scores_1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sn\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "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>Unnamed: 0</th>\n",
       "      <th>gender</th>\n",
       "      <th>race</th>\n",
       "      <th>education</th>\n",
       "      <th>lunch</th>\n",
       "      <th>test preparation course</th>\n",
       "      <th>math</th>\n",
       "      <th>reading</th>\n",
       "      <th>writing</th>\n",
       "      <th>Average</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>female</td>\n",
       "      <td>group B</td>\n",
       "      <td>bachelor's degree</td>\n",
       "      <td>standard</td>\n",
       "      <td>none</td>\n",
       "      <td>72</td>\n",
       "      <td>72</td>\n",
       "      <td>74</td>\n",
       "      <td>72.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>group C</td>\n",
       "      <td>some college</td>\n",
       "      <td>standard</td>\n",
       "      <td>completed</td>\n",
       "      <td>69</td>\n",
       "      <td>90</td>\n",
       "      <td>88</td>\n",
       "      <td>82.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>female</td>\n",
       "      <td>group B</td>\n",
       "      <td>master's degree</td>\n",
       "      <td>standard</td>\n",
       "      <td>none</td>\n",
       "      <td>90</td>\n",
       "      <td>95</td>\n",
       "      <td>93</td>\n",
       "      <td>92.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>group A</td>\n",
       "      <td>associate's degree</td>\n",
       "      <td>free/reduced</td>\n",
       "      <td>none</td>\n",
       "      <td>47</td>\n",
       "      <td>57</td>\n",
       "      <td>44</td>\n",
       "      <td>49.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>male</td>\n",
       "      <td>group C</td>\n",
       "      <td>some college</td>\n",
       "      <td>standard</td>\n",
       "      <td>none</td>\n",
       "      <td>76</td>\n",
       "      <td>78</td>\n",
       "      <td>75</td>\n",
       "      <td>76.333333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0  gender     race           education         lunch  \\\n",
       "0           0  female  group B   bachelor's degree      standard   \n",
       "1           1  female  group C        some college      standard   \n",
       "2           2  female  group B     master's degree      standard   \n",
       "3           3    male  group A  associate's degree  free/reduced   \n",
       "4           4    male  group C        some college      standard   \n",
       "\n",
       "  test preparation course  math  reading  writing    Average  \n",
       "0                    none    72       72       74  72.666667  \n",
       "1               completed    69       90       88  82.333333  \n",
       "2                    none    90       95       93  92.666667  \n",
       "3                    none    47       57       44  49.333333  \n",
       "4                    none    76       78       75  76.333333  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"new_degree.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "## I will try to create a model aiming to predict the score of the math course based on students features\n",
    "df.drop(columns = [\"Average\", \"Unnamed: 0\"], axis=1, inplace=True)\n",
    "df.head()\n",
    "df_dum = pd.get_dummies(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>math</th>\n",
       "      <th>reading</th>\n",
       "      <th>writing</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Race</th>\n",
       "      <th>Education</th>\n",
       "      <th>Preperation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>72</td>\n",
       "      <td>72</td>\n",
       "      <td>74</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>69</td>\n",
       "      <td>90</td>\n",
       "      <td>88</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>90</td>\n",
       "      <td>95</td>\n",
       "      <td>93</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>47</td>\n",
       "      <td>57</td>\n",
       "      <td>44</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>76</td>\n",
       "      <td>78</td>\n",
       "      <td>75</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   math  reading  writing  Gender  Race  Education  Preperation\n",
       "0    72       72       74       0     1          1            1\n",
       "1    69       90       88       0     2          4            0\n",
       "2    90       95       93       0     1          3            1\n",
       "3    47       57       44       1     0          0            1\n",
       "4    76       78       75       1     2          4            1"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Get dummy data\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "lb_make = LabelEncoder()\n",
    "df[\"Gender\"] = lb_make.fit_transform(df[\"gender\"])\n",
    "df[\"Race\"] = lb_make.fit_transform(df[\"race\"])\n",
    "df[\"Education\"] = lb_make.fit_transform(df[\"education\"])\n",
    "df[\"Preperation\"] = lb_make.fit_transform(df[\"test preparation course\"])\n",
