DSL-CODE.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1689,
     "status": "ok",
     "timestamp": 1626097353304,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "YcgXFI2mhKF1",
    "outputId": "6fbce2d5-c40b-4021-bf19-54fee1cdef77"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dev set size 70000, eval set size 30000\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "df_dev=pd.read_csv(\"development.tsv\",sep='\\t')\n",
    "df_eval=pd.read_csv(\"evaluation.tsv\",sep='\\t')\n",
    "df = pd.concat([df_dev, df_eval], sort=False)\n",
    "print(f\"dev set size {df_dev.shape[0]}, eval set size {df_eval.shape[0]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 11,
     "status": "ok",
     "timestamp": 1626097353305,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "1ARPGIivhKF3",
    "outputId": "52bda3c9-3f5d-40d8-fac7-dad589c0a6c4",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 70000 entries, 0 to 69999\n",
      "Data columns (total 14 columns):\n",
      " #   Column             Non-Null Count  Dtype  \n",
      "---  ------             --------------  -----  \n",
      " 0   beer/ABV           66893 non-null  float64\n",
      " 1   beer/name          70000 non-null  object \n",
      " 2   beer/style         70000 non-null  object \n",
      " 3   review/appearance  70000 non-null  float64\n",
      " 4   review/aroma       70000 non-null  float64\n",
      " 5   review/overall     70000 non-null  float64\n",
      " 6   review/palate      70000 non-null  float64\n",
      " 7   review/taste       70000 non-null  float64\n",
      " 8   review/text        69982 non-null  object \n",
      " 9   user/ageInSeconds  14645 non-null  float64\n",
      " 10  user/birthdayRaw   14645 non-null  object \n",
      " 11  user/birthdayUnix  14645 non-null  float64\n",
      " 12  user/gender        28181 non-null  object \n",
      " 13  user/profileName   69986 non-null  object \n",
      "dtypes: float64(8), object(6)\n",
      "memory usage: 7.5+ MB\n"
     ]
    }
   ],
   "source": [
    "df_dev.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 296
    },
    "executionInfo": {
     "elapsed": 10,
     "status": "ok",
     "timestamp": 1626097353305,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "n5e4z9sVhKF4",
    "outputId": "b910c7b9-4955-409d-b1e9-0044aecab2da",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='review/overall', ylabel='count'>"
      ]
     },
     "execution_count": 45,
     "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.countplot(x='review/overall',data=df_dev,color='darkred')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 7,
     "status": "ok",
     "timestamp": 1626097353306,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "gZdVGU-fhKF4",
    "outputId": "4606e0e5-ef4d-442f-a922-bde9bd3c65f5",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The skewness of target variable is -1.0235674030883761 and the kutosis is 1.6288439531922938\n"
     ]
    }
   ],
   "source": [
    "print(f\"The skewness of target variable is {df_dev['review/overall'].skew()} and the kutosis is {df_dev['review/overall'].kurtosis()}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 296
    },
    "executionInfo": {
     "elapsed": 11,
     "status": "ok",
     "timestamp": 1626097353825,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "ZX1VSX6chKF4",
    "outputId": "ec7a3d8a-722c-4427-f052-d05073329c1b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='user/ageInYear'>"
      ]
     },
     "execution_count": 47,
     "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": [
    "converter=31556952\n",
    "df_dev['user/ageInYear']=df_dev['user/ageInSeconds']/converter\n",
    "sns.boxplot(x='user/ageInYear',data=df_dev)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "executionInfo": {
     "elapsed": 11,
     "status": "ok",
     "timestamp": 1626097353827,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "J2B2WMEshKF5"
   },
   "outputs": [],
   "source": [
    "df_dev.drop(columns='user/ageInYear',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 296
    },
    "executionInfo": {
     "elapsed": 11,
