Legency (1).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "id": "NctXmLHTKtMp"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import sklearn as sk\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.neural_network import MLPRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn.metrics import mean_absolute_percentage_error as mape\n",
    "from sklearn.metrics import mean_absolute_error as mae\n",
    "from sklearn.metrics import mean_squared_error as mse\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "plt.style.use('seaborn-whitegrid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 488
    },
    "id": "x05FKJgBOHtC",
    "outputId": "1c24892b-0c67-49cf-8a84-c7dab1543d3c"
   },
   "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>total_sessions_day0</th>\n",
       "      <th>total_sessions_day1</th>\n",
       "      <th>total_sessions_day3</th>\n",
       "      <th>total_sessions_day7</th>\n",
       "      <th>chapters_finished_day0</th>\n",
       "      <th>chapters_finished_day1</th>\n",
       "      <th>chapters_finished_day3</th>\n",
       "      <th>chapters_finished_day7</th>\n",
       "      <th>chapters_opened_day0</th>\n",
       "      <th>chapters_opened_day1</th>\n",
       "      <th>...</th>\n",
       "      <th>install_date</th>\n",
       "      <th>country_code</th>\n",
       "      <th>ad_ltv_day0</th>\n",
       "      <th>ad_ltv_day1</th>\n",
       "      <th>ad_ltv_day3</th>\n",
       "      <th>platform</th>\n",
       "      <th>target_sub_ltv_day30</th>\n",
       "      <th>target_iap_ltv_day30</th>\n",
       "      <th>target_ad_ltv_day30</th>\n",
       "      <th>target_full_ltv_day30</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</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>...</td>\n",
       "      <td>2021-12-03</td>\n",
       "      <td>COUNTRY_135</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ios</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</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>...</td>\n",
       "      <td>2021-12-03</td>\n",
       "      <td>COUNTRY_141</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>android</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</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>...</td>\n",
       "      <td>2021-12-03</td>\n",
       "      <td>COUNTRY_141</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>android</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</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>...</td>\n",
       "      <td>2021-12-03</td>\n",
       "      <td>COUNTRY_141</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ios</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</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>...</td>\n",
       "      <td>2021-12-03</td>\n",
       "      <td>COUNTRY_203</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>android</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\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",
       "      <th>1467827</th>\n",
       "      <td>11.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-12-30</td>\n",
       "      <td>COUNTRY_141</td>\n",
       "      <td>0.037158</td>\n",
       "      <td>0.037158</td>\n",
       "      <td>0.037158</td>\n",
       "      <td>ios</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.174198</td>\n",
       "      <td>0.037158</td>\n",
       "      <td>4.211356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1467828</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>13</td>\n",
       "      <td>6</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-12-30</td>\n",
       "      <td>COUNTRY_141</td>\n",
       "      <td>0.039256</td>\n",
       "      <td>0.039256</td>\n",
       "      <td>0.091267</td>\n",
       "      <td>ios</td>\n",
       "      <td>4.705161</td>\n",
       "      <td>10.967580</td>\n",
       "      <td>0.091267</td>\n",
       "      <td>15.764008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1467829</th>\n",
       "      <td>3.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "      <td>15</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-12-30</td>\n",
       "      <td>COUNTRY_141</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ios</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.938221</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.938221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1467830</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>24</td>\n",
       "      <td>24</td>\n",
       "      <td>24</td>\n",
       "      <td>24</td>\n",
       "      <td>24</td>\n",
       "      <td>24</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-12-30</td>\n",
       "      <td>COUNTRY_141</td>\n",
       "      <td>0.177550</td>\n",
       "      <td>0.177550</td>\n",
       "      <td>0.177550</td>\n",
       "      <td>ios</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.402224</td>\n",
       "      <td>0.177550</td>\n",
       "      <td>7.579774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1467831</th>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>16</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-12-30</td>\n",
       "      <td>COUNTRY_141</td>\n",
       "      <td>0.114909</td>\n",
