linear regression IMDb score_current.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import warnings\n",
    "from sklearn import metrics\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.preprocessing import StandardScaler, MinMaxScaler\n",
    "from sklearn.metrics import accuracy_score, recall_score, precision_score, roc_auc_score \n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.model_selection import train_test_split \n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.model_selection import cross_validate\n",
    "from sklearn import linear_model\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.linear_model import Lasso, ElasticNet\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.preprocessing import PolynomialFeatures"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "movie_df = pd.read_csv(r'data/data_regression.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3852, 42)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "imdb_id                       object\n",
       "director_name                 object\n",
       "num_critic_for_reviews       float64\n",
       "duration                     float64\n",
       "director_facebook_likes      float64\n",
       "actor_3_facebook_likes       float64\n",
       "actor_2_name                  object\n",
       "actor_1_facebook_likes       float64\n",
       "gross                        float64\n",
       "genres                        object\n",
       "actor_1_name                  object\n",
       "movie_title                   object\n",
       "num_voted_users                int64\n",
       "cast_total_facebook_likes      int64\n",
       "actor_3_name                  object\n",
       "facenumber_in_poster         float64\n",
       "plot_keywords                 object\n",
       "num_user_for_reviews         float64\n",
       "language                      object\n",
       "country                       object\n",
       "content_rating                object\n",
       "budget                       float64\n",
       "title_year                   float64\n",
       "actor_2_facebook_likes       float64\n",
       "imdb_score                   float64\n",
       "movie_facebook_likes           int64\n",
       "Biography                      int64\n",
       "Comedy                         int64\n",
       "Crime                          int64\n",
       "Drama                          int64\n",
       "Romance                        int64\n",
       "Mystery_Thriller_Horror        int64\n",
       "Sci-Fi_Fantasy                 int64\n",
       "Family_Animation               int64\n",
       "Action_Adventure               int64\n",
       "History_War                    int64\n",
       "Others                         int64\n",
       "director_rank                  int64\n",
       "actor1_rank                    int64\n",
       "actor2_rank                    int64\n",
       "actor3_rank                    int64\n",
       "gross_budget_ratio           float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(movie_df.shape)\n",
    "movie_df.dtypes"
   ]
  },
  {
   "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>imdb_id</th>\n",
       "      <th>director_name</th>\n",
       "      <th>num_critic_for_reviews</th>\n",
       "      <th>duration</th>\n",
       "      <th>director_facebook_likes</th>\n",
       "      <th>actor_3_facebook_likes</th>\n",
       "      <th>actor_2_name</th>\n",
       "      <th>actor_1_facebook_likes</th>\n",
       "      <th>gross</th>\n",
       "      <th>genres</th>\n",
       "      <th>...</th>\n",
       "      <th>Sci-Fi_Fantasy</th>\n",
       "      <th>Family_Animation</th>\n",
       "      <th>Action_Adventure</th>\n",
       "      <th>History_War</th>\n",
       "      <th>Others</th>\n",
       "      <th>director_rank</th>\n",
       "      <th>actor1_rank</th>\n",
       "      <th>actor2_rank</th>\n",
       "      <th>actor3_rank</th>\n",
       "      <th>gross_budget_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>tt0035423</td>\n",
       "      <td>James Mangold</td>\n",
       "      <td>125.0</td>\n",
       "      <td>123.0</td>\n",
       "      <td>446.0</td>\n",
       "      <td>821.0</td>\n",
       "      <td>Natasha Lyonne</td>\n",
       "      <td>20000.0</td>\n",
       "      <td>47095453.0</td>\n",
       "      <td>Comedy|Fantasy|Romance</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0.981155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>tt0080339</td>\n",
       "      <td>Jim Abrahams</td>\n",
       "      <td>134.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>104.0</td>\n",
       "      <td>318.0</td>\n",
       "      <td>Lloyd Bridges</td>\n",
       "      <td>628.0</td>\n",
       "      <td>83400000.0</td>\n",
       "      <td>Comedy</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>23.828571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>tt0080453</td>\n",
       "      <td>Randal Kleiser</td>\n",
       "      <td>36.0</td>\n",
       "      <td>104.0</td>\n",
       "      <td>116.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>Christopher Atkins</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>58853106.0</td>\n",
       "      <td>Adventure|Drama|Romance</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>13.078468</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>tt0080455</td>\n",
       "      <td>John Landis</td>\n",
       "      <td>125.0</td>\n",
       "      <td>148.0</td>\n",
       "      <td>644.0</td>\n",
       "      <td>326.0</td>\n",
       "      <td>Aretha Franklin</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>54200000.0</td>\n",
       "      <td>Action|Comedy|Crime|Music</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>2.007407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>tt0080487</td>\n",
       "      <td>Harold Ramis</td>\n",
       "      <td>71.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>11000.0</td>\n",
       "      <td>484.0</td>\n",
       "      <td>Rodney Dangerfield</td>\n",
       "      <td>13000.0</td>\n",
       "      <td>39800000.0</td>\n",
       "      <td>Comedy|Sport</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>6.633333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 42 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     imdb_id   director_name  num_critic_for_reviews  duration  \\\n",
       "0  tt0035423   James Mangold                   125.0     123.0   \n",
       "1  tt0080339    Jim Abrahams                   134.0      88.0   \n",
       "2  tt0080453  Randal Kleiser                    36.0     104.0   \n",
       "3  tt0080455     John Landis                   125.0     148.0   \n",
       "4  tt0080487    Harold Ramis                    71.0      98.0   \n",
       "\n",
       "   director_facebook_likes  actor_3_facebook_likes        actor_2_name  \\\n",
       "0                    446.0                   821.0      Natasha Lyonne   \n",
       "1                    104.0                   318.0       Lloyd Bridges   \n",
       "2                    116.0                    83.0  Christopher Atkins   \n",
       "3                    644.0                   326.0     Aretha Franklin   \n",
       "4                  11000.0                   484.0  Rodney Dangerfield   \n",
       "\n",
       "   actor_1_facebook_likes       gross                     genres  ...  \\\n",
       "0                 20000.0  47095453.0     Comedy|Fantasy|Romance  ...   \n",
       "1                   628.0  83400000.0                     Comedy  ...   \n",
       "2                  1000.0  58853106.0    Adventure|Drama|Romance  ...   \n",
       "3                  1000.0  54200000.0  Action|Comedy|Crime|Music  ...   \n",
       "4                 13000.0  39800000.0               Comedy|Sport  ...   \n",
       "\n",
       "  Sci-Fi_Fantasy Family_Animation  Action_Adventure  History_War Others  \\\n",
       "0              1                0                 0            0      0   \n",
       "1              0                0                 0            0      0   \n",
       "2              0                0                 1            0      0   \n",
       "3              0                0                 1            0      1   \n",
       "4              0                0                 0            0      1   \n",
       "\n",
       "   director_rank actor1_rank  actor2_rank actor3_rank gross_budget_ratio  \n",
       "0              8           6            3           4           0.981155  \n",
       "1              8           2            3           4          23.828571  \n",
       "2              7           3            3           3          13.078468  \n",
       "3              8           3            3           4           2.007407  \n",
       "4              7           7            3           5           6.633333  \n",
       "\n",
       "[5 rows x 42 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "movie_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "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>num_critic_for_reviews</th>\n",
       "      <th>duration</th>\n",
       "      <th>director_facebook_likes</th>\n",
       "      <th>actor_3_facebook_likes</th>\n",
       "      <th>actor_1_facebook_likes</th>\n",
       "      <th>gross</th>\n",
       "      <th>num_voted_users</th>\n",
       "      <th>cast_total_facebook_likes</th>\n",
       "      <th>facenumber_in_poster</th>\n",
       "      <th>num_user_for_reviews</th>\n",
       "      <th>...</th>\n",
       "      <th>Sci-Fi_Fantasy</th>\n",
       "      <th>Family_Animation</th>\n",
       "      <th>Action_Adventure</th>\n",
       "      <th>History_War</th>\n",
       "      <th>Others</th>\n",
       "      <th>director_rank</th>\n",
       "      <th>actor1_rank</th>\n",
       "      <th>actor2_rank</th>\n",
       "      <th>actor3_rank</th>\n",
       "      <th>gross_budget_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>count</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3.852000e+03</td>\n",
       "      <td>3.852000e+03</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>3852.00000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "      <td>3852.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>mean</td>\n",
       "      <td>160.910436</td>\n",
       "      <td>109.049325</td>\n",
       "      <td>762.463396</td>\n",
       "      <td>730.276739</td>\n",
       "      <td>7563.897715</td>\n",
       "      <td>4.929950e+07</td>\n",
       "      <td>9.813550e+04</td>\n",
       "      <td>11170.345275</td>\n",
       "      <td>1.379283</td>\n",
       "      <td>313.869938</td>\n",
       "      <td>...</td>\n",
       "      <td>0.23702</td>\n",
       "      <td>0.119678</td>\n",
       "      <td>0.336708</td>\n",
       "      <td>0.063084</td>\n",
       "      <td>0.132918</td>\n",
       "      <td>7.076064</td>\n",
       "      <td>4.046729</td>\n",
       "      <td>3.432503</td>\n",
       "      <td>4.333333</td>\n",
       "      <td>5.690951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>std</td>\n",
       "      <td>121.840785</td>\n",
       "      <td>21.549751</td>\n",
       "      <td>2988.626715</td>\n",
       "      <td>1805.214756</td>\n",
       "      <td>15431.666830</td>\n",
       "      <td>7.034357e+07</td>\n",
       "      <td>1.468241e+05</td>\n",
       "      <td>18845.309090</td>\n",
       "      <td>2.045882</td>\n",
       "      <td>397.309921</td>\n",
       "      <td>...</td>\n",
       "      <td>0.42531</td>\n",
       "      <td>0.324627</td>\n",
       "      <td>0.472646</td>\n",
       "      <td>0.243146</td>\n",
       "      <td>0.339530</td>\n",
       "      <td>1.090874</td>\n",
       "      <td>1.987470</td>\n",
       "      <td>1.347786</td>\n",
       "      <td>1.547315</td>\n",
       "      <td>129.951090</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>min</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>34.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.180800e+01</td>\n",
       "      <td>4.000000e+01</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25%</td>\n",
       "      <td>72.750000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>183.000000</td>\n",
