JG+XuHHJjmNPAhPC4W8nmbeuNj3UXnHznEnwvfoq8HOgIGbaieHzccBaoEOCtk9WxoHLw2ED1nP4t9o84Mvp/vz1yJyHuuZKJpkA/D4cnh++zgEWunsV8C8Ls2zAp4HTgGfCREYWsC211W263P0tC67/mkAQ6NflifDzLzGzk4/wrQYTBLAvh+2US/Cl9zFwALjHzBYTfNEmcoWZXQAcBL7l7rvC5VTrDiywIBueC1R3PZpD8KPr9wRf6PeG4z8P9I9ZRlszOx74LGH2y90Xm9nuI1zPdKheiYuAM2Iyf+0IAqn4bljJ1h1gsQdnwg+a2Q6CAHcN8Bsz+yVBMPZi7MLcfWeYQR0MbCDYJ18G/pPgR9GK8L2OI/ixk8hDZtaGYP89KxyXbJup9nDM8+/C4fM4nL18APgV9Xvd3UsBzOxNoBfBSZpN7r4hHP8gQaDfUiTqmnsB8AcAd19vZv8E+jZweXVuQ5ISx4XbNwQZ0dkEwejj7r4XIMxQDQWK4+ZNtI+8FFfmH0CfsDfEYuCvjbESLUSy49h5wFfC4XlAoh5EF9CwNj3E3d80sz4E3yGfJzhmn+fubwNTzezSsGgPgu+UD+IWkaxMFHg0fI/qe1JcaWb3hutyVZ2fgrQoCkQlI5hZB2AEwbWLTvDD1IHHk80CrHP381JUxeZoEcEX2nCgQx3lYq8vrI5iKqnZtb91knmN4Gz5hFoTzAYBnyPohj2FoP3jLXD3KXXU7Q/Ab9195bT3VgAABJtJREFUUdgt8BYAd99iZtvNbATwGWBiWD4CnOfBtbGxdYFge2sSwh8PUYIAz4Dr3b2ontnqWvfYNo4C2e7+NzM7G7gY+F8z+6u7/zRumQuAywnOeD8e/ugw4D53/1EDVmUisBq4FZhJ8CMs6TYT8iTDicoc2k7DeuXGlKm1zvUss6VK1k293mNAA7chaVy1Ti5Ud0NvgGT7yCHuvtvMBgCjCE5CXU5w8k8+uSM5Fh3V5STuXgY8BjxmZlXAxeEJ588TfF/sM7NlxO3f4fdtsjIHvOZ1ofcSZHsPECQWKo+mrtI86RpRyRTjgPvdvae793L3HgQZnfeBsRZc53cyQdAEQbfOTmZ2HoCZ5ZjZv6ej4k3YHOCn7r7mKOb9J0F2rZWZtSMIKKvtAaozbcuBIXb4Wr48M+trwTV/7dz9KYJumUd7g5R2HL7GZVLctHsIbqjxp5gvxb8SBL2E9al+3xcIg1UzG03Q5TcjWXDt650E3SIdKAKuM7OccHrfMMsYL9m6J3ufrsA+d3+Q4ITFWQmKPUZwpn4CQVAKsBQYZ8FNlTCzE82sZ4J5AXD3CuAmYLCZ9SPJNhMzyxUxz9WZ0lcIrysnaMfqrM27BNlZgEsIeljUZT3Q28xOCV8nC4Zbkth9oy9Bt+53CD7bM8Njcw9gUPyMDdyGJPVeAL4S7lttgEsJsqUNdegYb8EdVCPu/ijwY9TGn0Sy49hygm7uxEyPd8RtamZDzKx9OJxL0BPlnwTfq7vDAPNUgl4q8RpSBgB33wpsJTjOz62rTtLyKCMqmWICQVYk1qME15uVElx/8DfgNeAjdy8PuyLeHgZC2QTdMNelrspNW9jl6rajnHeLmf0JeIugW2Zs95+7gafNbJu7X2hmk4GHzaxVOP0mgh8yfzaz1gRnco/2Bhe3AAvN7D2CL+veMdMWEZyJvTdm3FRgppm9RbDNvEBwzc2MsI6rgOcJroPLJNXd63IIMlEPcPjOl/cQdJlbFWY6dnK4G1esZOuezOnAr8Oz5BUE3flqCLMhJUB/d389HFdiZjcBf7XgL2YqCDIl/0z2Ru6+38z+D5ju7tck2Wb+Fg63suAmGBEOB4pTgTlm9oNw/a8Ox88i2M5eJwiQ99axvrj7AQtuurHYzN4n+CF4Wl3ztAB/BO40szUE295kdz9oZi8TnCxcQ3B8XpVg3nq3IUk9d19lZnOB18NR97h70i6cCRw6xhOcSLzXDv+dVEN6QgjkmVlpzOvfkvw4Ng140My+T9D9+aP4hR1lm54C/L/weyMSLvtRgp4j3w6/K94h+G6N95cGlIn1EMF1oiX1lJMWpvomHyIZy8zy3b0s7L77OjDE3f+V7npJZjOzc4DfufvQdNdFRETkaFjwn637w0sfxhPcuOiSdNfrSFhwh+1id5+d7rpIZlFGVJqCJ83sBIKzdD9TECr1MbMfEmRfJtZXVkREJIOdDdwRZi4/pIldg2tmbxD0Rvl+uusimUcZUREREREREUkp3axIREREREREUkqBqIiIiIiIiKSUAlERERERERFJKQWiIiIiIiIiklIKREVERERERCSl/j+26Yl3Lf0F/gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1152x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Plot Correlation of Salary to Features\n",
    "\n",
    "plot.figure(figsize=(16, 8))\n",
    "snsPlot = sns.heatmap(onlyUsefulColumns.corr(), xticklabels=['Age', 'Minutes Played', 'Defensive Rebound', 'Fouls', 'Points', 'Log of Salary'], yticklabels=['Age', 'Minutes Played', 'Defensive Rebound', 'Fouls', 'Points', 'Log of Salary'],cmap=\"seismic\", annot=True, vmin=-1, vmax=1)\n",
    "snsPlot.set_xticklabels(snsPlot.get_xticklabels(), rotation = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We notice that all the input are positively correlated with our salary so this makes our problem ideal for a Linear Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 4: PREDICTION USING FEATURES SELECTED BY LASSO"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Validation R^2 score was: 0.43864438613970036\n",
      "Feature coefficient results:\n",
      "Age : 0.08\n",
      "MP : 0.05\n",
      "DRB : 0.15\n",
      "PF : 0.07\n",
      "PointsPerGame : 0.04\n"
     ]
    }
   ],
   "source": [
    "X_train, X_val, y_train, y_val, residuals, model = split_and_validate(X,y) ## Generated new Scores\n",
    "\n",
    "y_pred = model.predict(X)\n",
    "\n",
    "onlyPlayerStats[\"Salary_Log_predict\"] = y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Age</th>\n",
       "      <th>MP</th>\n",
       "      <th>ThreePoints</th>\n",
       "      <th>AST</th>\n",
       "      <th>FG</th>\n",