       "      <td>0.114909</td>\n",
       "      <td>0.114909</td>\n",
       "      <td>android</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.949385</td>\n",
       "      <td>0.114909</td>\n",
       "      <td>4.064295</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1467832 rows × 63 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         total_sessions_day0  total_sessions_day1  total_sessions_day3  \\\n",
       "0                        1.0                  1.0                  1.0   \n",
       "1                        1.0                  1.0                  1.0   \n",
       "2                        1.0                  2.0                  2.0   \n",
       "3                        1.0                  1.0                  1.0   \n",
       "4                        1.0                  1.0                  1.0   \n",
       "...                      ...                  ...                  ...   \n",
       "1467827                 11.0                 11.0                 11.0   \n",
       "1467828                  2.0                  3.0                  6.0   \n",
       "1467829                  3.0                  5.0                 11.0   \n",
       "1467830                  5.0                  5.0                  5.0   \n",
       "1467831                  6.0                  7.0                  7.0   \n",
       "\n",
       "         total_sessions_day7  chapters_finished_day0  chapters_finished_day1  \\\n",
       "0                        1.0                       0                       0   \n",
       "1                        1.0                       0                       0   \n",
       "2                        2.0                       0                       0   \n",
       "3                        1.0                       0                       0   \n",
       "4                        1.0                       0                       0   \n",
       "...                      ...                     ...                     ...   \n",
       "1467827                 11.0                       9                       9   \n",
       "1467828                  7.0                       5                       9   \n",
       "1467829                 13.0                       5                       9   \n",
       "1467830                  5.0                      24                      24   \n",
       "1467831                  7.0                      15                      15   \n",
       "\n",
       "         chapters_finished_day3  chapters_finished_day7  chapters_opened_day0  \\\n",
       "0                             0                       0                     0   \n",
       "1                             0                       0                     0   \n",
       "2                             0                       0                     0   \n",
       "3                             0                       0                     0   \n",
       "4                             0                       0                     0   \n",
       "...                         ...                     ...                   ...   \n",
       "1467827                       9                       9                    10   \n",
       "1467828                      11                      13                     6   \n",
       "1467829                      15                      18                     5   \n",
       "1467830                      24                      24                    24   \n",
       "1467831                      15                      15                    15   \n",
       "\n",
       "         chapters_opened_day1  ...  install_date  country_code  ad_ltv_day0  \\\n",
       "0                           0  ...    2021-12-03   COUNTRY_135     0.000000   \n",
       "1                           0  ...    2021-12-03   COUNTRY_141     0.000000   \n",
       "2                           0  ...    2021-12-03   COUNTRY_141     0.000000   \n",
       "3                           0  ...    2021-12-03   COUNTRY_141     0.000000   \n",
       "4                           0  ...    2021-12-03   COUNTRY_203     0.000000   \n",
       "...                       ...  ...           ...           ...          ...   \n",
       "1467827                    10  ...    2021-12-30   COUNTRY_141     0.037158   \n",
       "1467828                    10  ...    2021-12-30   COUNTRY_141     0.039256   \n",
       "1467829                    10  ...    2021-12-30   COUNTRY_141     0.000000   \n",
       "1467830                    24  ...    2021-12-30   COUNTRY_141     0.177550   \n",
       "1467831                    16  ...    2021-12-30   COUNTRY_141     0.114909   \n",
       "\n",
       "         ad_ltv_day1  ad_ltv_day3  platform  target_sub_ltv_day30  \\\n",
       "0           0.000000     0.000000       ios              0.000000   \n",
       "1           0.000000     0.000000   android              0.000000   \n",
       "2           0.000000     0.000000   android              0.000000   \n",
       "3           0.000000     0.000000       ios              0.000000   \n",
       "4           0.000000     0.000000   android              0.000000   \n",
       "...              ...          ...       ...                   ...   \n",
       "1467827     0.037158     0.037158       ios              0.000000   \n",