       "      <td>726.250000</td>\n",
       "      <td>6.108100e+06</td>\n",
       "      <td>1.647275e+04</td>\n",
       "      <td>1833.750000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.422215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50%</td>\n",
       "      <td>132.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>427.500000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>2.626054e+07</td>\n",
       "      <td>4.757350e+04</td>\n",
       "      <td>3897.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>193.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.020191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75%</td>\n",
       "      <td>217.000000</td>\n",
       "      <td>119.000000</td>\n",
       "      <td>218.000000</td>\n",
       "      <td>681.000000</td>\n",
       "      <td>12000.000000</td>\n",
       "      <td>6.334179e+07</td>\n",
       "      <td>1.148290e+05</td>\n",
       "      <td>15823.750000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>375.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>2.088305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>max</td>\n",
       "      <td>813.000000</td>\n",
       "      <td>330.000000</td>\n",
       "      <td>23000.000000</td>\n",
       "      <td>23000.000000</td>\n",
       "      <td>640000.000000</td>\n",
       "      <td>1.200264e+09</td>\n",
       "      <td>1.689764e+06</td>\n",
       "      <td>656730.000000</td>\n",
       "      <td>43.000000</td>\n",
       "      <td>5060.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>7194.485533</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       num_critic_for_reviews     duration  director_facebook_likes  \\\n",
       "count             3852.000000  3852.000000              3852.000000   \n",
       "mean               160.910436   109.049325               762.463396   \n",
       "std                121.840785    21.549751              2988.626715   \n",
       "min                  1.000000    34.000000                 0.000000   \n",
       "25%                 72.750000    95.000000                10.000000   \n",
       "50%                132.000000   105.000000                58.000000   \n",
       "75%                217.000000   119.000000               218.000000   \n",
       "max                813.000000   330.000000             23000.000000   \n",
       "\n",
       "       actor_3_facebook_likes  actor_1_facebook_likes         gross  \\\n",
       "count             3852.000000             3852.000000  3.852000e+03   \n",
       "mean               730.276739             7563.897715  4.929950e+07   \n",
       "std               1805.214756            15431.666830  7.034357e+07   \n",
       "min                  0.000000                0.000000  6.180800e+01   \n",
       "25%                183.000000              726.250000  6.108100e+06   \n",
       "50%                427.500000             1000.000000  2.626054e+07   \n",
       "75%                681.000000            12000.000000  6.334179e+07   \n",
       "max              23000.000000           640000.000000  1.200264e+09   \n",
       "\n",
       "       num_voted_users  cast_total_facebook_likes  facenumber_in_poster  \\\n",
       "count     3.852000e+03                3852.000000           3852.000000   \n",
       "mean      9.813550e+04               11170.345275              1.379283   \n",
       "std       1.468241e+05               18845.309090              2.045882   \n",
       "min       4.000000e+01                   0.000000              0.000000   \n",
       "25%       1.647275e+04                1833.750000              0.000000   \n",
       "50%       4.757350e+04                3897.500000              1.000000   \n",
       "75%       1.148290e+05               15823.750000              2.000000   \n",
       "max       1.689764e+06              656730.000000             43.000000   \n",
       "\n",
       "       num_user_for_reviews  ...  Sci-Fi_Fantasy  Family_Animation  \\\n",
       "count           3852.000000  ...      3852.00000       3852.000000   \n",
       "mean             313.869938  ...         0.23702          0.119678   \n",
       "std              397.309921  ...         0.42531          0.324627   \n",
       "min                1.000000  ...         0.00000          0.000000   \n",
       "25%               98.000000  ...         0.00000          0.000000   \n",
       "50%              193.000000  ...         0.00000          0.000000   \n",
       "75%              375.000000  ...         0.00000          0.000000   \n",
       "max             5060.000000  ...         1.00000          1.000000   \n",
       "\n",
       "       Action_Adventure  History_War       Others  director_rank  actor1_rank  \\\n",
       "count       3852.000000  3852.000000  3852.000000    3852.000000  3852.000000   \n",
       "mean           0.336708     0.063084     0.132918       7.076064     4.046729   \n",
       "std            0.472646     0.243146     0.339530       1.090874     1.987470   \n",
       "min            0.000000     0.000000     0.000000       2.000000     0.000000   \n",
       "25%            0.000000     0.000000     0.000000       7.000000     3.000000   \n",
       "50%            0.000000     0.000000     0.000000       7.000000     3.000000   \n",
       "75%            1.000000     0.000000     0.000000       8.000000     5.000000   \n",
       "max            1.000000     1.000000     1.000000      10.000000    10.000000   \n",
       "\n",
       "       actor2_rank  actor3_rank  gross_budget_ratio  \n",
       "count  3852.000000  3852.000000         3852.000000  \n",
       "mean      3.432503     4.333333            5.690951  \n",
       "std       1.347786     1.547315          129.951090  \n",
       "min       0.000000     0.000000            0.000018  \n",
       "25%       3.000000     3.000000            0.422215  \n",
       "50%       3.000000     4.000000            1.020191  \n",
       "75%       4.000000     5.000000            2.088305  \n",
       "max      10.000000    10.000000         7194.485533  \n",
       "\n",
       "[8 rows x 31 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "movie_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# IMDb rating prediction\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data1 = movie_df[['duration','budget','Biography','Comedy','Crime','Drama','Romance',\n",
    "                  'Mystery_Thriller_Horror','Sci-Fi_Fantasy','Family_Animation','Action_Adventure',\n",
    "                  'History_War','Others','director_rank',\n",
    "                 'actor1_rank','actor2_rank','actor3_rank','movie_facebook_likes','cast_total_facebook_likes'\n",
    "                 ,'facenumber_in_poster']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "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>duration</th>\n",
       "      <th>budget</th>\n",
       "      <th>Biography</th>\n",
       "      <th>Comedy</th>\n",
       "      <th>Crime</th>\n",
       "      <th>Drama</th>\n",
       "      <th>Romance</th>\n",
       "      <th>Mystery_Thriller_Horror</th>\n",
       "      <th>Sci-Fi_Fantasy</th>\n",
       "      <th>Family_Animation</th>\n",
       "      <th>Action_Adventure</th>\n",
       "      <th>History_War</th>\n",
       "      <th>Others</th>\n",
       "      <th>director_rank</th>\n",
       "      <th>actor1_rank</th>\n",
       "      <th>actor2_rank</th>\n",
       "      <th>actor3_rank</th>\n",
       "      <th>movie_facebook_likes</th>\n",
       "      <th>cast_total_facebook_likes</th>\n",
       "      <th>facenumber_in_poster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>123.0</td>\n",
       "      <td>48000000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>22209</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>88.0</td>\n",
       "      <td>3500000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>16000</td>\n",
       "      <td>2726</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>104.0</td>\n",
       "      <td>4500000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1662</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>148.0</td>\n",
       "      <td>27000000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>14000</td>\n",
       "      <td>2566</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>98.0</td>\n",
       "      <td>6000000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>14921</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   duration      budget  Biography  Comedy  Crime  Drama  Romance  \\\n",
       "0     123.0  48000000.0          0       1      0      0        1   \n",
       "1      88.0   3500000.0          0       1      0      0        0   \n",
       "2     104.0   4500000.0          0       0      0      1        1   \n",
       "3     148.0  27000000.0          0       1      1      0        0   \n",
       "4      98.0   6000000.0          0       1      0      0        0   \n",
       "\n",
       "   Mystery_Thriller_Horror  Sci-Fi_Fantasy  Family_Animation  \\\n",
       "0                        0               1                 0   \n",
       "1                        0               0                 0   \n",
       "2                        0               0                 0   \n",
       "3                        0               0                 0   \n",
       "4                        0               0                 0   \n",
       "\n",
       "   Action_Adventure  History_War  Others  director_rank  actor1_rank  \\\n",
       "0                 0            0       0              8            6   \n",
       "1                 0            0       0              8            2   \n",
       "2                 1            0       0              7            3   \n",
       "3                 1            0       1              8            3   \n",
       "4                 0            0       1              7            7   \n",
       "\n",
       "   actor2_rank  actor3_rank  movie_facebook_likes  cast_total_facebook_likes  \\\n",
       "0            3            4                     0                      22209   \n",
       "1            3            4                 16000                       2726   \n",
       "2            3            3                     0                       1662   \n",
       "3            3            4                 14000                       2566   \n",
       "4            3            5                     0                      14921   \n",
       "\n",
       "   facenumber_in_poster  \n",
       "0                   1.0  \n",
       "1                   0.0  \n",
       "2                   0.0  \n",
       "3                   1.0  \n",
       "4                   3.0  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = movie_df['imdb_score']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(data1, y, test_size=0.30)\n",
    "\n",
    "MM = MinMaxScaler(feature_range=(0, 1), copy=True)\n",
    "X_train_MM = MM.fit_transform(X_train) # standardize X_train\n",
    "X_test_MM = MM.transform(X_test) # standardize X_test\n",
    "X_train_scaled = pd.DataFrame(data=X_train_MM,  columns=X_train.columns)\n",
    "X_test_scaled = pd.DataFrame(data=X_test_MM,  columns=X_test.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "linear regression R2: 0.6572611188687424\n",
      "linear regression MSE: [0.35689835]\n"
     ]
    }
   ],
   "source": [
    "lr = LinearRegression()\n",
    "lr.fit(X_train_scaled, y_train)\n",
    "y_pred= lr.predict(X_test_scaled)\n",
    "print(\"linear regression R2:\", r2_score(y_test, y_pred))\n",
    "print(\"linear regression MSE:\", mean_squared_error(y_test, y_pred, multioutput='raw_values'))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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",
       "      <td>3500</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.501149</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>527</td>\n",
       "      <td>6.4</td>\n",