     "status": "ok",
     "timestamp": 1626097353828,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "84bX4B74hKF5",
    "outputId": "31265c87-3be9-4cfc-878a-4a7cd6575e16"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='beer/ABV'>"
      ]
     },
     "execution_count": 49,
     "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.boxplot(x='beer/ABV',data=df_dev)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5MkQ1rJzhKF5"
   },
   "source": [
    "Some beers have different ABV and may even have missing values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 465
    },
    "executionInfo": {
     "elapsed": 4,
     "status": "ok",
     "timestamp": 1626097354284,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "5oxGdfbChKF6",
    "outputId": "8a944f72-9c6e-4831-ac0a-c87d006ce6d5",
    "scrolled": false
   },
   "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>beer/ABV</th>\n",
       "      <th>beer/name</th>\n",
       "      <th>beer/style</th>\n",
       "      <th>review/appearance</th>\n",
       "      <th>review/aroma</th>\n",
       "      <th>review/overall</th>\n",
       "      <th>review/palate</th>\n",
       "      <th>review/taste</th>\n",
       "      <th>review/text</th>\n",
       "      <th>user/ageInSeconds</th>\n",
       "      <th>user/birthdayRaw</th>\n",
       "      <th>user/birthdayUnix</th>\n",
       "      <th>user/gender</th>\n",
       "      <th>user/profileName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Witbier</td>\n",
       "      <td>Witbier</td>\n",
       "      <td>3.5</td>\n",
       "      <td>2.5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>From the notes on the menu: \"A Belgian style W...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>HeroJH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6196</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Witbier</td>\n",
       "      <td>Witbier</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.0</td>\n",
       "      <td>One first pour you are just blown away by the ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Ave</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33865</th>\n",
       "      <td>4.6</td>\n",
       "      <td>Witbier</td>\n",
       "      <td>Witbier</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.5</td>\n",
       "      <td>3.5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Aroma: A nice Belgian yeast nose with the oran...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>HardTarget</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50578</th>\n",
       "      <td>4.0</td>\n",
       "      <td>Witbier</td>\n",
       "      <td>Witbier</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>This beer came a washed out gold that looked m...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>DarkerTheBetter</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       beer/ABV beer/name beer/style  review/appearance  review/aroma  \\\n",
       "97          NaN   Witbier    Witbier                3.5           2.5   \n",
       "6196        NaN   Witbier    Witbier                4.0           2.5   \n",
       "33865       4.6   Witbier    Witbier                4.5           4.5   \n",
       "50578       4.0   Witbier    Witbier                2.0           4.0   \n",
       "\n",
       "       review/overall  review/palate  review/taste  \\\n",
       "97                3.0            4.0           3.0   \n",
       "6196              5.0            4.5           4.0   \n",
       "33865             3.5            3.0           3.0   \n",
       "50578             4.0            4.0           3.5   \n",
       "\n",
       "                                             review/text  user/ageInSeconds  \\\n",
       "97     From the notes on the menu: \"A Belgian style W...                NaN   \n",
       "6196   One first pour you are just blown away by the ...                NaN   \n",
       "33865  Aroma: A nice Belgian yeast nose with the oran...                NaN   \n",
       "50578  This beer came a washed out gold that looked m...                NaN   \n",
       "\n",