       "1467828     0.039256     0.091267       ios              4.705161   \n",
       "1467829     0.000000     0.000000       ios              0.000000   \n",
       "1467830     0.177550     0.177550       ios              0.000000   \n",
       "1467831     0.114909     0.114909   android              0.000000   \n",
       "\n",
       "         target_iap_ltv_day30  target_ad_ltv_day30  target_full_ltv_day30  \n",
       "0                    0.000000             0.000000               0.000000  \n",
       "1                    0.000000             0.000000               0.000000  \n",
       "2                    0.000000             0.000000               0.000000  \n",
       "3                    0.000000             0.000000               0.000000  \n",
       "4                    0.000000             0.000000               0.000000  \n",
       "...                       ...                  ...                    ...  \n",
       "1467827              4.174198             0.037158               4.211356  \n",
       "1467828             10.967580             0.091267              15.764008  \n",
       "1467829             14.938221             0.000000              14.938221  \n",
       "1467830              7.402224             0.177550               7.579774  \n",
       "1467831              3.949385             0.114909               4.064295  \n",
       "\n",
       "[1467832 rows x 63 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = 'TEST TASK (dataset).csv'         # шлях до датасету\n",
    "data = pd.read_csv(path)\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "y-b0ltV2BrLe"
   },
   "source": [
    "Переведемо колонку дату в тип datetime та розділимо її на 3 колонки: рік, номер місяця та номер дня в місяці."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "k-DuDh9PC_wg",
    "outputId": "73a52cbf-13c7-4d9f-a6ac-94f1835542b2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min date =  2021-12-01 00:00:00\n",
      "Max date =  2022-01-31 00:00:00\n"
     ]
    }
   ],
   "source": [
    "data['install_date'] = pd.to_datetime(data['install_date'])\n",
    "print('Min date = ', data['install_date'].min())\n",
    "print('Max date = ', data['install_date'].max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "-xPTjn7ZF6RR"
   },
   "outputs": [],
   "source": [
    "date = data['install_date']\n",
    "data['year'] = date.dt.year\n",
    "data['month'] = date.dt.month\n",
    "data['day'] = date.dt.day\n",
    "\n",
    "data = data.drop(['install_date'],axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UXH5SFMuPExn"
   },
   "source": [
    "Замінюємо nan values на нульові значення (в даному випадку nan values відносно дуже мало і тому така проста заміна не буде впливати на результат)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "YYfWVYr0HGfv",
    "outputId": "0cac9445-12fe-472f-fc23-aee4a2d3d477"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Кількість nan values у колонках:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "chapters_finished_session3       0\n",
       "chapters_opened_session1         0\n",
       "chapters_opened_session3         0\n",
       "chapters_opened_session9         0\n",
       "chapters_closed_session1         0\n",
       "                              ... \n",
       "tickets_spent_day0               0\n",
       "total_sessions_day0           1365\n",
       "total_sessions_day1           1367\n",
       "total_sessions_day3           1368\n",
       "total_sessions_day7           1374\n",
       "Length: 65, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('Кількість nan values у колонках:')\n",
    "(data.shape[0]-data.count()).sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "nEsnt9yzVESF"
   },
   "outputs": [],
   "source": [
    "data = data.fillna(0.0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ivFyVLKDWT0J"
   },
   "source": [
    "Кодуємо колонки типу string"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "Z3Y8Uj-dTqoo"
   },
   "outputs": [],
   "source": [
    "cat_col = ['media_source', 'country_code', 'platform']   # виокремимо в окрему змінну категоріальні дані"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "YybaEEMOPIES"
   },
   "outputs": [],
   "source": [
    "# кодуємо дані з відповідних колонок\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "le = LabelEncoder()\n",
    "for col_name in cat_col:\n",
    "  le.fit(data[col_name])\n",
    "  data[col_name] = le.transform(data[col_name])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "yGFHkVfilPGj",
    "outputId": "f2d6941c-dfaa-4311-dc3d-ffeb0930df6f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['android', 'ios'], dtype=object)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "le.classes_        # перевірімо в які класи закодувалися назви платформ, щоб потім побудувати діаграма "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "454QqzaJQGgG",
    "outputId": "fad2a5b4-7ba7-4d38-8fad-2af470be9c25"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 41,  48, 117, 156,  38, 164,  61,  28,  63, 208,  35,  18,  17,\n",
       "       126, 194, 202,  43, 150, 172, 189, 153,  57, 106,  34,  14, 143,\n",
       "         9, 133, 141,  51, 203, 140,   5, 114, 168,  74, 122,  64, 161,\n",