       "      <td>6.223268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>342</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.168241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3845</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.414720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1295</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.570921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2830</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.472629</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3793</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.041705</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>149</td>\n",
       "      <td>7.8</td>\n",
       "      <td>6.864923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1657</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.126586</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2298</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.428024</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "3500     5.3   5.501149\n",
       "527      6.4   6.223268\n",
       "342      7.2   7.168241\n",
       "3845     6.8   6.414720\n",
       "1295     6.0   5.570921\n",
       "...      ...        ...\n",
       "2830     7.1   6.472629\n",
       "3793     7.2   7.041705\n",
       "149      7.8   6.864923\n",
       "1657     5.2   6.126586\n",
       "2298     6.8   6.428024\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df0 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ridge best parameters: {'alpha': 0.4}\n",
      "ridge score: 0.6572666561979906\n",
      "ridge MSE: 0.35689257892946546\n",
      "ridge best estimator coef: [ 0.44179323 -0.26678507  0.18630665 -0.04527685  0.04007917  0.15104169\n",
      " -0.05694108 -0.09288349 -0.07188622  0.04797377 -0.04022566 -0.07803767\n",
      "  0.04048215  5.52511664  0.16220107 -0.08505961 -0.2149567   0.73581104\n",
      " -0.01378189 -0.41686259]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.1,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "ridge = linear_model.Ridge()\n",
    "gridridge = GridSearchCV(ridge, parameters, scoring ='r2')\n",
    "gridridge.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridridge.predict(X_test_scaled)\n",
    "\n",
    "print(\"ridge best parameters:\", gridridge.best_params_)\n",
    "print(\"ridge score:\", gridridge.score(X_test_scaled, y_test))\n",
    "print(\"ridge MSE:\", mean_squared_error(y_test, gridridge.predict(X_test_scaled)))\n",
    "print(\"ridge best estimator coef:\", gridridge.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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",
       "      <td>3500</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.504587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>527</td>\n",
       "      <td>6.4</td>\n",
       "      <td>6.218511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>342</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.162719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3845</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.412220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1295</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.574537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2830</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.473589</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3793</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.037383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>149</td>\n",
       "      <td>7.8</td>\n",
       "      <td>6.855763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1657</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.123514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2298</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.428073</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "3500     5.3   5.504587\n",
       "527      6.4   6.218511\n",
       "342      7.2   7.162719\n",
       "3845     6.8   6.412220\n",
       "1295     6.0   5.574537\n",
       "...      ...        ...\n",
       "2830     7.1   6.473589\n",
       "3793     7.2   7.037383\n",
       "149      7.8   6.855763\n",
       "1657     5.2   6.123514\n",
       "2298     6.8   6.428073\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### lasso\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso best parameters: {'alpha': 0.01}\n",
      "lasso score: 0.6359643128214648\n",
      "lasso MSE: 0.3790749793360078\n",
      "lasso best estimator coef: [ 0.00000000e+00 -0.00000000e+00  9.44766563e-02 -4.09906758e-02\n",
      "  0.00000000e+00  1.80397817e-01 -0.00000000e+00 -3.41501646e-02\n",
      " -3.09732571e-03 -0.00000000e+00 -0.00000000e+00  0.00000000e+00\n",
      "  0.00000000e+00  5.34122847e+00  0.00000000e+00  0.00000000e+00\n",
      " -0.00000000e+00  0.00000000e+00  0.00000000e+00 -0.00000000e+00]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "lasso = linear_model.Lasso()\n",
    "gridlasso = GridSearchCV(lasso, parameters, scoring ='r2')\n",
    "gridlasso.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridlasso.predict(X_test_scaled)\n",
    "\n",
    "print(\"lasso best parameters:\", gridlasso.best_params_)\n",
    "print(\"lasso score:\", gridlasso.score(X_test_scaled, y_test))\n",
    "print(\"lasso MSE:\", mean_squared_error(y_test, gridlasso.predict(X_test_scaled)))\n",
    "print(\"lasso best estimator coef:\", gridlasso.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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",
       "      <td>3500</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.621908</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>527</td>\n",
       "      <td>6.4</td>\n",
       "      <td>6.289562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>342</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.178604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3845</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.296403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1295</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.628749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2830</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.510951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3793</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.178604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>149</td>\n",
       "      <td>7.8</td>\n",
       "      <td>6.960959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1657</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.252315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2298</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.469960</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "3500     5.3   5.621908\n",
       "527      6.4   6.289562\n",
       "342      7.2   7.178604\n",
       "3845     6.8   6.296403\n",
       "1295     6.0   5.628749\n",
       "...      ...        ...\n",
       "2830     7.1   6.510951\n",
       "3793     7.2   7.178604\n",
       "149      7.8   6.960959\n",
       "1657     5.2   6.252315\n",
       "2298     6.8   6.469960\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### elastic net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Elastic Net best parameters: {'alpha': 0.01}\n",
      "Elastic Net score: 0.618808546588226\n",
      "Elastic Net MSE: 0.3969394963583978\n",
      "Elastic Net best estimator coef: [ 1.07433441e-01  0.00000000e+00  1.63009991e-01 -9.68725888e-02\n",
      "  2.19052826e-02  2.50805431e-01 -4.03697251e-02 -9.41675714e-02\n",
      " -2.47621184e-02  0.00000000e+00 -2.55161477e-03  0.00000000e+00\n",
      "  1.67853679e-02  4.19994097e+00  1.13656546e-01 -0.00000000e+00\n",
      " -0.00000000e+00  4.80127657e-01  0.00000000e+00 -0.00000000e+00]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "en = lasso = linear_model.ElasticNet()\n",
    "griden = GridSearchCV(en, parameters, scoring ='r2')\n",
    "griden.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = griden.predict(X_test_scaled)\n",
    "\n",
    "print(\"Elastic Net best parameters:\", griden.best_params_)\n",
    "print(\"Elastic Net score:\", griden.score(X_test_scaled, y_test))\n",
    "print(\"Elastic Net MSE:\", mean_squared_error(y_test, griden.predict(X_test_scaled)))\n",
    "print(\"Elastic Net best estimator coef:\", griden.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "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",
       "      <td>3500</td>\n",
       "      <td>5.3</td>\n",
       "      <td>5.660698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>527</td>\n",
       "      <td>6.4</td>\n",
       "      <td>6.169874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>342</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.047027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3845</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.324273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1295</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.699873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2830</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.531543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3793</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.079266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>149</td>\n",
       "      <td>7.8</td>\n",
       "      <td>6.755058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1657</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.109619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2298</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.456987</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "3500     5.3   5.660698\n",
       "527      6.4   6.169874\n",
       "342      7.2   7.047027\n",
       "3845     6.8   6.324273\n",
       "1295     6.0   5.699873\n",
       "...      ...        ...\n",
       "2830     7.1   6.531543\n",
       "3793     7.2   7.079266\n",
       "149      7.8   6.755058\n",
       "1657     5.2   6.109619\n",
       "2298     6.8   6.456987\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model without genres"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we exclude genre from 1 Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "data2 = movie_df[['duration','budget','director_rank',\n",
    "                 'actor1_rank','actor2_rank','actor3_rank'\n",
    "                  ,'movie_facebook_likes','cast_total_facebook_likes','facenumber_in_poster']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = movie_df['imdb_score']\n",
    "X_train, X_test, y_train, y_test = train_test_split(data2, y, test_size=0.30)\n",
    "\n",
    "MM = MinMaxScaler(feature_range=(0, 1), copy=True)\n",
    "X_train_MM = MM.fit_transform(X_train) # standardize X_train\n",
    "X_test_MM = MM.transform(X_test) # standardize X_test\n",
    "X_train_scaled = pd.DataFrame(data=X_train_MM,  columns=X_train.columns)\n",
    "X_test_scaled = pd.DataFrame(data=X_test_MM,  columns=X_test.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "linear regression MSE: [0.41047127]\n",
      "linear regression R2: 0.6288561315098145\n"
     ]
    }
   ],
   "source": [
    "lr = LinearRegression()\n",
    "lr.fit(X_train_scaled, y_train)\n",
    "y_pred= lr.predict(X_test_scaled)\n",
    "print(\"linear regression MSE:\", mean_squared_error(y_test, y_pred, multioutput='raw_values'))\n",
    "print(\"linear regression R2:\", r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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",
       "      <td>3564</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.317368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3394</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.623572</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1385</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.439094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3604</td>\n",
       "      <td>7.8</td>\n",
       "      <td>7.475219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2625</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.372983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>583</td>\n",
       "      <td>8.2</td>\n",
       "      <td>7.169249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3217</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.312341</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>8.3</td>\n",
       "      <td>7.932686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2530</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.177772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>328</td>\n",