       "      user/birthdayRaw  user/birthdayUnix user/gender user/profileName  \n",
       "97                 NaN                NaN         NaN           HeroJH  \n",
       "6196               NaN                NaN         NaN              Ave  \n",
       "33865              NaN                NaN         NaN       HardTarget  \n",
       "50578              NaN                NaN         NaN  DarkerTheBetter  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dev[df_dev['beer/name']=='Witbier']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 168
    },
    "executionInfo": {
     "elapsed": 4,
     "status": "ok",
     "timestamp": 1626097354727,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "lGXX9vdThKF6",
    "outputId": "e3572f7a-c1e1-4bea-8438-7d2e74ce5660"
   },
   "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>beer/ABV</th>\n",
       "      <th>beer/name</th>\n",
       "      <th>beer/style</th>\n",
       "      <th>review/appearance</th>\n",
       "      <th>review/aroma</th>\n",
       "      <th>review/overall</th>\n",
       "      <th>review/palate</th>\n",
       "      <th>review/taste</th>\n",
       "      <th>review/text</th>\n",
       "      <th>user/ageInSeconds</th>\n",
       "      <th>user/birthdayRaw</th>\n",
       "      <th>user/birthdayUnix</th>\n",
       "      <th>user/gender</th>\n",
       "      <th>user/profileName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>577</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Dark Marc</td>\n",
       "      <td>American Black Ale</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>On draught at the Dog and Pony, Dark Marc is b...</td>\n",
       "      <td>1.137507e+09</td>\n",
       "      <td>Nov 25, 1978</td>\n",
       "      <td>280828800.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>ccrida</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     beer/ABV  beer/name          beer/style  review/appearance  review/aroma  \\\n",
       "577       NaN  Dark Marc  American Black Ale                4.0           4.0   \n",
       "\n",
       "     review/overall  review/palate  review/taste  \\\n",
       "577             4.0            4.0           4.0   \n",
       "\n",
       "                                           review/text  user/ageInSeconds  \\\n",
       "577  On draught at the Dog and Pony, Dark Marc is b...       1.137507e+09   \n",
       "\n",
       "    user/birthdayRaw  user/birthdayUnix user/gender user/profileName  \n",
       "577     Nov 25, 1978        280828800.0        Male           ccrida  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dev[df_dev['beer/name']=='Dark Marc']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "executionInfo": {
     "elapsed": 134624,
     "status": "ok",
     "timestamp": 1626097489348,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "sYYeJQXBhKF6"
   },
   "outputs": [],
   "source": [
    "#Filling missing values for beer/ABV\n",
    "for i in df.loc[df['beer/ABV'].isna(),'beer/name']:\n",
    "    if df.loc[df['beer/name']==i,'beer/ABV'].isna().sum()==df[df['beer/name']==i].shape[0]:\n",
    "        df.loc[df['beer/name']==i,'beer/ABV']=df.loc[df['beer/name']==i,'beer/ABV'].fillna((df['beer/ABV'].mean()))\n",
    "    else:\n",
    "        df.loc[(df['beer/name']==i)&(df['beer/ABV'].isna()),'beer/ABV']=\\\n",
    "        df.loc[(df['beer/name']==i)&(df['beer/ABV'].isna()),'beer/ABV']\\\n",
    "        .fillna((df.loc[df['beer/name']==i,'beer/ABV'].mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 282
    },
    "executionInfo": {
     "elapsed": 11,
     "status": "ok",
     "timestamp": 1626097489349,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "Sk1AB9bHhKF6",
    "outputId": "3409814b-981b-4fd5-b21d-56d2c1679799",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 53,
     "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.boxplot(data=df_dev[['review/appearance','review/aroma','review/palate','review/taste']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 334
    },
    "executionInfo": {
     "elapsed": 521,
     "status": "ok",