       "       188, 180,  46, 132,  66, 191, 181, 213, 169, 102, 104, 139, 160,\n",
       "       212,  73, 176, 100, 118, 215,  12, 219,  19,  58, 144, 204,  70,\n",
       "       216, 207, 124, 125,  40, 137,  81,  78,  99,   7,  92,   1,  98,\n",
       "         3,  79,  96, 128,  54,  56,  11,  91,  62, 115, 121,  36, 146,\n",
       "       136,  75,  72, 170,   4,  13,  45, 178,   8,  77,  71,  76, 190,\n",
       "       185, 209,  47,   2, 162, 192,  24, 110,  26, 101,  65, 217, 173,\n",
       "       183, 220,   0,  39, 211,  55, 130, 195,  80,  89, 201, 147, 142,\n",
       "        10, 134,  60,  25,  42,  15, 167,  90,  88, 151,  31, 148,  97,\n",
       "        87,  52, 218, 158, 159,  30, 206, 205, 119,  50,  49, 221,  82,\n",
       "        94,  27,  84, 171, 200,  37, 108, 186, 157, 223,  23, 149, 113,\n",
       "       129,  22,  16, 182, 193,  86, 163, 197, 138, 116,  29, 177, 165,\n",
       "       120,  69, 166, 152,  95,  85, 222,  20,  59, 105, 111, 127, 174,\n",
       "       109, 196, 175, 103, 187, 131, 184,   6,  44, 154, 199, 145, 198,\n",
       "       210,  67,  93,  21,  68, 179,  33,  53, 112, 214,  83,  32, 123,\n",
       "       107, 135, 155])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['country_code'].unique()             # Перевіримо чи спрацювало кодування"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xHwG9aaB5OkS"
   },
   "source": [
    "Візуалізація даних"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "zYyYLKQo5VlF"
   },
   "outputs": [],
   "source": [
    "def vis(data,main_atrr,secon):     # функція, яка зображує точки даних у площині двох координат\n",
    "  y = data[main_atrr]\n",
    "  x = data[secon]\n",
    "  plt.plot(x,y,'o')\n",
    "  plt.xlabel(secon)\n",
    "  plt.ylabel(main_atrr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 276
    },
    "id": "dXNDHBID6c5u",
    "outputId": "28de0a37-6fe1-456b-afbf-7f649698d868"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vis(data, 'target_full_ltv_day30', 'app_iap_ltv_day3')   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZYIJHFc9kit6"
   },
   "source": [
    "Гістограма "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 294
    },
    "id": "zawvT9zmo9q4",
    "outputId": "2e8e437a-98b7-422b-d52d-32da199c22c6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='chapters_finished_day7', ylabel='Count'>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "var = 'chapters_finished_day7'\n",
    "sns.histplot(data=data,x=var,bins=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "sHM6J_ZvsJ0k",
    "outputId": "a98c22d5-c281-4d10-b4f6-80f4dc95f9cc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Дохід від гравців, у яких chapters_finished_day7=0 6113.190245197164\n",
      "Дохід від гравців, у яких chapters_finished_day7!=0 370932.99803929316\n",
      "Середній дохід від гравців, у яких chapters_finished_day7=0 0.0068819912519331155\n",
      "Середній дохід від гравців, у яких chapters_finished_day7!=0 0.6400428579008551\n"
     ]
    }
   ],
   "source": [
    "print('Дохід від гравців, у яких '+str(var)+'=0', data[data[var]==0]['target_full_ltv_day30'].sum())\n",
    "print('Дохід від гравців, у яких '+str(var)+'!=0', data[data[var]!=0]['target_full_ltv_day30'].sum())\n",
    "print('Середній дохід від гравців, у яких '+str(var)+'=0', data[data[var]==0]['target_full_ltv_day30'].mean())\n",
    "print('Середній дохід від гравців, у яких '+str(var)+'!=0', data[data[var]!=0]['target_full_ltv_day30'].mean())     "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "w7fxOne2kWqX"
   },
   "source": [
    "Зображуємо діаграму, щоб побачити як розподілені різні платформи у користувачів"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 355
    },
    "id": "e5qi2fJEg2fY",
    "outputId": "07c2abb3-dc06-4bb0-de20-bd4df9a69f27"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([<matplotlib.patches.Wedge at 0x12f3303d0>,\n",
       "  <matplotlib.patches.Wedge at 0x12f330af0>],\n",
       " [Text(0.35494489906126164, 1.041159987048288, 'ios'),\n",
       "  Text(-0.3549449965416937, -1.0411599538159433, 'android')],\n",
       " [Text(0.19360630857886996, 0.5679054474808842, '39.5%'),\n",
       "  Text(-0.19360636175001475, -0.5679054293541509, '60.5%')])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.pie([data['platform'].sum(),data['platform'].shape[0] - data['platform'].sum()], labels = ['ios','android'], autopct='%1.1f%%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Gcp8p9-7Pu-V"
   },
   "source": [
    "Кореляція"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "h_-pvHfkPf1c"
   },
   "outputs": [],
   "source": [
    "target_columns = ['target_sub_ltv_day30', 'target_iap_ltv_day30', 'target_ad_ltv_day30']\n",
    "ord_col = list(set(data.columns).difference(set(target_columns)))   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "iRfzsXH9S9mN",