       "      <th>FT</th>\n",
       "      <th>TOV</th>\n",
       "      <th>DRB</th>\n",
       "      <th>FGA</th>\n",
       "      <th>...</th>\n",
       "      <th>ORB</th>\n",
       "      <th>FTA</th>\n",
       "      <th>STL</th>\n",
       "      <th>BLK</th>\n",
       "      <th>PF</th>\n",
       "      <th>PointsPerGame</th>\n",
       "      <th>Salary</th>\n",
       "      <th>Season</th>\n",
       "      <th>Salary_Log</th>\n",
       "      <th>Salary_Log_predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Giannis Antetokounmpo</td>\n",
       "      <td>24</td>\n",
       "      <td>32.8</td>\n",
       "      <td>0.7</td>\n",
       "      <td>5.9</td>\n",
       "      <td>10.0</td>\n",
       "      <td>6.9</td>\n",
       "      <td>3.7</td>\n",
       "      <td>10.3</td>\n",
       "      <td>17.3</td>\n",
       "      <td>...</td>\n",
       "      <td>2.2</td>\n",
       "      <td>9.5</td>\n",
       "      <td>1.3</td>\n",
       "      <td>1.5</td>\n",
       "      <td>3.2</td>\n",
       "      <td>27.7</td>\n",
       "      <td>24157304.0</td>\n",
       "      <td>NBA_2019</td>\n",
       "      <td>17.000097</td>\n",
       "      <td>17.290277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Khris Middleton</td>\n",
       "      <td>27</td>\n",
       "      <td>31.1</td>\n",
       "      <td>2.3</td>\n",
       "      <td>4.3</td>\n",
       "      <td>6.6</td>\n",
       "      <td>2.8</td>\n",
       "      <td>2.3</td>\n",
       "      <td>5.3</td>\n",
       "      <td>14.9</td>\n",
       "      <td>...</td>\n",
       "      <td>0.6</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>2.2</td>\n",
       "      <td>18.3</td>\n",
       "      <td>13000000.0</td>\n",
       "      <td>NBA_2019</td>\n",
       "      <td>16.380460</td>\n",
       "      <td>16.212891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Eric Bledsoe</td>\n",
       "      <td>29</td>\n",
       "      <td>29.1</td>\n",
       "      <td>1.6</td>\n",
       "      <td>5.5</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>2.1</td>\n",
       "      <td>3.6</td>\n",
       "      <td>12.4</td>\n",
       "      <td>...</td>\n",
       "      <td>1.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.4</td>\n",
       "      <td>2.0</td>\n",
       "      <td>15.9</td>\n",
       "      <td>15000000.0</td>\n",
       "      <td>NBA_2019</td>\n",
       "      <td>16.523561</td>\n",
       "      <td>15.897635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Brook Lopez</td>\n",
       "      <td>30</td>\n",
       "      <td>28.7</td>\n",
       "      <td>2.3</td>\n",
       "      <td>1.2</td>\n",
       "      <td>4.4</td>\n",
       "      <td>1.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.5</td>\n",
       "      <td>9.7</td>\n",
       "      <td>...</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.6</td>\n",
       "      <td>2.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>12.5</td>\n",
       "      <td>3382000.0</td>\n",
       "      <td>NBA_2019</td>\n",
       "      <td>15.033978</td>\n",
       "      <td>15.971340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Malcolm Brogdon</td>\n",
       "      <td>26</td>\n",
       "      <td>28.6</td>\n",
       "      <td>1.6</td>\n",
       "      <td>3.2</td>\n",
       "      <td>5.9</td>\n",
       "      <td>2.2</td>\n",
       "      <td>1.4</td>\n",
       "      <td>3.5</td>\n",
       "      <td>11.7</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.2</td>\n",
       "      <td>1.6</td>\n",
       "      <td>15.6</td>\n",
       "      <td>1544951.0</td>\n",
       "      <td>NBA_2019</td>\n",
       "      <td>14.250503</td>\n",
       "      <td>15.576392</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Name  Age    MP  ThreePoints  AST    FG   FT  TOV   DRB  \\\n",