     "timestamp": 1626097489860,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "aWlPg9KvhKF7",
    "outputId": "6c3e2c9d-fb78-42f5-9f18-43aab312c80e",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_dev['user/gender'].fillna(\"unknown\",inplace=True)\n",
    "plt.figure(figsize=(6,5))\n",
    "ax = sns.countplot(x=\"user/gender\", data=df_dev, order=['unknown','Male','Female'])\n",
    "\n",
    "\n",
    "for p in ax.patches:\n",
    "        ax.annotate('{}'.format(p.get_height()), (p.get_x()+0.1, p.get_height()+50))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 5,
     "status": "ok",
     "timestamp": 1626097489861,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "yFjagJYdhKF7",
    "outputId": "1d15f283-f9a0-4ee7-de57-944e77d9b7b3",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "beer/name           14770\n",
       "beer/style            104\n",
       "user/gender             3\n",
       "user/profileName    10573\n",
       "dtype: int64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dev[['beer/name','beer/style','user/gender','user/profileName']].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "executionInfo": {
     "elapsed": 13037,
     "status": "ok",
     "timestamp": 1626097502895,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "OCNZGADPhKF7"
   },
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "vect =TfidfVectorizer (token_pattern=r'\\w{2,}',stop_words=\"english\",use_idf=True,max_df=0.8,min_df=0.01)\n",
    "wpm=vect.fit_transform(df[\"review/text\"].fillna(\"\"))\n",
    "freq = sorted(zip(vect.get_feature_names(), wpm.sum(axis=0).tolist()[0]), key=lambda x: x[1], reverse=True)\n",
    "words = [ word for word, _ in freq ]\n",
    "mask = [ w in words for w in vect.get_feature_names() ]\n",
    "words_ = [ w for w in vect.get_feature_names() if w in words ]\n",
    "words_df = pd.DataFrame(data=wpm[:, np.array(mask)].toarray(), columns=[f\"word_{word}\" for word in words_],index=df.index)\n",
    "df_new=df.drop(columns=['review/text','user/birthdayUnix','user/birthdayRaw','beer/name','user/gender'])\n",
    "df_new = pd.concat([df_new, words_df], axis=1)\n",
    "converter=31556952\n",
    "df_new['user/ageInSeconds'].fillna((df_new['user/ageInSeconds'].mean()),inplace=True)\n",
    "df_new['user/ageInSeconds']=df_new['user/ageInSeconds']/converter\n",
    "df_new['user/ageInSeconds'] = (df_new['user/ageInSeconds']+1).transform(np.log)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "841"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(freq)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 198
    },
    "executionInfo": {
     "elapsed": 53513,
     "status": "ok",
     "timestamp": 1626097556402,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "p2xMOy7mhKF8",
    "outputId": "6dd594c6-a87b-45eb-92ea-f81708f86041"
   },
   "outputs": [
    {
     "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": [
    "from wordcloud import WordCloud\n",
    "feature_names = vect.get_feature_names()\n",
    "dense = wpm.todense()\n",
    "lst1 = dense.tolist()\n",
    "wc = pd.DataFrame(lst1, columns=feature_names)\n",
    "wc.T.sum(axis=1)\n",
    "Cloud = WordCloud(background_color=\"white\", max_words=50).generate_from_frequencies(wc.T.sum(axis=1))\n",
    "plt.figure()\n",
    "plt.imshow(Cloud, interpolation=\"bilinear\")\n",
    "plt.axis(\"off\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "executionInfo": {
     "elapsed": 1143,
     "status": "ok",
     "timestamp": 1626097557537,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "QYJCo8dZhKF8"
   },
   "outputs": [],
   "source": [
    "df_new.loc[df_new['user/profileName'].isin(list(df_new['user/profileName'].value_counts()[671:].index)),'user/profileName']='other'\n",
    "df_new = pd.get_dummies(df_new, columns=['user/profileName'])\n",
    "df_new.loc[df_new['beer/style'].isin(list(df_new['beer/style'].value_counts()[27:].index)),'beer/style']='other'\n",
    "df_new = pd.get_dummies(df_new, columns=['beer/style'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "executionInfo": {
     "elapsed": 5657,
     "status": "ok",