    "outputId": "a61ec956-7c8a-46b0-f745-9f7fcb5fc9c2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['chapters_closed_day7', 'app_iap_ltv_day3', 'total_sessions_day1', 'diamonds_spent_day1', 'chapters_closed_session9', 'chapters_opened_session1', 'total_sessions_day7', 'retained_day1', 'chapters_finished_session3', 'diamonds_received_day7', 'day', 'media_source', 'diamonds_spent_day3', 'diamonds_spent_day0', 'chapters_finished_day3', 'chapters_opened_session3', 'tickets_spent_day0', 'month', 'retained_day7', 'chapters_finished_day1', 'diamonds_spent_session1', 'platform', 'chapters_opened_day7', 'chapters_closed_session1', 'chapters_finished_day0', 'retained_day3', 'target_full_ltv_day30', 'diamonds_received_day0', 'chapters_closed_day1', 'total_sessions_day0', 'tickets_spent_session3', 'chapters_closed_session3', 'app_sub_ltv_day1', 'chapters_finished_session1', 'diamonds_spent_session3', 'app_sub_ltv_day3', 'chapters_finished_day7', 'tickets_spent_day1', 'chapters_finished_session9', 'tickets_spent_day3', 'chapters_opened_day3', 'app_iap_ltv_day0', 'country_code', 'chapters_opened_day0', 'chapters_closed_day0', 'ad_ltv_day3', 'chapters_opened_day1', 'chapters_closed_day3', 'diamonds_spent_session9', 'diamonds_received_day1', 'diamonds_spent_day7', 'tickets_spent_day7', 'total_sessions_day3', 'ad_ltv_day1', 'year', 'ad_ltv_day0', 'app_sub_ltv_day0', 'chapters_opened_session9', 'app_iap_ltv_day1', 'tickets_spent_session9', 'diamonds_received_day3', 'tickets_spent_session1']\n"
     ]
    }
   ],
   "source": [
    "print(ord_col)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "3afQuLXT28K0"
   },
   "outputs": [],
   "source": [
    "data = data.astype('float64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Y1noin863vOE",
    "outputId": "93409742-7d6c-4198-ad03-323aab790e0e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.all(data.dtypes=='float64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "-TkCjcM0SQnJ",
    "outputId": "c442f6db-9484-4d49-cfcb-06c2c3f4494b"
   },
   "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>total_sessions_day0</th>\n",
       "      <th>total_sessions_day1</th>\n",
       "      <th>total_sessions_day3</th>\n",
       "      <th>total_sessions_day7</th>\n",
       "      <th>chapters_finished_day0</th>\n",
       "      <th>chapters_finished_day1</th>\n",
       "      <th>chapters_finished_day3</th>\n",
       "      <th>chapters_finished_day7</th>\n",
       "      <th>chapters_opened_day0</th>\n",
       "      <th>chapters_opened_day1</th>\n",
       "      <th>...</th>\n",
       "      <th>ad_ltv_day1</th>\n",
       "      <th>ad_ltv_day3</th>\n",
       "      <th>platform</th>\n",
       "      <th>target_sub_ltv_day30</th>\n",
       "      <th>target_iap_ltv_day30</th>\n",
       "      <th>target_ad_ltv_day30</th>\n",
       "      <th>target_full_ltv_day30</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>...</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "      <td>1.467832e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.845580e+00</td>\n",
       "      <td>2.191869e+00</td>\n",
       "      <td>2.606548e+00</td>\n",
       "      <td>3.110570e+00</td>\n",
       "      <td>1.234763e+00</td>\n",
       "      <td>1.498150e+00</td>\n",
       "      <td>1.772233e+00</td>\n",
       "      <td>2.055170e+00</td>\n",
       "      <td>2.099693e+00</td>\n",
       "      <td>2.382293e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>1.090574e-02</td>\n",
       "      <td>1.325368e-02</td>\n",
       "      <td>3.954172e-01</td>\n",
       "      <td>2.203037e-02</td>\n",
       "      <td>2.139268e-01</td>\n",
       "      <td>2.091572e-02</td>\n",
       "      <td>2.568728e-01</td>\n",
       "      <td>2.021392e+03</td>\n",
       "      <td>7.685820e+00</td>\n",
       "      <td>1.861522e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.530274e+00</td>\n",
       "      <td>3.620912e+00</td>\n",
       "      <td>5.345442e+00</td>\n",
       "      <td>7.884984e+00</td>\n",
       "      <td>3.048333e+00</td>\n",
       "      <td>3.975010e+00</td>\n",
       "      <td>5.033945e+00</td>\n",
       "      <td>6.325564e+00</td>\n",
       "      <td>3.117375e+00</td>\n",
       "      <td>4.077269e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>6.009280e-02</td>\n",
       "      <td>7.472105e-02</td>\n",
       "      <td>4.889403e-01</td>\n",
       "      <td>3.088758e-01</td>\n",
       "      <td>2.097916e+00</td>\n",
       "      <td>1.601929e-01</td>\n",
       "      <td>2.274466e+00</td>\n",
       "      <td>4.882406e-01</td>\n",
       "      <td>5.370647e+00</td>\n",
       "      <td>9.568710e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-2.829604e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-2.777195e+00</td>\n",
       "      <td>2.021000e+03</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>2.021000e+03</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>2.021000e+03</td>\n",
       "      <td>1.200000e+01</td>\n",
       "      <td>2.100000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>2.022000e+03</td>\n",
       "      <td>1.200000e+01</td>\n",
       "      <td>2.700000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.710000e+02</td>\n",
       "      <td>7.710000e+02</td>\n",
       "      <td>7.710000e+02</td>\n",
       "      <td>9.050000e+02</td>\n",
       "      <td>7.300000e+01</td>\n",
       "      <td>3.590000e+02</td>\n",
       "      <td>3.600000e+02</td>\n",
       "      <td>3.600000e+02</td>\n",
       "      <td>7.700000e+01</td>\n",
       "      <td>3.600000e+02</td>\n",
       "      <td>...</td>\n",
       "      <td>1.974900e+01</td>\n",
       "      <td>1.997062e+01</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>4.545741e+01</td>\n",