       "      <td>7.5</td>\n",
       "      <td>7.055980</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "3564     6.8   6.317368\n",
       "3394     5.5   5.623572\n",
       "1385     7.1   6.439094\n",
       "3604     7.8   7.475219\n",
       "2625     6.1   6.372983\n",
       "...      ...        ...\n",
       "583      8.2   7.169249\n",
       "3217     8.0   7.312341\n",
       "14       8.3   7.932686\n",
       "2530     7.4   7.177772\n",
       "328      7.5   7.055980\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df4 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ridge best parameters: {'alpha': 0.8}\n",
      "ridge score: 0.6285559131640261\n",
      "ridge MSE: 0.41080330221112404\n",
      "ridge best estimator coef: [ 0.87512296 -0.51741653  5.69517694  0.12465005 -0.0543251  -0.41498827\n",
      "  0.98077289 -0.0158353  -0.34579399]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.1,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "ridge = linear_model.Ridge()\n",
    "gridridge = GridSearchCV(ridge, parameters, scoring ='r2')\n",
    "gridridge.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridridge.predict(X_test_scaled)\n",
    "\n",
    "print(\"ridge best parameters:\", gridridge.best_params_)\n",
    "print(\"ridge score:\", gridridge.score(X_test_scaled, y_test))\n",
    "print(\"ridge MSE:\", mean_squared_error(y_test, gridridge.predict(X_test_scaled)))\n",
    "print(\"ridge best estimator coef:\", gridridge.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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",
       "      <td>3564</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.318677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3394</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.633174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1385</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.437614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3604</td>\n",
       "      <td>7.8</td>\n",
       "      <td>7.453455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2625</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.371992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>583</td>\n",
       "      <td>8.2</td>\n",
       "      <td>7.158986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3217</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.297635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>8.3</td>\n",
       "      <td>7.916281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2530</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.168222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>328</td>\n",
       "      <td>7.5</td>\n",
       "      <td>7.043440</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "3564     6.8   6.318677\n",
       "3394     5.5   5.633174\n",
       "1385     7.1   6.437614\n",
       "3604     7.8   7.453455\n",
       "2625     6.1   6.371992\n",
       "...      ...        ...\n",
       "583      8.2   7.158986\n",
       "3217     8.0   7.297635\n",
       "14       8.3   7.916281\n",
       "2530     7.4   7.168222\n",
       "328      7.5   7.043440\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df5 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso best parameters: {'alpha': 0.01}\n",
      "lasso score: 0.613440550279975\n",
      "lasso MSE: 0.42752032963719183\n",
      "lasso best estimator coef: [ 0.         -0.          5.58836012  0.         -0.         -0.\n",
      "  0.          0.         -0.        ]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "lasso = linear_model.Lasso()\n",
    "gridlasso = GridSearchCV(lasso, parameters, scoring ='r2')\n",
    "gridlasso.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridlasso.predict(X_test_scaled)\n",
    "\n",
    "print(\"lasso best parameters:\", gridlasso.best_params_)\n",
    "print(\"lasso score:\", gridlasso.score(X_test_scaled, y_test))\n",
    "print(\"lasso MSE:\", mean_squared_error(y_test, gridlasso.predict(X_test_scaled)))\n",
    "print(\"lasso best estimator coef:\", gridlasso.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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",
       "      <td>3564</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.393130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3394</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.694585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1385</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.393130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3604</td>\n",
       "      <td>7.8</td>\n",
       "      <td>7.091675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2625</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.393130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>583</td>\n",
       "      <td>8.2</td>\n",
       "      <td>7.091675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3217</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.091675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>8.3</td>\n",
       "      <td>7.790220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2530</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.091675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>328</td>\n",
       "      <td>7.5</td>\n",
       "      <td>7.091675</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "3564     6.8   6.393130\n",
       "3394     5.5   5.694585\n",
       "1385     7.1   6.393130\n",
       "3604     7.8   7.091675\n",
       "2625     6.1   6.393130\n",
       "...      ...        ...\n",
       "583      8.2   7.091675\n",
       "3217     8.0   7.091675\n",
       "14       8.3   7.790220\n",
       "2530     7.4   7.091675\n",
       "328      7.5   7.091675\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df6 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df6"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### elastic net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso best parameters: {'alpha': 0.01}\n",
      "lasso score: 0.5885915512325238\n",
      "lasso MSE: 0.45500239551765315\n",
      "lasso best estimator coef: [ 0.44074999 -0.          4.49671059  0.09912373 -0.         -0.15709035\n",
      "  0.25589799  0.         -0.        ]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "en = linear_model.ElasticNet()\n",
    "gridlasso = GridSearchCV(en, parameters, scoring ='r2')\n",
    "gridlasso.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridlasso.predict(X_test_scaled)\n",
    "\n",
    "print(\"lasso best parameters:\", gridlasso.best_params_)\n",
    "print(\"lasso score:\", gridlasso.score(X_test_scaled, y_test))\n",
    "print(\"lasso MSE:\", mean_squared_error(y_test, gridlasso.predict(X_test_scaled)))\n",
    "print(\"lasso best estimator coef:\", gridlasso.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "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",
       "      <td>3564</td>\n",
       "      <td>6.8</td>\n",
       "      <td>6.378513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3394</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.815971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1385</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.408776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3604</td>\n",
       "      <td>7.8</td>\n",
       "      <td>7.060800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2625</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.388401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>583</td>\n",
       "      <td>8.2</td>\n",
       "      <td>7.002821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3217</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.031396</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>8.3</td>\n",
       "      <td>7.606460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2530</td>\n",
       "      <td>7.4</td>\n",
       "      <td>6.995934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>328</td>\n",
       "      <td>7.5</td>\n",
       "      <td>6.938634</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "3564     6.8   6.378513\n",
       "3394     5.5   5.815971\n",
       "1385     7.1   6.408776\n",
       "3604     7.8   7.060800\n",
       "2625     6.1   6.388401\n",
       "...      ...        ...\n",
       "583      8.2   7.002821\n",
       "3217     8.0   7.031396\n",
       "14       8.3   7.606460\n",
       "2530     7.4   6.995934\n",
       "328      7.5   6.938634\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df7 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df7"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Model with logs\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we will transform  budget s to log values. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "data3 = movie_df[['duration','Biography','Comedy','Crime','Drama','Romance',\n",
    "                  'Mystery_Thriller_Horror','Sci-Fi_Fantasy','Family_Animation','Action_Adventure',\n",
    "                  'History_War','Others','director_rank',\n",
    "                 'actor1_rank','actor2_rank','actor3_rank','movie_facebook_likes','cast_total_facebook_likes'\n",
    "                 ,'facenumber_in_poster']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "budget = np.log(movie_df['budget'])\n",
    "data3['log_budget'] = budget"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = movie_df['imdb_score']\n",
    "X_train, X_test, y_train, y_test = train_test_split(data3, y, test_size=0.30)\n",
    "MM = MinMaxScaler(feature_range=(0, 1), copy=True)\n",
    "X_train_MM = MM.fit_transform(X_train) # standardize X_train\n",
    "X_test_MM = MM.transform(X_test) # standardize X_test\n",
    "X_train_scaled = pd.DataFrame(data=X_train_MM,  columns=X_train.columns)\n",
    "X_test_scaled = pd.DataFrame(data=X_test_MM,  columns=X_test.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "linear regression MSE: [0.34414592]\n",
      "linear regression R2: 0.6734455512999777\n"
     ]
    }
   ],
   "source": [
    "lr = LinearRegression()\n",
    "lr.fit(X_train_scaled, y_train)\n",
    "y_pred= lr.predict(X_test_scaled)\n",
    "print(\"linear regression MSE:\", mean_squared_error(y_test, y_pred, multioutput='raw_values'))\n",
    "print(\"linear regression R2:\", r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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",
       "      <td>1055</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.507544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>362</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.189431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>944</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.117240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>313</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.819918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2863</td>\n",
       "      <td>5.6</td>\n",
       "      <td>6.290665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1885</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.352130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2895</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.467326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2595</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.458672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3085</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.195311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2982</td>\n",
       "      <td>6.6</td>\n",
       "      <td>6.425428</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "1055     7.0   7.507544\n",
       "362      6.9   7.189431\n",
       "944      7.2   7.117240\n",
       "313      8.0   7.819918\n",
       "2863     5.6   6.290665\n",
       "...      ...        ...\n",
       "1885     6.0   6.352130\n",
       "2895     6.0   6.467326\n",
       "2595     5.5   5.458672\n",
       "3085     7.2   7.195311\n",
       "2982     6.6   6.425428\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df8 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df8"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ridge best parameters: {'alpha': 0.8}\n",