     "timestamp": 1626097914138,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "3_02Em5ahKF8"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.experimental import enable_hist_gradient_boosting\n",
    "from sklearn.ensemble import RandomForestRegressor,HistGradientBoostingRegressor\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.linear_model import Ridge,Lasso,LinearRegression\n",
    "\n",
    "train_valid_mask = ~df_new[\"review/overall\"].isna()\n",
    "feature_names = df_new[train_valid_mask].drop(columns=[\"review/overall\"]).columns\n",
    "X = df_new.drop(columns=[\"review/overall\"]).values\n",
    "y = df_new[\"review/overall\"].values\n",
    "X_train_valid = X[train_valid_mask]\n",
    "y_train_valid = y[train_valid_mask]\n",
    "X_eval = X[~train_valid_mask]\n",
    "y_eval = y[~train_valid_mask]\n",
    "train_ratio = 0.75\n",
    "validation_ratio = 0.15\n",
    "test_ratio = 0.10\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_train_valid, y_train_valid, shuffle=True, random_state=42,test_size=1-train_ratio)\n",
    "X_valid, X_test, y_valid, y_test = train_test_split(X_test, y_test, shuffle=True, random_state=42,test_size=test_ratio/(test_ratio + validation_ratio))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tuning hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 6691095,
     "status": "ok",
     "timestamp": 1626104690720,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "O_zibXmWhKGB",
    "outputId": "8f2f7c8c-dae7-4b30-b9df-ba3074b7cf10"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6879010003639854"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svr1=SVR(C=0.1)\n",
    "svr1.fit(X_train, y_train)\n",
    "r2_score(y_valid, svr1.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 7068991,
     "status": "ok",
     "timestamp": 1626111759705,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "hc2wKsqPhKGB",
    "outputId": "99b3fb12-c59a-4559-e64b-0d911c70cf5c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7055615048904049"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svr2=SVR(C=1)\n",
    "svr2.fit(X_train, y_train)\n",
    "r2_score(y_valid, svr2.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "bemFxBh9hKGB"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7098730015036638"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svr3=SVR(C=10)\n",
    "svr3.fit(X_train, y_train)\n",
    "r2_score(y_valid, svr3.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "id": "zHhfclyoyoKV"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6990734440164093"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr1=LinearRegression(fit_intercept=True)\n",
    "lr1.fit(X_train, y_train)\n",
    "r2_score(y_valid, lr1.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "id": "VRAGNGl6hKGC"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6990637949806444"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr2=LinearRegression(fit_intercept=False)\n",
    "lr2.fit(X_train, y_train)\n",
    "r2_score(y_valid, lr2.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "bMMgsvcKhKGC"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6856542441810598"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfr1=RandomForestRegressor(n_estimators=250,max_depth=None,n_jobs=-1)\n",
    "rfr1.fit(X_train, y_train)\n",
    "r2_score(y_valid, rfr1.predict(X_valid))\n",
    "#0.6842776354754101"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "jJeRTaPUhKGC"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6867483210451666"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfr2=RandomForestRegressor(n_estimators=500,max_depth=None,n_jobs=-1)\n",
    "rfr2.fit(X_train, y_train)\n",
    "r2_score(y_valid, rfr2.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "id": "E4Nc1_YbhKGC"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.684081528458959"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfr3=RandomForestRegressor(n_estimators=100,max_depth=None,n_jobs=-1)\n",
    "rfr3.fit(X_train, y_train)\n",
    "r2_score(y_valid, rfr3.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6833680980093217"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfr4=RandomForestRegressor(n_estimators=100,max_depth=50,n_jobs=-1)\n",