       "      <td>2.687878e+02</td>\n",
       "      <td>3.393084e+01</td>\n",
       "      <td>2.720042e+02</td>\n",
       "      <td>2.022000e+03</td>\n",
       "      <td>1.200000e+01</td>\n",
       "      <td>3.100000e+01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       total_sessions_day0  total_sessions_day1  total_sessions_day3  \\\n",
       "count         1.467832e+06         1.467832e+06         1.467832e+06   \n",
       "mean          1.845580e+00         2.191869e+00         2.606548e+00   \n",
       "std           2.530274e+00         3.620912e+00         5.345442e+00   \n",
       "min           0.000000e+00         0.000000e+00         0.000000e+00   \n",
       "25%           1.000000e+00         1.000000e+00         1.000000e+00   \n",
       "50%           1.000000e+00         1.000000e+00         1.000000e+00   \n",
       "75%           2.000000e+00         2.000000e+00         2.000000e+00   \n",
       "max           7.710000e+02         7.710000e+02         7.710000e+02   \n",
       "\n",
       "       total_sessions_day7  chapters_finished_day0  chapters_finished_day1  \\\n",
       "count         1.467832e+06            1.467832e+06            1.467832e+06   \n",
       "mean          3.110570e+00            1.234763e+00            1.498150e+00   \n",
       "std           7.884984e+00            3.048333e+00            3.975010e+00   \n",
       "min           0.000000e+00            0.000000e+00            0.000000e+00   \n",
       "25%           1.000000e+00            0.000000e+00            0.000000e+00   \n",
       "50%           1.000000e+00            0.000000e+00            0.000000e+00   \n",
       "75%           2.000000e+00            1.000000e+00            1.000000e+00   \n",
       "max           9.050000e+02            7.300000e+01            3.590000e+02   \n",
       "\n",
       "       chapters_finished_day3  chapters_finished_day7  chapters_opened_day0  \\\n",
       "count            1.467832e+06            1.467832e+06          1.467832e+06   \n",
       "mean             1.772233e+00            2.055170e+00          2.099693e+00   \n",
       "std              5.033945e+00            6.325564e+00          3.117375e+00   \n",
       "min              0.000000e+00            0.000000e+00          0.000000e+00   \n",
       "25%              0.000000e+00            0.000000e+00          1.000000e+00   \n",
       "50%              0.000000e+00            0.000000e+00          1.000000e+00   \n",
       "75%              1.000000e+00            1.000000e+00          2.000000e+00   \n",
       "max              3.600000e+02            3.600000e+02          7.700000e+01   \n",
       "\n",
       "       chapters_opened_day1  ...   ad_ltv_day1   ad_ltv_day3      platform  \\\n",
       "count          1.467832e+06  ...  1.467832e+06  1.467832e+06  1.467832e+06   \n",
       "mean           2.382293e+00  ...  1.090574e-02  1.325368e-02  3.954172e-01   \n",
       "std            4.077269e+00  ...  6.009280e-02  7.472105e-02  4.889403e-01   \n",
       "min            0.000000e+00  ...  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "25%            1.000000e+00  ...  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "50%            1.000000e+00  ...  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "75%            2.000000e+00  ...  0.000000e+00  0.000000e+00  1.000000e+00   \n",
       "max            3.600000e+02  ...  1.974900e+01  1.997062e+01  1.000000e+00   \n",
       "\n",
       "       target_sub_ltv_day30  target_iap_ltv_day30  target_ad_ltv_day30  \\\n",
       "count          1.467832e+06          1.467832e+06         1.467832e+06   \n",
       "mean           2.203037e-02          2.139268e-01         2.091572e-02   \n",
       "std            3.088758e-01          2.097916e+00         1.601929e-01   \n",
       "min           -2.829604e+00          0.000000e+00         0.000000e+00   \n",
       "25%            0.000000e+00          0.000000e+00         0.000000e+00   \n",
       "50%            0.000000e+00          0.000000e+00         0.000000e+00   \n",
       "75%            0.000000e+00          0.000000e+00         0.000000e+00   \n",
       "max            4.545741e+01          2.687878e+02         3.393084e+01   \n",
       "\n",
       "       target_full_ltv_day30          year         month           day  \n",
       "count           1.467832e+06  1.467832e+06  1.467832e+06  1.467832e+06  \n",
       "mean            2.568728e-01  2.021392e+03  7.685820e+00  1.861522e+01  \n",
       "std             2.274466e+00  4.882406e-01  5.370647e+00  9.568710e+00  \n",
       "min            -2.777195e+00  2.021000e+03  1.000000e+00  1.000000e+00  \n",
       "25%             0.000000e+00  2.021000e+03  1.000000e+00  1.000000e+01  \n",
       "50%             0.000000e+00  2.021000e+03  1.200000e+01  2.100000e+01  \n",
       "75%             0.000000e+00  2.022000e+03  1.200000e+01  2.700000e+01  \n",
       "max             2.720042e+02  2.022000e+03  1.200000e+01  3.100000e+01  \n",
       "\n",
       "[8 rows x 65 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()   # статистичні характеристики наших даних"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "A5FEJ50wA0MX",
    "outputId": "b5c3a8ef-078d-4017-edc7-f780a592e514"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "chapters_opened_session1      0.150980\n",