      "ridge score: 0.673469671877164\n",
      "ridge MSE: 0.3441205021597735\n",
      "ridge best estimator coef: [ 7.97079283e-01  1.65655443e-01 -4.83601526e-02  5.54794981e-02\n",
      "  1.59724261e-01 -6.02647863e-02 -1.05283042e-01 -3.57563307e-02\n",
      "  8.08446776e-02  6.25446972e-03 -8.16055139e-02  2.90919083e-02\n",
      "  5.42791837e+00  2.26491766e-01 -4.30623374e-03 -2.30952422e-01\n",
      "  1.49171699e+00  1.35627148e-01 -5.31778270e-01 -1.37629729e+00]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.1,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "ridge = linear_model.Ridge()\n",
    "gridridge = GridSearchCV(ridge, parameters, scoring ='r2')\n",
    "gridridge.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridridge.predict(X_test_scaled)\n",
    "\n",
    "print(\"ridge best parameters:\", gridridge.best_params_)\n",
    "print(\"ridge score:\", gridridge.score(X_test_scaled, y_test))\n",
    "print(\"ridge MSE:\", mean_squared_error(y_test, gridridge.predict(X_test_scaled)))\n",
    "print(\"ridge best estimator coef:\", gridridge.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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",
       "      <td>1055</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.481161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>362</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.178747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>944</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.107600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>313</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.792566</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2863</td>\n",
       "      <td>5.6</td>\n",
       "      <td>6.301613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1885</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.365336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2895</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.473477</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2595</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.466789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3085</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.183231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2982</td>\n",
       "      <td>6.6</td>\n",
       "      <td>6.436457</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "1055     7.0   7.481161\n",
       "362      6.9   7.178747\n",
       "944      7.2   7.107600\n",
       "313      8.0   7.792566\n",
       "2863     5.6   6.301613\n",
       "...      ...        ...\n",
       "1885     6.0   6.365336\n",
       "2895     6.0   6.473477\n",
       "2595     5.5   5.466789\n",
       "3085     7.2   7.183231\n",
       "2982     6.6   6.436457\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df9 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df9"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### lasso\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso best parameters: {'alpha': 0.01}\n",
      "lasso score: 0.6539096857639983\n",
      "lasso MSE: 0.36473418384195\n",
      "lasso best estimator coef: [ 0.          0.08402777 -0.05897211  0.          0.20697237 -0.0105958\n",
      " -0.04670987 -0.         -0.         -0.         -0.          0.\n",
      "  5.28582972  0.         -0.         -0.          0.          0.\n",
      " -0.         -0.        ]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "lasso = linear_model.Lasso()\n",
    "gridlasso = GridSearchCV(lasso, parameters, scoring ='r2')\n",
    "gridlasso.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridlasso.predict(X_test_scaled)\n",
    "\n",
    "print(\"lasso best parameters:\", gridlasso.best_params_)\n",
    "print(\"lasso score:\", gridlasso.score(X_test_scaled, y_test))\n",
    "print(\"lasso MSE:\", mean_squared_error(y_test, gridlasso.predict(X_test_scaled)))\n",
    "print(\"lasso best estimator coef:\", gridlasso.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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",
       "      <td>1055</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.190864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>362</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.131891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>944</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.180268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>313</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.190864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2863</td>\n",
       "      <td>5.6</td>\n",
       "      <td>6.323162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1885</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.530135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2895</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.483425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2595</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.592866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3085</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.180268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2982</td>\n",
       "      <td>6.6</td>\n",
       "      <td>6.471163</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "1055     7.0   7.190864\n",
       "362      6.9   7.131891\n",
       "944      7.2   7.180268\n",
       "313      8.0   7.190864\n",
       "2863     5.6   6.323162\n",
       "...      ...        ...\n",
       "1885     6.0   6.530135\n",
       "2895     6.0   6.483425\n",
       "2595     5.5   5.592866\n",
       "3085     7.2   7.180268\n",
       "2982     6.6   6.471163\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df10 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df10"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### elastic net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso best parameters: {'alpha': 0.01}\n",
      "lasso score: 0.6254588297680825\n",
      "lasso MSE: 0.39471768616614095\n",
      "lasso best estimator coef: [ 7.63940570e-02  1.55393489e-01 -1.24683372e-01  2.46963222e-02\n",
      "  2.64556436e-01 -5.52614942e-02 -1.03472613e-01 -7.67372318e-03\n",
      " -0.00000000e+00 -1.57608953e-02  0.00000000e+00  3.03373986e-03\n",
      "  4.20452615e+00  8.84849107e-02 -0.00000000e+00 -2.79264489e-02\n",
      "  3.66037082e-01  0.00000000e+00 -0.00000000e+00 -4.12337124e-02]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "en = linear_model.ElasticNet()\n",
    "gridlasso = GridSearchCV(en, parameters, scoring ='r2')\n",
    "gridlasso.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridlasso.predict(X_test_scaled)\n",
    "\n",
    "print(\"lasso best parameters:\", gridlasso.best_params_)\n",
    "print(\"lasso score:\", gridlasso.score(X_test_scaled, y_test))\n",
    "print(\"lasso MSE:\", mean_squared_error(y_test, gridlasso.predict(X_test_scaled)))\n",
    "print(\"lasso best estimator coef:\", gridlasso.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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",
       "      <td>1055</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.131565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>362</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.015727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>944</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.073311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>313</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.216350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2863</td>\n",
       "      <td>5.6</td>\n",
       "      <td>6.364951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1885</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.600321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2895</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.546913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2595</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.640523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3085</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.101906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2982</td>\n",
       "      <td>6.6</td>\n",
       "      <td>6.529827</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "1055     7.0   7.131565\n",
       "362      6.9   7.015727\n",
       "944      7.2   7.073311\n",
       "313      8.0   7.216350\n",
       "2863     5.6   6.364951\n",
       "...      ...        ...\n",
       "1885     6.0   6.600321\n",
       "2895     6.0   6.546913\n",
       "2595     5.5   5.640523\n",
       "3085     7.2   7.101906\n",
       "2982     6.6   6.529827\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df11 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df11"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "scrolled": true
   },
   "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>duration</th>\n",
       "      <th>Biography</th>\n",
       "      <th>Comedy</th>\n",
       "      <th>Crime</th>\n",
       "      <th>Drama</th>\n",
       "      <th>Romance</th>\n",
       "      <th>Mystery_Thriller_Horror</th>\n",
       "      <th>Sci-Fi_Fantasy</th>\n",
       "      <th>Family_Animation</th>\n",
       "      <th>Action_Adventure</th>\n",
       "      <th>History_War</th>\n",
       "      <th>Others</th>\n",
       "      <th>director_rank</th>\n",
       "      <th>actor1_rank</th>\n",
       "      <th>actor2_rank</th>\n",
       "      <th>actor3_rank</th>\n",
       "      <th>movie_facebook_likes</th>\n",
       "      <th>cast_total_facebook_likes</th>\n",
       "      <th>facenumber_in_poster</th>\n",
       "      <th>log_budget</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>123.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>22209</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.686712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>88.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>16000</td>\n",
       "      <td>2726</td>\n",
       "      <td>0.0</td>\n",
       "      <td>15.068274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>104.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1662</td>\n",
       "      <td>0.0</td>\n",
       "      <td>15.319588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>148.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>14000</td>\n",
       "      <td>2566</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.111347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>98.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>14921</td>\n",
       "      <td>3.0</td>\n",
       "      <td>15.607270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3847</td>\n",
       "      <td>92.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>17860</td>\n",
       "      <td>0.0</td>\n",
       "      <td>17.370859</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3848</td>\n",
       "      <td>100.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>18000</td>\n",
       "      <td>18786</td>\n",
       "      <td>9.0</td>\n",
       "      <td>16.811243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3849</td>\n",
       "      <td>92.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>11000</td>\n",
       "      <td>2848</td>\n",
       "      <td>4.0</td>\n",
       "      <td>15.424948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3850</td>\n",
       "      <td>81.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3344</td>\n",
       "      <td>0.0</td>\n",
       "      <td>15.404746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3851</td>\n",
       "      <td>120.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1810</td>\n",
       "      <td>11.0</td>\n",
       "      <td>15.424948</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3852 rows × 20 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      duration  Biography  Comedy  Crime  Drama  Romance  \\\n",
       "0        123.0          0       1      0      0        1   \n",
       "1         88.0          0       1      0      0        0   \n",
       "2        104.0          0       0      0      1        1   \n",
       "3        148.0          0       1      1      0        0   \n",
       "4         98.0          0       1      0      0        0   \n",