       "0  Giannis Antetokounmpo   24  32.8          0.7  5.9  10.0  6.9  3.7  10.3   \n",
       "1        Khris Middleton   27  31.1          2.3  4.3   6.6  2.8  2.3   5.3   \n",
       "2           Eric Bledsoe   29  29.1          1.6  5.5   6.0  2.3  2.1   3.6   \n",
       "3            Brook Lopez   30  28.7          2.3  1.2   4.4  1.4  1.0   4.5   \n",
       "4        Malcolm Brogdon   26  28.6          1.6  3.2   5.9  2.2  1.4   3.5   \n",
       "\n",
       "    FGA  ...  ORB  FTA  STL  BLK   PF  PointsPerGame      Salary    Season  \\\n",
       "0  17.3  ...  2.2  9.5  1.3  1.5  3.2           27.7  24157304.0  NBA_2019   \n",
       "1  14.9  ...  0.6  3.4  1.0  0.1  2.2           18.3  13000000.0  NBA_2019   \n",
       "2  12.4  ...  1.1  3.0  1.5  0.4  2.0           15.9  15000000.0  NBA_2019   \n",
       "3   9.7  ...  0.4  1.6  0.6  2.2  2.3           12.5   3382000.0  NBA_2019   \n",
       "4  11.7  ...  1.0  2.4  0.7  0.2  1.6           15.6   1544951.0  NBA_2019   \n",
       "\n",
       "  Salary_Log  Salary_Log_predict  \n",
       "0  17.000097           17.290277  \n",
       "1  16.380460           16.212891  \n",
       "2  16.523561           15.897635  \n",
       "3  15.033978           15.971340  \n",
       "4  14.250503           15.576392  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "onlyPlayerStats.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we have our predicted Salary in Logarithm but we need to do some inverse log transformation to get our specific players salary. We create a function for that below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "This function returns a dataframe containing a particular user, their stats, \n",
    "log of their actual and predicted salaries, and their actual and predicted salaries\n",
    "'''\n",
    "def getDataStatsSalary(onlyPlayerStats,usefulColumns,playerName):\n",
    "    sortByPlayerStats = onlyPlayerStats.sort_values(by=['Name','Season'])\n",
    "\n",
    "    ### Get Players, and Important Columns\n",
    "    usefulColumns = [\"Name\"] + usefulColumns[0:5] + [\"Season\",\"Salary\"] + [usefulColumns[5]] + [\"Salary_Log_predict\"]\n",
    "\n",
    "    importedColumnSortedPlayers = sortByPlayerStats[usefulColumns]\n",
    "\n",
    "    GiannisHistory = importedColumnSortedPlayers [importedColumnSortedPlayers[\"Name\"] == playerName]\n",
    "\n",
    "    arrayIndex = []\n",
    "    for res in GiannisHistory[\"Salary_Log_predict\"]:\n",
    "        arrayIndex.append(math.exp(res))\n",
    "    \n",
    "    GiannisHistory[\"Salary_predicted\"] = arrayIndex\n",
    "    \n",
    "    CompareData = GiannisHistory.drop(columns=[\"Salary_Log_predict\", \"Salary_Log\"])\n",
    "    \n",
    "    return CompareData"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 5: Evaluate Results and Draw Insights"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Insights 1:\n",
    "Victor Oladipo who won the most improved player in 2018 so an increase in his actual salary from $6.5 M to $ 21 M but my model he predicted he needs to be making less based on his stats. In particular, Victor's stats are lower in 2018 than 2019 because of an out of season injury he had. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:19: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Age</th>\n",
       "      <th>MP</th>\n",
       "      <th>DRB</th>\n",
       "      <th>PF</th>\n",
       "      <th>PointsPerGame</th>\n",
       "      <th>Season</th>\n",
       "      <th>Salary</th>\n",
       "      <th>Salary_predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1260</th>\n",
       "      <td>Victor Oladipo</td>\n",
       "      <td>24</td>\n",
       "      <td>33.20</td>\n",
       "      <td>3.80</td>\n",
       "      <td>2.30</td>\n",
       "      <td>15.90</td>\n",
       "      <td>NBA_2017</td>\n",