    "\n",
    "\n",
    "df = df[[\"math\",\"reading\", \"writing\", \"Gender\", \"Race\", \"Education\", \"Preperation\"]]\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Examine the correlation of our data features, with a Correlation Matrix \n",
    "plt.figure(figsize = (10,10))\n",
    "corrMatrix = df.corr()\n",
    "sn.heatmap(corrMatrix, annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Preparing the data before the splitting \n",
    "X = df_dum.drop(\"math\", axis=1)\n",
    "y = df.math.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Spltting the data\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>            <td>y</td>        <th>  R-squared:         </th> <td>   0.877</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.875</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   500.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 16 Aug 2020</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>00:53:33</td>     <th>  Log-Likelihood:    </th> <td> -3090.7</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>  1000</td>      <th>  AIC:               </th> <td>   6211.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>   985</td>      <th>  BIC:               </th> <td>   6285.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    14</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                  <td></td>                     <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th>                             <td>   -0.2160</td> <td>    0.338</td> <td>   -0.638</td> <td> 0.523</td> <td>   -0.880</td> <td>    0.448</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>reading</th>                           <td>    0.2635</td> <td>    0.042</td> <td>    6.266</td> <td> 0.000</td> <td>    0.181</td> <td>    0.346</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>writing</th>                           <td>    0.7016</td> <td>    0.044</td> <td>   16.120</td> <td> 0.000</td> <td>    0.616</td> <td>    0.787</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>gender_female</th>                     <td>   -6.7282</td> <td>    0.287</td> <td>  -23.472</td> <td> 0.000</td> <td>   -7.291</td> <td>   -6.166</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>gender_male</th>                       <td>    6.5122</td> <td>    0.210</td> <td>   30.950</td> <td> 0.000</td> <td>    6.099</td> <td>    6.925</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race_group A</th>                      <td>   -1.2811</td> <td>    0.487</td> <td>   -2.628</td> <td> 0.009</td> <td>   -2.238</td> <td>   -0.325</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race_group B</th>                      <td>   -0.4458</td> <td>    0.359</td> <td>   -1.240</td> <td> 0.215</td> <td>   -1.151</td> <td>    0.260</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race_group C</th>                      <td>   -1.1029</td> <td>    0.302</td> <td>   -3.648</td> <td> 0.000</td> <td>   -1.696</td> <td>   -0.510</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race_group D</th>                      <td>   -1.1827</td> <td>    0.339</td> <td>   -3.487</td> <td> 0.001</td> <td>   -1.848</td> <td>   -0.517</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>race_group E</th>                      <td>    3.7966</td> <td>    0.415</td> <td>    9.143</td> <td> 0.000</td> <td>    2.982</td> <td>    4.611</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>education_associate's degree</th>      <td>    0.1945</td> <td>    0.355</td> <td>    0.548</td> <td> 0.583</td> <td>   -0.501</td> <td>    0.890</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>education_bachelor's degree</th>       <td>   -0.8524</td> <td>    0.462</td> <td>   -1.845</td> <td> 0.065</td> <td>   -1.759</td> <td>    0.054</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>education_high school</th>             <td>    0.7622</td> <td>    0.370</td> <td>    2.059</td> <td> 0.040</td> <td>    0.036</td> <td>    1.488</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>education_master's degree</th>         <td>   -1.6616</td> <td>    0.624</td> <td>   -2.664</td> <td> 0.008</td> <td>   -2.886</td> <td>   -0.438</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>education_some college</th>            <td>    0.5947</td> <td>    0.349</td> <td>    1.702</td> <td> 0.089</td> <td>   -0.091</td> <td>    1.280</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>education_some high school</th>        <td>    0.7466</td> <td>    0.384</td> <td>    1.943</td> <td> 0.052</td> <td>   -0.007</td> <td>    1.501</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>lunch_free/reduced</th>                <td>   -1.7144</td> <td>    0.221</td> <td>   -7.770</td> <td> 0.000</td> <td>   -2.147</td> <td>   -1.281</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>lunch_standard</th>                    <td>    1.4983</td> <td>    0.280</td> <td>    5.344</td> <td> 0.000</td> <td>    0.948</td> <td>    2.049</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>test preparation course_completed</th> <td>   -1.8592</td> <td>    0.303</td> <td>   -6.133</td> <td> 0.000</td> <td>   -2.454</td> <td>   -1.264</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>test preparation course_none</th>      <td>    1.6431</td> <td>    0.210</td> <td>    7.831</td> <td> 0.000</td> <td>    1.231</td> <td>    2.055</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 0.330</td> <th>  Durbin-Watson:     </th> <td>   1.986</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.848</td> <th>  Jarque-Bera (JB):  </th> <td>   0.402</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.034</td> <th>  Prob(JB):          </th> <td>   0.818</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.930</td> <th>  Cond. No.          </th> <td>1.23e+18</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The smallest eigenvalue is 6.55e-30. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   R-squared:                       0.877\n",
       "Model:                            OLS   Adj. R-squared:                  0.875\n",
       "Method:                 Least Squares   F-statistic:                     500.3\n",
       "Date:                Sun, 16 Aug 2020   Prob (F-statistic):               0.00\n",
       "Time:                        00:53:33   Log-Likelihood:                -3090.7\n",
       "No. Observations:                1000   AIC:                             6211.\n",
       "Df Residuals:                     985   BIC:                             6285.\n",
       "Df Model:                          14                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=====================================================================================================\n",
       "                                        coef    std err          t      P>|t|      [0.025      0.975]\n",
       "-----------------------------------------------------------------------------------------------------\n",
       "const                                -0.2160      0.338     -0.638      0.523      -0.880       0.448\n",
       "reading                               0.2635      0.042      6.266      0.000       0.181       0.346\n",
       "writing                               0.7016      0.044     16.120      0.000       0.616       0.787\n",
       "gender_female                        -6.7282      0.287    -23.472      0.000      -7.291      -6.166\n",
       "gender_male                           6.5122      0.210     30.950      0.000       6.099       6.925\n",
       "race_group A                         -1.2811      0.487     -2.628      0.009      -2.238      -0.325\n",
       "race_group B                         -0.4458      0.359     -1.240      0.215      -1.151       0.260\n",
       "race_group C                         -1.1029      0.302     -3.648      0.000      -1.696      -0.510\n",
       "race_group D                         -1.1827      0.339     -3.487      0.001      -1.848      -0.517\n",
       "race_group E                          3.7966      0.415      9.143      0.000       2.982       4.611\n",
       "education_associate's degree          0.1945      0.355      0.548      0.583      -0.501       0.890\n",
       "education_bachelor's degree          -0.8524      0.462     -1.845      0.065      -1.759       0.054\n",
       "education_high school                 0.7622      0.370      2.059      0.040       0.036       1.488\n",
       "education_master's degree            -1.6616      0.624     -2.664      0.008      -2.886      -0.438\n",
       "education_some college                0.5947      0.349      1.702      0.089      -0.091       1.280\n",
       "education_some high school            0.7466      0.384      1.943      0.052      -0.007       1.501\n",
       "lunch_free/reduced                   -1.7144      0.221     -7.770      0.000      -2.147      -1.281\n",
       "lunch_standard                        1.4983      0.280      5.344      0.000       0.948       2.049\n",
       "test preparation course_completed    -1.8592      0.303     -6.133      0.000      -2.454      -1.264\n",
       "test preparation course_none          1.6431      0.210      7.831      0.000       1.231       2.055\n",
       "==============================================================================\n",
       "Omnibus:                        0.330   Durbin-Watson:                   1.986\n",
       "Prob(Omnibus):                  0.848   Jarque-Bera (JB):                0.402\n",
       "Skew:                          -0.034   Prob(JB):                        0.818\n",
       "Kurtosis:                       2.930   Cond. No.                     1.23e+18\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The smallest eigenvalue is 6.55e-30. This might indicate that there are\n",
       "strong multicollinearity problems or that the design matrix is singular.\n",
       "\"\"\""
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Exporte a summary and examine wich feature affected most the target value  \n",
    "import statsmodels.api as sm\n",
    "\n",
    "X_cm = X = sm.add_constant(X)\n",
    "model = sm.OLS(y,X_cm)\n",
    "model.fit().summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##Initially, I'm traing a Multiple Linear Regression model\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "model = LinearRegression()\n",