       "...        ...        ...     ...    ...    ...      ...   \n",
       "3847      92.0          0       1      0      0        0   \n",
       "3848     100.0          0       1      0      0        0   \n",
       "3849      92.0          0       1      0      0        0   \n",
       "3850      81.0          0       0      0      0        0   \n",
       "3851     120.0          0       0      0      1        0   \n",
       "\n",
       "      Mystery_Thriller_Horror  Sci-Fi_Fantasy  Family_Animation  \\\n",
       "0                           0               1                 0   \n",
       "1                           0               0                 0   \n",
       "2                           0               0                 0   \n",
       "3                           0               0                 0   \n",
       "4                           0               0                 0   \n",
       "...                       ...             ...               ...   \n",
       "3847                        0               0                 0   \n",
       "3848                        0               0                 0   \n",
       "3849                        0               0                 0   \n",
       "3850                        1               0                 0   \n",
       "3851                        0               0                 0   \n",
       "\n",
       "      Action_Adventure  History_War  Others  director_rank  actor1_rank  \\\n",
       "0                    0            0       0              8            6   \n",
       "1                    0            0       0              8            2   \n",
       "2                    1            0       0              7            3   \n",
       "3                    1            0       1              8            3   \n",
       "4                    0            0       1              7            7   \n",
       "...                ...          ...     ...            ...          ...   \n",
       "3847                 0            0       0              7            5   \n",
       "3848                 0            0       0              7            5   \n",
       "3849                 0            0       0              5            2   \n",
       "3850                 0            0       0              7            3   \n",
       "3851                 0            0       0              3            2   \n",
       "\n",
       "      actor2_rank  actor3_rank  movie_facebook_likes  \\\n",
       "0               3            4                     0   \n",
       "1               3            4                 16000   \n",
       "2               3            3                     0   \n",
       "3               3            4                 14000   \n",
       "4               3            5                     0   \n",
       "...           ...          ...                   ...   \n",
       "3847            3            4                     0   \n",
       "3848            5            5                 18000   \n",
       "3849            4            4                 11000   \n",
       "3850            3            4                     0   \n",
       "3851            3            4                     0   \n",
       "\n",
       "      cast_total_facebook_likes  facenumber_in_poster  log_budget  \n",
       "0                         22209                   1.0   17.686712  \n",
       "1                          2726                   0.0   15.068274  \n",
       "2                          1662                   0.0   15.319588  \n",
       "3                          2566                   1.0   17.111347  \n",
       "4                         14921                   3.0   15.607270  \n",
       "...                         ...                   ...         ...  \n",
       "3847                      17860                   0.0   17.370859  \n",
       "3848                      18786                   9.0   16.811243  \n",
       "3849                       2848                   4.0   15.424948  \n",
       "3850                       3344                   0.0   15.404746  \n",
       "3851                       1810                  11.0   15.424948  \n",
       "\n",
       "[3852 rows x 20 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model only with films with likes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a17680510>"
      ]
     },
     "execution_count": 40,
     "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": [
    "sns.regplot(movie_df['movie_facebook_likes'], movie_df['imdb_score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1a4eb950>"
      ]
     },
     "execution_count": 41,
     "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": [
    "sns.regplot(movie_df['title_year'],movie_df['movie_facebook_likes'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we include only movies with facebook likes and facebook cast likes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "new = movie_df.query(\"movie_facebook_likes != 0\")\n",
    "new = new.query(\"cast_total_facebook_likes != 0\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "data4 = new[['duration','budget','Biography','Comedy','Crime','Drama','Romance',\n",
    "                  'Mystery_Thriller_Horror','Sci-Fi_Fantasy','Family_Animation','Action_Adventure',\n",
    "                  'History_War','Others','director_rank',\n",
    "                 'actor1_rank','actor2_rank','actor3_rank'\n",
    "                  ,'movie_facebook_likes','cast_total_facebook_likes','facenumber_in_poster']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = movie_df['imdb_score']\n",
    "X_train, X_test, y_train, y_test = train_test_split(data4, y, test_size=0.30)\n",
    "MM = MinMaxScaler(feature_range=(0, 1), copy=True)\n",
    "X_train_MM = MM.fit_transform(X_train) # standardize X_train\n",
    "X_test_MM = MM.transform(X_test) # standardize X_test\n",
    "X_train_scaled = pd.DataFrame(data=X_train_MM,  columns=X_train.columns)\n",
    "X_test_scaled = pd.DataFrame(data=X_test_MM,  columns=X_test.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "linear regression MSE: [0.34414592]\n",
      "linear regression R2: 0.6734455512999777\n"
     ]
    }
   ],
   "source": [
    "lr = LinearRegression()\n",
    "lr.fit(X_train_scaled, y_train)\n",
    "y_pred= lr.predict(X_test_scaled)\n",
    "print(\"linear regression MSE:\", mean_squared_error(y_test, y_pred, multioutput='raw_values'))\n",
    "print(\"linear regression R2:\", r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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",
       "      <td>1055</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.507544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>362</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.189431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>944</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.117240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>313</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.819918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2863</td>\n",
       "      <td>5.6</td>\n",
       "      <td>6.290665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1885</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.352130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2895</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.467326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2595</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.458672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3085</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.195311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2982</td>\n",
       "      <td>6.6</td>\n",
       "      <td>6.425428</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "1055     7.0   7.507544\n",
       "362      6.9   7.189431\n",
       "944      7.2   7.117240\n",
       "313      8.0   7.819918\n",
       "2863     5.6   6.290665\n",
       "...      ...        ...\n",
       "1885     6.0   6.352130\n",
       "2895     6.0   6.467326\n",
       "2595     5.5   5.458672\n",
       "3085     7.2   7.195311\n",
       "2982     6.6   6.425428\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df12 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df12"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ridge best parameters: {'alpha': 0.8}\n",
      "ridge score: 0.673469671877164\n",
      "ridge MSE: 0.3441205021597735\n",
      "ridge best estimator coef: [ 7.97079283e-01  1.65655443e-01 -4.83601526e-02  5.54794981e-02\n",
      "  1.59724261e-01 -6.02647863e-02 -1.05283042e-01 -3.57563307e-02\n",
      "  8.08446776e-02  6.25446972e-03 -8.16055139e-02  2.90919083e-02\n",
      "  5.42791837e+00  2.26491766e-01 -4.30623374e-03 -2.30952422e-01\n",
      "  1.49171699e+00  1.35627148e-01 -5.31778270e-01 -1.37629729e+00]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.1,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "ridge = linear_model.Ridge()\n",
    "gridridge = GridSearchCV(ridge, parameters, scoring ='r2')\n",
    "gridridge.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridridge.predict(X_test_scaled)\n",
    "\n",
    "print(\"ridge best parameters:\", gridridge.best_params_)\n",
    "print(\"ridge score:\", gridridge.score(X_test_scaled, y_test))\n",
    "print(\"ridge MSE:\", mean_squared_error(y_test, gridridge.predict(X_test_scaled)))\n",
    "print(\"ridge best estimator coef:\", gridridge.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "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",
       "      <td>1055</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.481161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>362</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.178747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>944</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.107600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>313</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.792566</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2863</td>\n",
       "      <td>5.6</td>\n",
       "      <td>6.301613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1885</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.365336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2895</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.473477</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2595</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.466789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3085</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.183231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2982</td>\n",
       "      <td>6.6</td>\n",
       "      <td>6.436457</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "1055     7.0   7.481161\n",
       "362      6.9   7.178747\n",
       "944      7.2   7.107600\n",
       "313      8.0   7.792566\n",
       "2863     5.6   6.301613\n",
       "...      ...        ...\n",
       "1885     6.0   6.365336\n",
       "2895     6.0   6.473477\n",
       "2595     5.5   5.466789\n",
       "3085     7.2   7.183231\n",
       "2982     6.6   6.436457\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df13 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df13"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso best parameters: {'alpha': 0.01}\n",
      "lasso score: 0.6539096857639983\n",
      "lasso MSE: 0.36473418384195\n",
      "lasso best estimator coef: [ 0.          0.08402777 -0.05897211  0.          0.20697237 -0.0105958\n",
      " -0.04670987 -0.         -0.         -0.         -0.          0.\n",
      "  5.28582972  0.         -0.         -0.          0.          0.\n",
      " -0.         -0.        ]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "lasso = linear_model.Lasso()\n",
    "gridlasso = GridSearchCV(lasso, parameters, scoring ='r2')\n",
    "gridlasso.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridlasso.predict(X_test_scaled)\n",
    "\n",
    "print(\"lasso best parameters:\", gridlasso.best_params_)\n",
    "print(\"lasso score:\", gridlasso.score(X_test_scaled, y_test))\n",
    "print(\"lasso MSE:\", mean_squared_error(y_test, gridlasso.predict(X_test_scaled)))\n",