       "      <td>6552960.00</td>\n",
       "      <td>6950453.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>555</th>\n",
       "      <td>Victor Oladipo</td>\n",
       "      <td>25</td>\n",
       "      <td>34.00</td>\n",
       "      <td>4.60</td>\n",
       "      <td>2.30</td>\n",
       "      <td>23.10</td>\n",
       "      <td>NBA_2018</td>\n",
       "      <td>21000000.00</td>\n",
       "      <td>12024670.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64</th>\n",
       "      <td>Victor Oladipo</td>\n",
       "      <td>26</td>\n",
       "      <td>31.90</td>\n",
       "      <td>5.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>18.80</td>\n",
       "      <td>NBA_2019</td>\n",
       "      <td>21000000.00</td>\n",
       "      <td>10163521.21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Name  Age    MP  DRB   PF  PointsPerGame    Season  \\\n",
       "1260  Victor Oladipo   24 33.20 3.80 2.30          15.90  NBA_2017   \n",
       "555   Victor Oladipo   25 34.00 4.60 2.30          23.10  NBA_2018   \n",
       "64    Victor Oladipo   26 31.90 5.00 2.00          18.80  NBA_2019   \n",
       "\n",
       "          Salary  Salary_predicted  \n",
       "1260  6552960.00        6950453.18  \n",
       "555  21000000.00       12024670.29  \n",
       "64   21000000.00       10163521.21  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Check for the most improved Player in 2017 and evaluate his growth\n",
    "playerComparedStatsVictor = getDataStatsSalary(onlyPlayerStats,usefulColumns,\"Victor Oladipo\")\n",
    "pd.options.display.float_format = '{:.2f}'.format\n",
    "playerComparedStatsVictor "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Insights 2:\n",
    "Giannis Antetokounmpo who won the most improved player in 2017 and plays for the Bucks saw an increase in his actual salary from $2.9 M to $ 24 M but my model predicted that he needs to be making a lot more money because of the impact he making on his team.\n",
    "\n",
    "Giannis personally cares more about building the franchise of the Bucks than making the most money from his team but we can use this model to see the value he brings to the Bucks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:19: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Age</th>\n",
       "      <th>MP</th>\n",
       "      <th>DRB</th>\n",
       "      <th>PF</th>\n",
       "      <th>PointsPerGame</th>\n",
       "      <th>Season</th>\n",
       "      <th>Salary</th>\n",
       "      <th>Salary_predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1034</th>\n",
       "      <td>Giannis Antetokounmpo</td>\n",
       "      <td>22</td>\n",
       "      <td>35.60</td>\n",
       "      <td>7.00</td>\n",
       "      <td>3.10</td>\n",
       "      <td>22.90</td>\n",
       "      <td>NBA_2017</td>\n",
       "      <td>2995421.00</td>\n",
       "      <td>15497674.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>585</th>\n",
       "      <td>Giannis Antetokounmpo</td>\n",
       "      <td>23</td>\n",
       "      <td>36.70</td>\n",
       "      <td>8.00</td>\n",
       "      <td>3.10</td>\n",
       "      <td>26.90</td>\n",
       "      <td>NBA_2018</td>\n",
       "      <td>22471910.00</td>\n",
       "      <td>24488340.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Giannis Antetokounmpo</td>\n",
       "      <td>24</td>\n",
       "      <td>32.80</td>\n",
       "      <td>10.30</td>\n",
       "      <td>3.20</td>\n",
       "      <td>27.70</td>\n",
       "      <td>NBA_2019</td>\n",
       "      <td>24157304.00</td>\n",
       "      <td>32290270.87</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Name  Age    MP   DRB   PF  PointsPerGame    Season  \\\n",
       "1034  Giannis Antetokounmpo   22 35.60  7.00 3.10          22.90  NBA_2017   \n",