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8804332983749567\n"
     ]
    }
   ],
   "source": [
    "accuracy = model.score(X_test, y_test)\n",
    "print(accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "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>Actual</th>\n",
       "      <th>Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>91</td>\n",
       "      <td>76.387970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>53</td>\n",
       "      <td>58.885970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>80</td>\n",
       "      <td>76.990265</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>74</td>\n",
       "      <td>76.851804</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>84</td>\n",
       "      <td>87.627378</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Actual  Predicted\n",
       "0      91  76.387970\n",
       "1      53  58.885970\n",
       "2      80  76.990265\n",
       "3      74  76.851804\n",
       "4      84  87.627378"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Prediction\n",
    "y_pred = model.predict(X_test)\n",
    "\n",
    "## Compare the predicted scores with the actuals scores\n",
    "df1 = pd.DataFrame({'Actual': y_test.flatten(), 'Predicted': y_pred.flatten()})\n",
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df2 = df1.head(20)\n",
    "df2.plot(kind='bar',figsize=(16,6))\n",
    "plt.grid(which=\"major\", linestyle='-', linewidth='0.5', color='green')\n",
    "plt.grid(which='minor', linestyle=':', linewidth='0.5', color='black')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['explained_variance', 'r2', 'max_error', 'neg_median_absolute_error', 'neg_mean_absolute_error', 'neg_mean_squared_error', 'neg_mean_squared_log_error', 'neg_root_mean_squared_error', 'neg_mean_poisson_deviance', 'neg_mean_gamma_deviance', 'accuracy', 'roc_auc', 'roc_auc_ovr', 'roc_auc_ovo', 'roc_auc_ovr_weighted', 'roc_auc_ovo_weighted', 'balanced_accuracy', 'average_precision', 'neg_log_loss', 'neg_brier_score', 'adjusted_rand_score', 'homogeneity_score', 'completeness_score', 'v_measure_score', 'mutual_info_score', 'adjusted_mutual_info_score', 'normalized_mutual_info_score', 'fowlkes_mallows_score', 'precision', 'precision_macro', 'precision_micro', 'precision_samples', 'precision_weighted', 'recall', 'recall_macro', 'recall_micro', 'recall_samples', 'recall_weighted', 'f1', 'f1_macro', 'f1_micro', 'f1_samples', 'f1_weighted', 'jaccard', 'jaccard_macro', 'jaccard_micro', 'jaccard_samples', 'jaccard_weighted'])"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import sklearn\n",
    "sklearn.metrics.SCORERS.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-4.38534844836657"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Evaluate the models using negative mean absolute error\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "scores = cross_val_score(model, X_train, y_train, scoring=\"neg_mean_absolute_error\", cv=3)\n",
    "np.mean(scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, ccp_alpha=0.0, criterion='mse',\n",
       "                      max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "                      max_samples=None, min_impurity_decrease=0.0,\n",
       "                      min_impurity_split=None, min_samples_leaf=1,\n",
       "                      min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "                      n_estimators=100, n_jobs=None, oob_score=False,\n",
       "                      random_state=1, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## The next model that I will train is a Random Forest algorithm\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "forest_model = RandomForestRegressor(random_state=1)\n",
    "forest_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8496466221268407\n"
     ]
    }
   ],
   "source": [
    "## Evaluation \n",
    "accuracy = forest_model.score(X_test, y_test)\n",
    "print(accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "y2_pred = forest_model.predict(X_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-4.969210878319394"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Negative mean absolute error\n",
    "scores_2 = cross_val_score(forest_model, X_train, y_train, scoring=\"neg_mean_absolute_error\", cv=3)\n",
    "np.mean(scores_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-5.155198036880054"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Lasso regression\n",
    "from sklearn.linear_model import Lasso\n",
    "\n",
    "las_model = Lasso()\n",
    "\n",
    "scores_3 = cross_val_score(las_model, X_train, y_train, scoring=\"neg_mean_absolute_error\", cv=3)\n",
    "np.mean(scores_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "## I will try to reduce the error finding a better alpha\n",
    "\n",
    "alpha = []\n",
    "error = []\n",
    "\n",
    "for i in range (1,100):\n",
    "    alpha.append(i/100)\n",
    "    la = Lasso(alpha=(i/100))\n",