    "print(\"lasso best estimator coef:\", gridlasso.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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",
       "      <td>1055</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.190864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>362</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.131891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>944</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.180268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>313</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.190864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2863</td>\n",
       "      <td>5.6</td>\n",
       "      <td>6.323162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1885</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.530135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2895</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.483425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2595</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.592866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3085</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.180268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2982</td>\n",
       "      <td>6.6</td>\n",
       "      <td>6.471163</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "1055     7.0   7.190864\n",
       "362      6.9   7.131891\n",
       "944      7.2   7.180268\n",
       "313      8.0   7.190864\n",
       "2863     5.6   6.323162\n",
       "...      ...        ...\n",
       "1885     6.0   6.530135\n",
       "2895     6.0   6.483425\n",
       "2595     5.5   5.592866\n",
       "3085     7.2   7.180268\n",
       "2982     6.6   6.471163\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df14 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df14"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### elastic net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso best parameters: {'alpha': 0.01}\n",
      "lasso score: 0.6254588297680825\n",
      "lasso MSE: 0.39471768616614095\n",
      "lasso best estimator coef: [ 7.63940570e-02  1.55393489e-01 -1.24683372e-01  2.46963222e-02\n",
      "  2.64556436e-01 -5.52614942e-02 -1.03472613e-01 -7.67372318e-03\n",
      " -0.00000000e+00 -1.57608953e-02  0.00000000e+00  3.03373986e-03\n",
      "  4.20452615e+00  8.84849107e-02 -0.00000000e+00 -2.79264489e-02\n",
      "  3.66037082e-01  0.00000000e+00 -0.00000000e+00 -4.12337124e-02]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "en = linear_model.ElasticNet()\n",
    "gridlasso = GridSearchCV(en, parameters, scoring ='r2')\n",
    "gridlasso.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridlasso.predict(X_test_scaled)\n",
    "\n",
    "print(\"lasso best parameters:\", gridlasso.best_params_)\n",
    "print(\"lasso score:\", gridlasso.score(X_test_scaled, y_test))\n",
    "print(\"lasso MSE:\", mean_squared_error(y_test, gridlasso.predict(X_test_scaled)))\n",
    "print(\"lasso best estimator coef:\", gridlasso.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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",
       "      <td>1055</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.131565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>362</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.015727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>944</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.073311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>313</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7.216350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2863</td>\n",
       "      <td>5.6</td>\n",
       "      <td>6.364951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1885</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.600321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2895</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.546913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2595</td>\n",
       "      <td>5.5</td>\n",
       "      <td>5.640523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3085</td>\n",
       "      <td>7.2</td>\n",
       "      <td>7.101906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2982</td>\n",
       "      <td>6.6</td>\n",
       "      <td>6.529827</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "1055     7.0   7.131565\n",
       "362      6.9   7.015727\n",
       "944      7.2   7.073311\n",
       "313      8.0   7.216350\n",
       "2863     5.6   6.364951\n",
       "...      ...        ...\n",
       "1885     6.0   6.600321\n",
       "2895     6.0   6.546913\n",
       "2595     5.5   5.640523\n",
       "3085     7.2   7.101906\n",
       "2982     6.6   6.529827\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df15 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df15"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model without likes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "MSE lower than in the model with likes, but R2 is lower"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "data5 = movie_df[['duration','budget','Biography','Comedy','Crime','Drama','Romance',\n",
    "                  'Mystery_Thriller_Horror','Sci-Fi_Fantasy','Family_Animation','Action_Adventure',\n",
    "                  'History_War','Others','director_rank',\n",
    "                 'actor1_rank','actor2_rank','actor3_rank'\n",
    "                 ,'facenumber_in_poster']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "scrolled": true
   },
   "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>duration</th>\n",
       "      <th>budget</th>\n",
       "      <th>Biography</th>\n",
       "      <th>Comedy</th>\n",
       "      <th>Crime</th>\n",
       "      <th>Drama</th>\n",
       "      <th>Romance</th>\n",
       "      <th>Mystery_Thriller_Horror</th>\n",
       "      <th>Sci-Fi_Fantasy</th>\n",
       "      <th>Family_Animation</th>\n",
       "      <th>Action_Adventure</th>\n",
       "      <th>History_War</th>\n",
       "      <th>Others</th>\n",
       "      <th>director_rank</th>\n",
       "      <th>actor1_rank</th>\n",
       "      <th>actor2_rank</th>\n",
       "      <th>actor3_rank</th>\n",
       "      <th>facenumber_in_poster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>123.0</td>\n",
       "      <td>48000000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>88.0</td>\n",
       "      <td>3500000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>104.0</td>\n",
       "      <td>4500000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>148.0</td>\n",
       "      <td>27000000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>98.0</td>\n",
       "      <td>6000000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   duration      budget  Biography  Comedy  Crime  Drama  Romance  \\\n",
       "0     123.0  48000000.0          0       1      0      0        1   \n",
       "1      88.0   3500000.0          0       1      0      0        0   \n",
       "2     104.0   4500000.0          0       0      0      1        1   \n",
       "3     148.0  27000000.0          0       1      1      0        0   \n",
       "4      98.0   6000000.0          0       1      0      0        0   \n",
       "\n",
       "   Mystery_Thriller_Horror  Sci-Fi_Fantasy  Family_Animation  \\\n",
       "0                        0               1                 0   \n",
       "1                        0               0                 0   \n",
       "2                        0               0                 0   \n",
       "3                        0               0                 0   \n",
       "4                        0               0                 0   \n",
       "\n",
       "   Action_Adventure  History_War  Others  director_rank  actor1_rank  \\\n",
       "0                 0            0       0              8            6   \n",
       "1                 0            0       0              8            2   \n",
       "2                 1            0       0              7            3   \n",
       "3                 1            0       1              8            3   \n",
       "4                 0            0       1              7            7   \n",
       "\n",
       "   actor2_rank  actor3_rank  facenumber_in_poster  \n",
       "0            3            4                   1.0  \n",
       "1            3            4                   0.0  \n",
       "2            3            3                   0.0  \n",
       "3            3            4                   1.0  \n",
       "4            3            5                   3.0  "
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data5.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = movie_df['imdb_score']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(data5, y, test_size=0.30)\n",
    "\n",
    "MM = MinMaxScaler(feature_range=(0, 1), copy=True)\n",
    "X_train_MM = MM.fit_transform(X_train) # standardize X_train\n",
    "X_test_MM = MM.transform(X_test) # standardize X_test\n",
    "X_train_scaled = pd.DataFrame(data=X_train_MM,  columns=X_train.columns)\n",
    "X_test_scaled = pd.DataFrame(data=X_test_MM,  columns=X_test.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "linear regression R2: 0.638246281983388\n",
      "linear regression MSE: [0.38319442]\n"
     ]
    }
   ],
   "source": [
    "lr = LinearRegression()\n",
    "lr.fit(X_train_scaled, y_train)\n",
    "y_pred= lr.predict(X_test_scaled)\n",
    "print(\"linear regression R2:\", r2_score(y_test, y_pred))\n",
    "print(\"linear regression MSE:\", mean_squared_error(y_test, y_pred, multioutput='raw_values'))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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",
       "      <td>2257</td>\n",
       "      <td>6.1</td>\n",
       "      <td>7.103313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3519</td>\n",
       "      <td>6.5</td>\n",
       "      <td>6.952845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3488</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.843445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1635</td>\n",
       "      <td>5.4</td>\n",
       "      <td>5.738196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1073</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.216591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3782</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.300338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2081</td>\n",
       "      <td>7.3</td>\n",
       "      <td>7.184396</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2372</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.015314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2753</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.455119</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2713</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.281529</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "2257     6.1   7.103313\n",
       "3519     6.5   6.952845\n",
       "3488     8.0   6.843445\n",
       "1635     5.4   5.738196\n",
       "1073     7.0   7.216591\n",
       "...      ...        ...\n",
       "3782     6.9   7.300338\n",
       "2081     7.3   7.184396\n",
       "2372     7.4   7.015314\n",
       "2753     6.1   6.455119\n",
       "2713     6.1   6.281529\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df16 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df16"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ridge best parameters: {'alpha': 0.2}\n",
      "ridge score: 0.6383946418952156\n",
      "ridge MSE: 0.3830372626272397\n",
      "ridge best estimator coef: [ 0.49108169 -0.34003549  0.21281498 -0.04399127  0.05561663  0.16299694\n",
      " -0.08150637 -0.08684578 -0.04728036  0.01104862 -0.0250949  -0.09142457\n",
      "  0.03796864  5.69734271  0.10771686  0.0558603  -0.24924464 -0.68902482]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.1,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "ridge = linear_model.Ridge()\n",
    "gridridge = GridSearchCV(ridge, parameters, scoring ='r2')\n",
    "gridridge.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridridge.predict(X_test_scaled)\n",
    "\n",