       "585   Giannis Antetokounmpo   23 36.70  8.00 3.10          26.90  NBA_2018   \n",
       "0     Giannis Antetokounmpo   24 32.80 10.30 3.20          27.70  NBA_2019   \n",
       "\n",
       "          Salary  Salary_predicted  \n",
       "1034  2995421.00       15497674.11  \n",
       "585  22471910.00       24488340.31  \n",
       "0    24157304.00       32290270.87  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Player Giannis\n",
    "playerComparedStatsGiannis = getDataStatsSalary(onlyPlayerStats,usefulColumns,\"Giannis Antetokounmpo\")\n",
    "pd.options.display.float_format = '{:.2f}'.format\n",
    "playerComparedStatsGiannis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Insights 3:\n",
    "Jimmy Butler who won the most improved player in 2015 and saw an increase in his actual salary from $17 M to $ 20 M but my model predicted that his salary needs to be decreasing over the different seasons because of his stats.\n",
    "\n",
    "Something to note here is that there are other factors that affect a players value such as social brand and popularity that are not measured in the statistics.\n",
    "\n",
    "So some teams are willing to pay a premium for some players because of the viewership they would bring to the team.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/opasina/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:19: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Age</th>\n",
       "      <th>MP</th>\n",
       "      <th>DRB</th>\n",
       "      <th>PF</th>\n",
       "      <th>PointsPerGame</th>\n",
       "      <th>Season</th>\n",
       "      <th>Salary</th>\n",
       "      <th>Salary_predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1067</th>\n",
       "      <td>Jimmy Butler</td>\n",
       "      <td>27</td>\n",
       "      <td>37.00</td>\n",
       "      <td>4.50</td>\n",
       "      <td>1.50</td>\n",
       "      <td>23.90</td>\n",
       "      <td>NBA_2017</td>\n",
       "      <td>17552209.00</td>\n",
       "      <td>15786995.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>829</th>\n",
       "      <td>Jimmy Butler</td>\n",
       "      <td>28</td>\n",
       "      <td>36.70</td>\n",
       "      <td>4.00</td>\n",
       "      <td>1.30</td>\n",
       "      <td>22.20</td>\n",
       "      <td>NBA_2018</td>\n",
       "      <td>19301070.00</td>\n",
       "      <td>14318452.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>Jimmy Butler</td>\n",
       "      <td>29</td>\n",
       "      <td>33.20</td>\n",
       "      <td>3.40</td>\n",
       "      <td>1.70</td>\n",
       "      <td>18.20</td>\n",
       "      <td>NBA_2019</td>\n",
       "      <td>20445779.00</td>\n",
       "      <td>10311500.10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Name  Age    MP  DRB   PF  PointsPerGame    Season      Salary  \\\n",
       "1067  Jimmy Butler   27 37.00 4.50 1.50          23.90  NBA_2017 17552209.00   \n",
       "829   Jimmy Butler   28 36.70 4.00 1.30          22.20  NBA_2018 19301070.00   \n",
       "36    Jimmy Butler   29 33.20 3.40 1.70          18.20  NBA_2019 20445779.00   \n",
       "\n",
       "      Salary_predicted  \n",
       "1067       15786995.85  \n",
       "829        14318452.25  \n",
       "36         10311500.10  "
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Player Jimmy\n",
    "playerComparedStatsJimmy = getDataStatsSalary(onlyPlayerStats,usefulColumns,\"Jimmy Butler\")\n",
    "playerComparedStatsJimmy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion:\n",
    "In conclusion, I was able to predict Nba salaries for players based on their stats. The model actually provides a lot more insights on the impact a player makes on the team games and can be used to either extend the players contract or trade the player."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}