    "rfr4.fit(X_train, y_train)\n",
    "r2_score(y_valid, rfr4.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.686670962392851"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfr5=RandomForestRegressor(n_estimators=250,max_depth=50,n_jobs=-1)\n",
    "rfr5.fit(X_train, y_train)\n",
    "r2_score(y_valid, rfr5.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6866039865930839"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfr6=RandomForestRegressor(n_estimators=500,max_depth=50,n_jobs=-1)\n",
    "rfr6.fit(X_train, y_train)\n",
    "r2_score(y_valid, rfr6.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6999686089459136"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r1=Ridge(alpha=1)\n",
    "r1.fit(X_train, y_train)\n",
    "r2_score(y_valid, r1.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7011181707571669"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r2=Ridge(alpha=5)\n",
    "r2.fit(X_train, y_train)\n",
    "r2_score(y_valid, r2.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7018964938276763"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r3=Ridge(alpha=15)\n",
    "r3.fit(X_train, y_train)\n",
    "r2_score(y_valid, r3.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6807140695740925"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l1=Lasso(alpha=0.001)\n",
    "l1.fit(X_train, y_train)\n",
    "r2_score(y_valid, l1.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6986780239038239"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l2=Lasso(alpha=0.0001)\n",
    "l2.fit(X_train, y_train)\n",
    "r2_score(y_valid, l2.predict(X_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7013428294131359"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l3=Lasso(alpha=0.00001)\n",
    "l3.fit(X_train, y_train)\n",
    "r2_score(y_valid, l3.predict(X_valid))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Model Selection of best configurations of the models on the test portion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_new=np.concatenate((X_train,X_valid))\n",
    "y_train_new=np.concatenate((y_train,y_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.716338340823122"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_svr=SVR(C=10)\n",
    "best_svr.fit(X_train_new,y_train_new)\n",
    "r2_score(y_test,best_svr.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7052023016660179"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_lr=LinearRegression(fit_intercept=True)\n",
    "best_lr.fit(X_train_new,y_train_new)\n",
    "r2_score(y_test,best_lr.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6978548678379191"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_rfr=RandomForestRegressor(n_estimators=500,max_depth=None,n_jobs=-1)\n",
    "best_rfr.fit(X_train_new,y_train_new)\n",
    "r2_score(y_test,best_rfr.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.708404941799504"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_r=Ridge(alpha=15)\n",
    "best_r.fit(X_train_new,y_train_new)\n",
    "r2_score(y_test,best_r.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7071211823793"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_l=Lasso(alpha=0.00001)\n",
    "best_l.fit(X_train_new,y_train_new)\n",
    "r2_score(y_test,best_l.predict(X_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Final Model (SVR) on evaluation set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Uw-bKYDxhKGC"
   },
   "outputs": [],
   "source": [
    "final_model=SVR(C=10)\n",
    "final_model.fit(X_train_valid, y_train_valid)\n",
    "y_pred1=final_model.predict(X_eval)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "executionInfo": {
     "elapsed": 5,
     "status": "ok",
     "timestamp": 1626094414141,
     "user": {
      "displayName": "ALMOUDI MOSTAFA",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Ggag6Su7xYEbeg02b3KD79PPONoKQ9i52rxvDSbYw=s64",
      "userId": "02569998257955289351"
     },
     "user_tz": -180
    },
    "id": "i4wFvLEYhKGC"
   },
   "outputs": [],
   "source": [