    "print(\"ridge best parameters:\", gridridge.best_params_)\n",
    "print(\"ridge score:\", gridridge.score(X_test_scaled, y_test))\n",
    "print(\"ridge MSE:\", mean_squared_error(y_test, gridridge.predict(X_test_scaled)))\n",
    "print(\"ridge best estimator coef:\", gridridge.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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",
       "      <td>2257</td>\n",
       "      <td>6.1</td>\n",
       "      <td>7.100834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3519</td>\n",
       "      <td>6.5</td>\n",
       "      <td>6.950437</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3488</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.843242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1635</td>\n",
       "      <td>5.4</td>\n",
       "      <td>5.742609</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1073</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.214138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3782</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.297838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2081</td>\n",
       "      <td>7.3</td>\n",
       "      <td>7.182400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2372</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.011702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2753</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.454707</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2713</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.280151</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "2257     6.1   7.100834\n",
       "3519     6.5   6.950437\n",
       "3488     8.0   6.843242\n",
       "1635     5.4   5.742609\n",
       "1073     7.0   7.214138\n",
       "...      ...        ...\n",
       "3782     6.9   7.297838\n",
       "2081     7.3   7.182400\n",
       "2372     7.4   7.011702\n",
       "2753     6.1   6.454707\n",
       "2713     6.1   6.280151\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df17 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df17"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### lasso\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso best parameters: {'alpha': 0.01}\n",
      "lasso score: 0.6312683923530386\n",
      "lasso MSE: 0.39058587620901936\n",
      "lasso best estimator coef: [ 0.00000000e+00 -0.00000000e+00  1.14742493e-01 -4.17976818e-02\n",
      "  0.00000000e+00  1.96085115e-01 -7.63965719e-03 -1.14300798e-02\n",
      " -3.08330311e-03 -0.00000000e+00 -0.00000000e+00  0.00000000e+00\n",
      "  0.00000000e+00  5.32755885e+00  0.00000000e+00  0.00000000e+00\n",
      " -0.00000000e+00 -0.00000000e+00]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "lasso = linear_model.Lasso()\n",
    "gridlasso = GridSearchCV(lasso, parameters, scoring ='r2')\n",
    "gridlasso.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = gridlasso.predict(X_test_scaled)\n",
    "\n",
    "print(\"lasso best parameters:\", gridlasso.best_params_)\n",
    "print(\"lasso score:\", gridlasso.score(X_test_scaled, y_test))\n",
    "print(\"lasso MSE:\", mean_squared_error(y_test, gridlasso.predict(X_test_scaled)))\n",
    "print(\"lasso best estimator coef:\", gridlasso.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "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",
       "      <td>2257</td>\n",
       "      <td>6.1</td>\n",
       "      <td>7.154443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3519</td>\n",
       "      <td>6.5</td>\n",
       "      <td>6.955275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3488</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.955275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1635</td>\n",
       "      <td>5.4</td>\n",
       "      <td>5.822554</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1073</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.165873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3782</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.165873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2081</td>\n",
       "      <td>7.3</td>\n",
       "      <td>7.162790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2372</td>\n",
       "      <td>7.4</td>\n",
       "      <td>7.113353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2753</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.458131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2713</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.254406</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "2257     6.1   7.154443\n",
       "3519     6.5   6.955275\n",
       "3488     8.0   6.955275\n",
       "1635     5.4   5.822554\n",
       "1073     7.0   7.165873\n",
       "...      ...        ...\n",
       "3782     6.9   7.165873\n",
       "2081     7.3   7.162790\n",
       "2372     7.4   7.113353\n",
       "2753     6.1   6.458131\n",
       "2713     6.1   6.254406\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df18 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df18"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### elastic net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Elastic Net best parameters: {'alpha': 0.01}\n",
      "Elastic Net score: 0.6055653031024515\n",
      "Elastic Net MSE: 0.417812355924941\n",
      "Elastic Net best estimator coef: [ 0.03138012  0.          0.18370308 -0.11435287  0.03204114  0.25521483\n",
      " -0.04816495 -0.0747615  -0.01828177 -0.         -0.          0.\n",
      "  0.0097674   4.29317948  0.08335239  0.         -0.         -0.        ]\n"
     ]
    }
   ],
   "source": [
    "parameters = {'alpha': np.concatenate((np.arange(0.01,2,0.1), np.arange(2, 5, 0.5), np.arange(5, 25, 1)))}\n",
    "en = lasso = linear_model.ElasticNet()\n",
    "griden = GridSearchCV(en, parameters, scoring ='r2')\n",
    "griden.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = griden.predict(X_test_scaled)\n",
    "\n",
    "print(\"Elastic Net best parameters:\", griden.best_params_)\n",
    "print(\"Elastic Net score:\", griden.score(X_test_scaled, y_test))\n",
    "print(\"Elastic Net MSE:\", mean_squared_error(y_test, griden.predict(X_test_scaled)))\n",
    "print(\"Elastic Net best estimator coef:\", griden.best_estimator_.coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "scrolled": true
   },
   "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",
       "      <td>2257</td>\n",
       "      <td>6.1</td>\n",
       "      <td>7.041470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3519</td>\n",
       "      <td>6.5</td>\n",
       "      <td>6.779032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3488</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.797773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1635</td>\n",
       "      <td>5.4</td>\n",
       "      <td>5.977164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1073</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.116449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3782</td>\n",
       "      <td>6.9</td>\n",
       "      <td>7.127197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2081</td>\n",
       "      <td>7.3</td>\n",
       "      <td>7.121974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2372</td>\n",
       "      <td>7.4</td>\n",
       "      <td>6.933579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2753</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.466865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2713</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.170840</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1156 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Actual  Predicted\n",
       "2257     6.1   7.041470\n",
       "3519     6.5   6.779032\n",
       "3488     8.0   6.797773\n",
       "1635     5.4   5.977164\n",
       "1073     7.0   7.116449\n",
       "...      ...        ...\n",
       "3782     6.9   7.127197\n",
       "2081     7.3   7.121974\n",
       "2372     7.4   6.933579\n",
       "2753     6.1   6.466865\n",
       "2713     6.1   6.170840\n",
       "\n",
       "[1156 rows x 2 columns]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df19 = pd.DataFrame({'Actual': y_test, 'Predicted': y_pred})\n",
    "df19"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Connection with the director rank and profitability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a17b2a390>"
      ]
     },
     "execution_count": 64,
     "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": [
    "sns.regplot(movie_df['director_rank'], movie_df['imdb_score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "new = movie_df.query(\"gross_budget_ratio <= 100\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1a508350>"
      ]
     },
     "execution_count": 66,
     "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": [
    "sns.regplot(new['director_rank'],new['gross_budget_ratio'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will divide films by director ranks: 2-4, 5-7,8-10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "director_low = movie_df.query('director_rank <= 4')\n",
    "director_low = director_low.query('gross_budget_ratio <= 10')\n",
    "director_middle = movie_df.query('director_rank <= 7')\n",
    "director_middle = director_middle.query('director_rank >4')\n",
    "director_middle = director_middle.query('gross_budget_ratio <= 10')\n",
    "director_high = movie_df.query('director_rank >= 8')\n",
    "director_high = director_high.query('gross_budget_ratio <= 10')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a17fc5450>"
      ]
     },
     "execution_count": 85,
     "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": [
    "sns.regplot(director_low['director_rank'],director_low['gross_budget_ratio'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1b175fd0>"
      ]
     },
     "execution_count": 86,
     "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": [
    "sns.regplot(director_middle['director_rank'],director_middle['gross_budget_ratio'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1b25c990>"
      ]
     },
     "execution_count": 87,
     "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": [
    "sns.regplot(director_high['director_rank'],director_high['gross_budget_ratio'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "d2 = movie_df.query('director_rank == 2')\n",
    "d3 = movie_df.query('director_rank == 3')\n",
    "d4 = movie_df.query('director_rank == 4')\n",
    "d5 = movie_df.query('director_rank == 5')\n",
    "d6 = movie_df.query('director_rank == 6')\n",
    "d7 = movie_df.query('director_rank == 7')\n",
    "d8 = movie_df.query('director_rank == 8')\n",
    "d9 = movie_df.query('director_rank == 9')\n",
    "d10 = movie_df.query('director_rank == 10')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "imdb_mean2 = d2['imdb_score'].mean()\n",
    "imdb_mean3 = d3['imdb_score'].mean()\n",
    "imdb_mean4 = d4['imdb_score'].mean()\n",
    "imdb_mean5 = d5['imdb_score'].mean()\n",
    "imdb_mean6 = d6['imdb_score'].mean()\n",
    "imdb_mean7 = d7['imdb_score'].mean()\n",
    "imdb_mean8 = d8['imdb_score'].mean()\n",
    "imdb_mean9 = d9['imdb_score'].mean()\n",
    "imdb_mean10 = d10['imdb_score'].mean()\n",
    "\n",
    "imdb_mean =[imdb_mean2,imdb_mean3,imdb_mean4,imdb_mean5,imdb_mean6,imdb_mean7,imdb_mean8,imdb_mean9,imdb_mean10]\n",
    "\n",
    "gb_mean2 = d2['gross_budget_ratio'].mean()\n",
    "gb_mean3 = d3['gross_budget_ratio'].mean()\n",
    "gb_mean4 = d4['gross_budget_ratio'].mean()\n",
    "gb_mean5 = d5['gross_budget_ratio'].mean()\n",
    "gb_mean6 = d6['gross_budget_ratio'].mean()\n",
    "gb_mean7 = d7['gross_budget_ratio'].mean()\n",
    "gb_mean8 = d8['gross_budget_ratio'].mean()\n",
    "gb_mean9 = d9['gross_budget_ratio'].mean()\n",
    "gb_mean10 = d10['gross_budget_ratio'].mean()\n",
    "\n",
    "gb_mean = [gb_mean2,gb_mean3,gb_mean4,gb_mean5,gb_mean6,gb_mean7,gb_mean8,gb_mean9,gb_mean10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1b34b7d0>"
      ]
     },
     "execution_count": 94,
     "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": [
    "sns.regplot(imdb_mean,gb_mean)"
   ]
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}