    "    error.append(np.mean(cross_val_score(la, X_train, y_train, scoring=\"neg_mean_absolute_error\", cv=3)))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x2918e2ea708>]"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(alpha, error)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "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>Aplha</th>\n",
       "      <th>Error</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.02</td>\n",
       "      <td>-4.38192</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Aplha    Error\n",
       "1   0.02 -4.38192"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "err = tuple(zip(alpha, error))\n",
    "df_err = pd.DataFrame(err, columns = [\"Aplha\", \"Error\"])\n",
    "df_err[df_err.Error == max(df_err.Error)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Lasso(alpha=0.02, copy_X=True, fit_intercept=True, max_iter=1000,\n",
       "      normalize=False, positive=False, precompute=False, random_state=None,\n",
       "      selection='cyclic', tol=0.0001, warm_start=False)"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lasso_impr = Lasso(alpha=0.02)\n",
    "lasso_impr.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Tune models GridsearchCV, I will try to improve the Random Forest model, finding the best parameters.\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "parameters = {\"n_estimators\":range(10,200,10), \"criterion\":(\"mse\", \"mae\"), \"max_features\":(\"auto\", \"sqrt\", \"log2\")}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=3, error_score=nan,\n",
       "             estimator=RandomForestRegressor(bootstrap=True, ccp_alpha=0.0,\n",
       "                                             criterion='mse', max_depth=None,\n",
       "                                             max_features='auto',\n",
       "                                             max_leaf_nodes=None,\n",
       "                                             max_samples=None,\n",
       "                                             min_impurity_decrease=0.0,\n",
       "                                             min_impurity_split=None,\n",
       "                                             min_samples_leaf=1,\n",
       "                                             min_samples_split=2,\n",
       "                                             min_weight_fraction_leaf=0.0,\n",
       "                                             n_estimators=100, n_jobs=None,\n",
       "                                             oob_score=False, random_state=1,\n",
       "                                             verbose=0, warm_start=False),\n",
       "             iid='deprecated', n_jobs=None,\n",
       "             param_grid={'criterion': ('mse', 'mae'),\n",
       "                         'max_features': ('auto', 'sqrt', 'log2'),\n",
       "                         'n_estimators': range(10, 200, 10)},\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring='neg_mean_absolute_error', verbose=0)"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gs = GridSearchCV(forest_model, parameters, scoring=\"neg_mean_absolute_error\", cv=3)\n",
    "gs.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-4.899720400829213"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gs.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, ccp_alpha=0.0, criterion='mae',\n",
       "                      max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "                      max_samples=None, min_impurity_decrease=0.0,\n",
       "                      min_impurity_split=None, min_samples_leaf=1,\n",
       "                      min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "                      n_estimators=130, n_jobs=None, oob_score=False,\n",
       "                      random_state=1, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gs.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Test ansembles\n",
    "tpred_linear = model.predict(X_test)\n",
    "tpred_lasso = lasso_impr.predict(X_test)\n",
    "tpred_rand_for = gs.best_estimator_.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4.214763142474848\n",
      "4.196016677309714\n",
      "4.600826923076923\n"
     ]
    }
   ],
   "source": [
    "## Examine the mean absolute error and pick the model which has the best results.\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "print(mean_absolute_error(y_test, tpred_linear))\n",
    "print(mean_absolute_error(y_test, tpred_lasso)) \n",
    "print(mean_absolute_error(y_test, tpred_rand_for))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Save the model\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"degrees_pred.pkl\", \"wb\") as f:\n",
    "     pickle.dump(lasso_impr,f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Load the model\n",
    "with open(\"degrees_pred.pkl\", \"rb\") as f:\n",
    "    loaded_lasso = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Lets make a prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_student = [70,75,0,1,1,0,0,0,0,0,1,0,0,0,0,1,0,0,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([75.56089091])"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loaded_lasso.predict([new_student])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}