    "pd.DataFrame(y_pred3, index=df_new[~train_valid_mask].index).to_csv(\"outputforest.csv\",index_label=\"Id\", header=[\"Predicted\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "l=sorted(zip(feature_names, best_rfr.feature_importances_), key=lambda x: x[1],reverse=True)[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "top_features=list(map(lambda x:x[0],l))\n",
    "importance=list(map(lambda x:x[1],l))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<BarContainer object of 5 artists>"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeMAAAEwCAYAAACE3Rm5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAXxElEQVR4nO3de7QdZ33e8e+DZCfcXC5SCsgWUom4iCwwIAQUQ8ytsaFBeGFiGygxmKpOY2ia0kRtWkpLk2DIam6YCoW6kEKilLsKwkoTMBhfgmQjbEvGRDGuLZzGxiWAqYux+fWPmYNGx/vobMv7+JX3+X7W2kvvzLwz855377OfmXfmjFJVSJKkdh7QugGSJC12hrEkSY0ZxpIkNWYYS5LUmGEsSVJjhrEkSY0tbbXjZcuW1apVq1rtXpKk+9zll1/+zapaPnt+szBetWoVu3btarV7SZLuc0n+16j5DlNLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjTV7NvWkrdr06dZNOKJc/46XtW6CJGlMnhlLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1NlYYJzkpybVJ9iXZNEedE5PsTrInyecn20xJkqbXvE/gSrIEOA94CbAf2JlkW1XtHdR5GPAe4KSquiHJTyxQeyVJmjrjnBmvB/ZV1XVVdQewFdgwq86rgY9V1Q0AVXXzZJspSdL0GieMVwA3Dqb39/OGHg88PMmFSS5P8rpJNVCSpGk3zn8UkRHzasR2ngG8CHggcGmSy6rqawdtKNkIbARYuXLlPW+tJElTaJwz4/3AcYPpY4GbRtS5oKq+V1XfBL4APHX2hqpqS1Wtq6p1y5cvP9w2S5I0VcYJ453AmiSrkxwNnA5sm1Xnk8DzkixN8iDgWcA1k22qJEnTad5h6qq6M8k5wA5gCXB+Ve1Jcna/fHNVXZPkAuBK4IfA+6rq6oVsuCRJ02Kca8ZU1XZg+6x5m2dNvwt41+SaJknS4uATuCRJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhobK4yTnJTk2iT7kmwasfzEJN9Osrt/vXXyTZUkaTotna9CkiXAecBLgP3AziTbqmrvrKoXVdU/XIA2SpI01cY5M14P7Kuq66rqDmArsGFhmyVJ0uIxThivAG4cTO/v5832nCRfSfKZJE+eSOskSVoE5h2mBjJiXs2avgJ4bFXdluSlwCeANXfbULIR2AiwcuXKe9ZSSZKm1DhnxvuB4wbTxwI3DStU1Xeq6ra+vB04Ksmy2Ruqqi1Vta6q1i1fvvxeNFuSpOkxThjvBNYkWZ3kaOB0YNuwQpJHJUlfXt9v99ZJN1aSpGk07zB1Vd2Z5BxgB7AEOL+q9iQ5u1++GTgV+IUkdwK3A6dX1eyhbEmSNMI414xnhp63z5q3eVB+N/DuyTZNkqTFwSdwSZLUmGEsSVJjhrEkSY0ZxpIkNWYYS5LUmGEsSVJjhrEkSY0ZxpIkNWYYS5LUmGEsSVJjhrEkSY0ZxpIkNWYYS5LUmGEsSVJjhrEkSY0ZxpIkNWYYS5LUmGEsSVJjhrEkSY0ZxpIkNWYYS5LUmGEsSVJjhrEkSY0ZxpIkNWYYS5LUmGEsSVJjhrEkSY0ZxpIkNWYYS5LUmGEsSVJjhrEkSY0ZxpIkNTZWGCc5Kcm1SfYl2XSIes9McleSUyfXREmSptu8YZxkCXAecDKwFjgjydo56p0L7Jh0IyVJmmbjnBmvB/ZV1XVVdQewFdgwot6bgI8CN0+wfZIkTb1xwngFcONgen8/70eSrABOATZPrmmSJC0O44RxRsyrWdO/A/xqVd11yA0lG5PsSrLrlltuGbOJkiRNt6Vj1NkPHDeYPha4aVaddcDWJADLgJcmubOqPjGsVFVbgC0A69atmx3okiQtSuOE8U5gTZLVwDeA04FXDytU1eqZcpL3A5+aHcSSJGm0ecO4qu5Mcg7dXdJLgPOrak+Ss/vlXieWJOleGOfMmKraDmyfNW9kCFfVmfe+WZIkLR4+gUuSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKmxscI4yUlJrk2yL8mmEcs3JLkyye4ku5KcMPmmSpI0nZbOVyHJEuA84CXAfmBnkm1VtXdQ7c+BbVVVSZ4C/HfgiQvRYEmSps04Z8brgX1VdV1V3QFsBTYMK1TVbVVV/eSDgUKSJI1lnDBeAdw4mN7fzztIklOSfBX4NPCGyTRPkqTpN04YZ8S8u535VtXHq+qJwCuAt4/cULKxv6a865ZbbrlHDZUkaVqNE8b7geMG08cCN81Vuaq+ADwuybIRy7ZU1bqqWrd8+fJ73FhJkqbROGG8E1iTZHWSo4HTgW3DCkl+Mkn68tOBo4FbJ91YSZKm0bx3U1fVnUnOAXYAS4Dzq2pPkrP75ZuBVwKvS/ID4HbgtMENXZIk6RDmDWOAqtoObJ81b/OgfC5w7mSbJknS4uATuCRJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhozjCVJaswwliSpMcNYkqTGDGNJkhobK4yTnJTk2iT7kmwasfw1Sa7sX5ckeerkmypJ0nSaN4yTLAHOA04G1gJnJFk7q9rXgZ+uqqcAbwe2TLqhkiRNq3HOjNcD+6rquqq6A9gKbBhWqKpLqupb/eRlwLGTbaYkSdNrnDBeAdw4mN7fz5vLWcBn7k2jJElaTJaOUScj5tXIiskL6ML4hDmWbwQ2AqxcuXLMJkqSNN3GOTPeDxw3mD4WuGl2pSRPAd4HbKiqW0dtqKq2VNW6qlq3fPnyw2mvJElTZ5ww3gmsSbI6ydHA6cC2YYUkK4GPAf+oqr42+WZKkjS95h2mrqo7k5wD7ACWAOdX1Z4kZ/fLNwNvBR4JvCcJwJ1VtW7hmi1J0vQY55oxVbUd2D5r3uZB+Y3AGyfbNEmSFgefwCVJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktSYYSxJUmOGsSRJjRnGkiQ1ZhhLktTYWGGc5KQk1ybZl2TTiOVPTHJpku8necvkmylJ0vRaOl+FJEuA84CXAPuBnUm2VdXeQbX/A7wZeMVCNFKSpGk2zpnxemBfVV1XVXcAW4ENwwpVdXNV7QR+sABtlCRpqo0TxiuAGwfT+/t5kiRpAsYJ44yYV4ezsyQbk+xKsuuWW245nE1IkjR1xgnj/cBxg+ljgZsOZ2dVtaWq1lXVuuXLlx/OJiRJmjrjhPFOYE2S1UmOBk4Hti1ssyRJWjzmvZu6qu5Mcg6wA1gCnF9Ve5Kc3S/fnORRwC7gGOCHSX4JWFtV31m4pkuSNB3mDWOAqtoObJ81b/Og/L/phq8lSdI95BO4JElqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKkxw1iSpMYMY0mSGjOMJUlqzDCWJKmxpa0boCPXqk2fbt2EI8r173hZ6yZImlKeGUuS1JhhLElSY4axJEmNGcaSJDVmGEuS1JhhLElSY4axJEmNGcaSJDXmQz+k+5APUjnAh6hIB4x1ZpzkpCTXJtmXZNOI5Unye/3yK5M8ffJNlSRpOs0bxkmWAOcBJwNrgTOSrJ1V7WRgTf/aCPznCbdTkqSpNc4w9XpgX1