       "chapters_closed_day1          0.152283\n",
       "tickets_spent_session1        0.157572\n",
       "chapters_finished_session1    0.163845\n",
       "retained_day1                 0.177551\n",
       "chapters_closed_day3          0.183482\n",
       "retained_day3                 0.187840\n",
       "retained_day7                 0.192130\n",
       "total_sessions_day0           0.195521\n",
       "chapters_closed_day7          0.217513\n",
       "total_sessions_day1           0.229253\n",
       "tickets_spent_session3        0.229660\n",
       "chapters_opened_session3      0.235190\n",
       "chapters_finished_session3    0.244130\n",
       "total_sessions_day3           0.256603\n",
       "tickets_spent_day0            0.269870\n",
       "app_sub_ltv_day3              0.274343\n",
       "total_sessions_day7           0.284839\n",
       "chapters_opened_day0          0.286784\n",
       "diamonds_received_day0        0.290158\n",
       "chapters_finished_day0        0.292734\n",
       "tickets_spent_session9        0.298779\n",
       "ad_ltv_day0                   0.300505\n",
       "tickets_spent_day1            0.310353\n",
       "chapters_opened_session9      0.313546\n",
       "chapters_finished_session9    0.320740\n",
       "chapters_opened_day1          0.328068\n",
       "chapters_finished_day1        0.333980\n",
       "ad_ltv_day1                   0.334121\n",
       "tickets_spent_day3            0.346518\n",
       "diamonds_spent_session1       0.368041\n",
       "ad_ltv_day3                   0.368362\n",
       "chapters_opened_day3          0.369447\n",
       "chapters_finished_day3        0.375659\n",
       "tickets_spent_day7            0.391570\n",
       "diamonds_received_day1        0.394815\n",
       "chapters_opened_day7          0.413583\n",
       "chapters_finished_day7        0.420365\n",
       "diamonds_received_day3        0.476700\n",
       "diamonds_spent_session3       0.528519\n",
       "diamonds_received_day7        0.584553\n",
       "diamonds_spent_day0           0.605621\n",
       "app_iap_ltv_day0              0.636444\n",
       "diamonds_spent_session9       0.662524\n",
       "diamonds_spent_day1           0.674307\n",
       "app_iap_ltv_day1              0.717920\n",
       "diamonds_spent_day3           0.751245\n",
       "app_iap_ltv_day3              0.799408\n",
       "diamonds_spent_day7           0.809262\n",
       "Name: target_full_ltv_day30, dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# виберемо змінні, які мають кореляцію більше ніж 0.15 з головною змінною\n",
    "corr = data[ord_col].corr()\n",
    "corr_target = corr['target_full_ltv_day30']\n",
    "corr_target = corr_target[corr_target<1]\n",
    "a = np.abs(corr_target).sort_values()\n",
    "cor_min = 0.15\n",
    "corr_list = a[a>cor_min]\n",
    "corr_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "1KGfuxkAAHSh",
    "outputId": "d78dde06-e916-44f8-d243-3210dff78e12"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "кількість змінних, у яких кореляція з головною змінною>0.15 =  49\n"
     ]
    }
   ],
   "source": [
    "print('кількість змінних, у яких кореляція з головною змінною>'+str(cor_min)+' = ', corr_list.shape[0])\n",
    "main_col = list(corr_list.index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OInJwk_YYkwn"
   },
   "source": [
    "Вибір моделі"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "AKJ6Bn0zHcgC"
   },
   "outputs": [],
   "source": [
    "X = np.array(data[main_col])\n",
    "y = np.array(data['target_full_ltv_day30'])\n",
    "X_train, X_test, y_train, y_test = sk.model_selection.train_test_split(X, y, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "1si8lPjzK54j",
    "outputId": "e56ac5ff-40a6-438d-91d7-baac82be8e9d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "shape(X) =  (1467832, 49)\n",
      "shape(y) =  (1467832,)\n"
     ]
    }
   ],
   "source": [
    "print('shape(X) = ', X.shape)\n",
    "print('shape(y) = ', y.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MbsnkziiIQKK"
   },
   "source": [
    "Linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "K69dEp0CIBLe",
    "outputId": "71c6de7a-bf2c-435f-af2b-b07149909a72"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test R2_score =  0.7638522762409652\n",
      "train R2_score =  0.7990838274415739\n",
      "test rmse_score =  1.072284493316548\n",
      "test mape_score =  99494980624846.48\n",
      "test mae_score =  0.13278890084457187\n"
     ]
    }
   ],
   "source": [
    "model1 = LinearRegression().fit(X_train, y_train)\n",
    "print('test R2_score = ', model1.score(X_test, y_test))\n",
    "print('train R2_score = ', model1.score(X_train, y_train))\n",
    "print('test rmse_score = ', np.sqrt(mse(y_test, model1.predict(X_test))))\n",
    "print('test mape_score = ', mape(y_test, model1.predict(X_test)))\n",
    "print('test mae_score = ', mae(y_test, model1.predict(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "spgMOO85u_Gn",
    "outputId": "cdc67fb6-03d3-4a5d-e9a3-f28de4c85923"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(mape для y!=0): 22.86196157083573\n"
     ]
    }
   ],
   "source": [
    "print('(mape для y!=0):',mape(y_test[y_test!=0], model1.predict(X_test)[y_test!=0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "kBGtxRQ_QD45"