VdB5BkK7AB2DuoswH4w6oq4LIkD0vy6Kr664m3WJJ6DvsfbBJD//bpAfflpZRxhqlXADcOpvf38+5pHUmSNMI4Z8YZMa8Oow5JNtINYwPcluTaMfZ/f7MM+GbrRuTc1i2YKPt0suzPybNPJ2ua+/Oxo2aOE8b7geMG08cCNx1GHapqC7BljH3ebyXZVVXrWrdjmtink2V/Tp59OlmLsT/HGabeCaxJsjrJ0cDpwLZZdbYBr+vvqn428G2vF0uSNJ55z4yr6s4k5wA7gCXA+VW1J8nZ/fLNwHbgpcA+4P8Cr1+4JkuSNF3GeuhHVW2nC9zhvM2DcgG/ONmm3W9N9TB8I/bpZNmfk2efTtai6890OSpJklrx2dSSJDVmGI+Q5DFJPjKhbb03yXOTnJnkMYe5jROT/P1JtOe+tBD9eBjrnZjkU/PUOT7JSw+/dQsjyaokVy/Qto9Kcvlg+pQkleSJs/Z/e5LdSb6S5JIkT+j79NJZ21ua5G+SPHoh2itNu0URxv1d3mP/rFV1U1WdOqHdPwu4DDgTOKwwBk4EmofxEdKPC+F4uhsQp16SmftETgAuGSw6A/gi3V9LDP1VVR1fVU8FPgD8a+ALwLFJVg3qvRi4ejH8FUV/YP3u1u04kt3TPkpy2yGWXTLXskGd65MsG3Nfcx7kJrkwSZM/qZraMO47/Jok7wGuAP5tkp39f2Tx7/s65yb5p4N13pbkXwzfrCRLkrxrsO4/6ee/J8nL+/LHk5zfl89K8h/78pOArwGnAOuAD/VnGQ9M8tZ+m1cn2ZIk/TpvTrK339fW/gvvbOCf9+s+L8nyJB/t1995OGeM98d+rKq7+l+W3+nP0q5Osr6vs76f9+WZM7gRP8vd6qT7c73/AJzW9+9pSR6c5Py+rV9OsmGh+ncMS5N8oO+zjyR5UJJnJPl8ksuT7Jg5G03yuCQX9PMvSn+Wm+T9Sf5Tks8BM48xOAn4TL/8IcBzgbO4exgPHQN8q6p+CHwYOG2w7HTgjyf5gx8p0j2fX4ewEH00s82qan4icp+oqql8AauAHwLPBv4B3d15oTsA+RTwfOBpwOcH6+wFVvbrXt3P2wj8m778Y8AuYDXdl8+7+vlfAi7ry/8V+Jm+/MvAG/ryhcC6wb4eMSj/N+Bn+/JNwI/15Yf1/74NeMug/h8BJ/TllcA1i6wf/6AvP3+w/WOApX35xcBH+/KJwKfmqXMm8O5B+38DeO3Me0B3QPXgRp/hAp7bT58P/Eu6M9rl/bzT6P7cEODPgTV9+VnAZ/vy+/v3aslg218CHtSXXwv8l758CfD0wf5vB3YDfwX8NbCyX/ZM4MuD9/Nm4OEtftfn6cNfAd7cl3970CcvAj5INyJwFXA1cO5gvdvoDtL+gm4U4fX95+DzwB8MPy8j9rkc+CjdMxp2Dt6/9X3/frn/9wn9/Cf378du4Eq6/3Dn7cA/G2zz12d+jinpo9XApX3/vB24rQ78vn6O7jtu78x+BssuBD4CfBX4EAduQr6e7qldDwQuAP4x8BC634kr+vZvGHyuv0o30nNlv72Z34UL6b+n6b7vLu3X/zDwkAX9rLb+ZVmwH6zr8K/35d/q36zd/WsfcFa/7Bq64eOnAhcP1p35kv9I/wGbWffr/Zu0gm7YdC3dl90ngUf3b/JD+3V3AI+Z/Sb306/sP8RXAd8ANvXzL+j3+dqZN5+7h/HNg/bs7td/6CLqxxcO2ncDXWAeB3yc7gvjKuCrg1/gmTCeq86ZHBzGu/o6M229AXhSo8/wDYPpFwJ/Bnxn0LargD+l++K5nYM/F9f0670f+PnBdh4DXDCY/jTwkr78Zg4cHP3o/eunT5u13l8CTwBeAWxr/Ts/Rx8+G/hwX76ILvSOAv5d/7qBLjyXAp8FXtHXLeDn+vKjB/WOBi7m0EEz8mCZuQ8Gfx94TV8+mi5QVgFX9PMeQHcw9Mgp6qNtwOv68i9ycOB+D1g9qDtc9m26Jzw+gC4oZ/r5+r7P/myw3aXAMX15Gd33VRh9kPuWvnwh3SjmMrrLMQ/u5/8q8NaF/KyO9XfG92Pf6/8N8JtV9d4RdT4CnAo8Ctg6YnmAN1XVjrstSB5ON9z3BeARwM/RfXC+m+RBdGe2d3ssaJIfB95DF843Jnkb8OP94pfRnfG9nG5I+Mkj2vQA4DlVdfvoH3vijrR+rFmbKLqj689V1Sn90P6FI9owTp2Ztr6yqo6EZ6fP/lm/C+ypqucMZyY5Bvjbqjp+ju18b1A+me4AhySPpAv5n0pSdA/2qSS/MmIb2+hGLGZspRvZeBJH7hD15cAzkjwU+D7dWc464HnA/wAurKpbAJJ8iO537xPAXXRnt9CNMgzr/Qnw+EPs88XA2v7KE8Ax/f7/DvCBJGvo3tej+uWXAr+W5FjgY1X1l8D1SW5N8jTg79KNQtx6r3pibi366Ll0JyTQjQwOnwL9par6+hzrfamq9vf72E0XrF/sl30SeGdVfaifDvAbSZ5PN7q3gq4vAW6sqov78gfpDkJ/a7CfZ9OdIFzcv49H071PC2ZqrxnPsgN4Q39tjCQrkvxEv2zmC+VUukAZte4vJDmqX/fxSR7cL7sU+CW6ELkIeEv/L8AL6IZbZnwXeGhfngneb/ZtOrXf9gOA46rqc3RDRw+jO+MZrgvdmdA5MxNJjh+jDybhSOhH6K9VJjmB7tGr36b7ovtGv/zMOdo/V53Z/bsDeFPyo+v4T5tje/eFlUlmgvcMulGE5TPz0t0V/eSq+g7w9SSv6ucnyVPn2OaPrhfTvV9/WFWPrapVVXUc3ajFCSPWO4HuDG3GH9ON4LyQuz8i94hQVT+gO2t6Pd3Q8EV0n6nH0Z3JzeX/VdVdw03dg93OHCwf379WVNV3OXAw+FPAz9J/D1TVH9EdfN8O7Ejywn4776P7nL6e7uxtQTTqo0PV/94c86E7WJhxFwc/uOpi4OSZ31vgNXRn6s/oD1L/hgPfvaMO6IcC/M/Be7i2qs46RLvutUURxlX1p3RDR5cmuYouLB7aL9vTl79Ro+8EfR/dNdAr+puR3suBD8BFdMNO++iOJh/BgRA5mW7Iecb7gc390dz36a6pXEV3hLmzr7ME+GDfxi8Dv11Vf0t3dHpKf4PR8+iO4tb1N/XspbvBa8EdIf0I8K10d1huprvpCOCdwG8muZiuH0eZq87n6M5kdic5je5L8yjgyr6tbz9Etyy0a4CfT3IlXb/8Pl2AnpvkK3TD0TM3uLwGOKufv4fu/xk/SH9TzJqq+mo/6wy6ofuhjwKv7suP6/vlK3TX0t84U6mq9tI9/vazVXWoL9DWvkB3gDdzsHc2Xb9dBvx0kmV9v5xBd71ztr8ATkzyyP5g8lXz7G+ug+WRB4NJ/h5wXVX9Ht1BzVP6RR+nO3B6Jv1IxgK6r/voYg7cLPiae998AN4K3Eo36ghdf99cVT9I8gIO/t+SZh/kfpGDXQY8N8lPAvQ3Th7qTP/eW8gx8MX8oguVo1q34/7+mt2PzLr27use9+cJwObW7biPf+YXAT/gwPW/rwG/3JdfzYGbk945WOe2WdsY3pz0uxz6eugy4E/obg7aO9PfwHP6bVxMd4B3fT//X9EdPO2mO/Ac3ty5GXjHFPbR8AauTRx8XfhTs+qOXAa8GzizL1/f93voLqW8s5++lO4ekPfRHdiu6l97+769ku7gc9QNXC/s23dl/3r5Qr4HPg5T9ytJLqS72WJX67ZIC6m/bHUF8KrqriNrik37DVyaMlV1Yus2SAstyVq6P0f7uEG8OHhmLEmHIcmvcfdrox+uql9v0Z4jkX00PsNYkqTGFsXd1JIkHckMY0mSGjOMJUlqzDCWJKkxw1iSpMb+P/BlAwOjfISNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize = (8, 5))\n",
    " \n",
    "plt.bar(top_features, importance)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "machine_shape": "hm",
   "name": "fff.ipynb",
   "provenance": []
  },
  "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}