   },
   "source": [
    "MLPRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "C8Njgms8QHhU"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"alph1 = [0.01,0.03,0.05,0.08,0.12]\\nalph2 = [0.001,0.003,0.005,0.008,0.012]\\n\\nfor i in alph1:\\n  model2 = MLPRegressor(random_state=42, hidden_layer_sizes=(64,32), max_iter=500, alpha=i, n_iter_no_change=5).fit(X_train, y_train)\\n  print('alpha = ', i)\\n  print('test score = ', model2.score(X_test, y_test))\\n  print('train score = ', model2.score(X_train, y_train))\\n  print('\\n')\\n  \\nfor i in alph2:\\n  model2 = MLPRegressor(random_state=42, hidden_layer_sizes=(64,32), max_iter=500, alpha=i, n_iter_no_change=5).fit(X_train, y_train)\\n  print('alpha = ', i)\\n  print('test score = ', model2.score(X_test, y_test))\\n  print('train score = ', model2.score(X_train, y_train))\\n  print('\\n')\""
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# процес вибору адекватних параметрів для нейромережі\n",
    "'''alph1 = [0.01,0.03,0.05,0.08,0.12]\n",
    "alph2 = [0.001,0.003,0.005,0.008,0.012]\n",
    "\n",
    "for i in alph1:\n",
    "  model2 = MLPRegressor(random_state=42, hidden_layer_sizes=(64,32), max_iter=500, alpha=i, n_iter_no_change=5).fit(X_train, y_train)\n",
    "  print('alpha = ', i)\n",
    "  print('test score = ', model2.score(X_test, y_test))\n",
    "  print('train score = ', model2.score(X_train, y_train))\n",
    "  print('\\n')\n",
    "  \n",
    "for i in alph2:\n",
    "  model2 = MLPRegressor(random_state=42, hidden_layer_sizes=(64,32), max_iter=500, alpha=i, n_iter_no_change=5).fit(X_train, y_train)\n",
    "  print('alpha = ', i)\n",
    "  print('test score = ', model2.score(X_test, y_test))\n",
    "  print('train score = ', model2.score(X_train, y_train))\n",
    "  print('\\n')'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "aVb-a_d7eINS",
    "outputId": "00f9db29-c665-4d0e-aa0a-72e91a7876d0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test R2_score =  0.7844315226453266\n",
      "train R2_score =  0.8005214303320115\n",
      "test rmse_score =  1.0244971881004095\n",
      "test mape_score =  28819444574364.25\n",
      "test mae_score =  0.09425775305817256\n"
     ]
    }
   ],
   "source": [
    "model2 = MLPRegressor(random_state=42, hidden_layer_sizes=(64,32), max_iter=500, alpha=0.001, n_iter_no_change=5).fit(X_train, y_train)\n",
    "print('test R2_score = ', model2.score(X_test, y_test))\n",
    "print('train R2_score = ', model2.score(X_train, y_train))\n",
    "print('test rmse_score = ', np.sqrt(mse(y_test, model2.predict(X_test))))\n",
    "print('test mape_score = ', mape(y_test, model2.predict(X_test)))\n",
    "print('test mae_score = ', mae(y_test, model2.predict(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ZlqZv-7H2gYs",
    "outputId": "f3d8a9c5-92cc-4e73-f68b-3b702aa9daab"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(mape для y!=0): 7.521882858943411\n"
     ]
    }
   ],
   "source": [
    "print('(mape для y!=0):',mape(y_test[y_test!=0], model2.predict(X_test)[y_test!=0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Fv5csvfFQQf8",
    "outputId": "c74e8445-86ba-40e2-aac9-98bdcd8cfe97"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test R2_score =  0.7825550247756997\n",
      "train R2_score =  0.884298616418279\n"
     ]
    }
   ],
   "source": [
    "model3 = GradientBoostingRegressor(n_estimators=200, learning_rate=0.1, max_depth=3, random_state=42, loss='squared_error').fit(X_train, y_train)\n",
    "print('test R2_score = ', model3.score(X_test, y_test))\n",
    "print('train R2_score = ', model3.score(X_train, y_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "MqOqQBk-yJwJ",
    "outputId": "e75a8691-21c5-4e3c-8b1b-175307dc0012"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test rmse_score =  1.0289465890850418\n",
      "test mape_score =  30619358834314.445\n",
      "test mae_score =  0.0895138132817083\n"
     ]
    }
   ],
   "source": [
    "print('test rmse_score = ', np.sqrt(mse(y_test, model3.predict(X_test))))\n",
    "print('test mape_score = ', mape(y_test, model3.predict(X_test)))\n",
    "print('test mae_score = ', mae(y_test, model3.predict(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "-Mi4rjTe3Wdi",
    "outputId": "f59af3d9-8351-4620-bb95-d56d84aadf18"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(mape для y!=0): 5.21730293656799\n"
     ]
    }
   ],
   "source": [
    "print('(mape для y!=0):',mape(y_test[y_test!=0], model3.predict(X_test)[y_test!=0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "id": "E14IjNQkSBWo"
   },
   "outputs": [],
   "source": [
    "# збереження натренованих моделей\n",
    "def download_models(models):\n",
    "  import pickle\n",
    "\n",
    "  with open('models1.pkl', 'wb') as f:\n",
    "    pickle.dump(models, f)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 17
    },
    "id": "8EFxhwBDSTFi",
    "outputId": "22acbdb5-2e7d-403a-d481-1e34bd7e4c69"
   },
   "outputs": [],
   "source": [
    "models = [model1, model2, model3]\n",
    "download_models(models)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('test.npy', 'wb') as f:\n",
    "    np.save(f, X_test)\n",
    "    np.save(f, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}