P7_bootstrap_petrolium.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Описание проекта"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Аннотация**\n",
    "\n",
    "Допустим, вы работаете в добывающей компании «ГлавРосГосНефть». Нужно решить, где бурить новую скважину.\n",
    "\n",
    "Вам предоставлены пробы нефти в трёх регионах: в каждом 10 000 месторождений, где измерили качество нефти и объём её запасов. Постройте модель машинного обучения, которая поможет определить регион, где добыча принесёт наибольшую прибыль. Проанализируйте возможную прибыль и риски техникой *Bootstrap.*\n",
    "\n",
    "Шаги для выбора локации:\n",
    "\n",
    "- В избранном регионе ищут месторождения, для каждого определяют значения признаков;\n",
    "- Строят модель и оценивают объём запасов;\n",
    "- Выбирают месторождения с самым высокими оценками значений. Количество месторождений зависит от бюджета компании и стоимости разработки одной скважины;\n",
    "- Прибыль равна суммарной прибыли отобранных месторождений.\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Загрузка и подготовка данных"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Регион 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [],
   "source": [
    "geo_0 = pd.read_csv('/datasets/geo_data_0.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>f0</th>\n",
       "      <th>f1</th>\n",
       "      <th>f2</th>\n",
       "      <th>product</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>txEyH</td>\n",
       "      <td>0.705745</td>\n",
       "      <td>-0.497823</td>\n",
       "      <td>1.221170</td>\n",
       "      <td>105.280062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2acmU</td>\n",
       "      <td>1.334711</td>\n",
       "      <td>-0.340164</td>\n",
       "      <td>4.365080</td>\n",
       "      <td>73.037750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>409Wp</td>\n",
       "      <td>1.022732</td>\n",
       "      <td>0.151990</td>\n",
       "      <td>1.419926</td>\n",
       "      <td>85.265647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>iJLyR</td>\n",
       "      <td>-0.032172</td>\n",
       "      <td>0.139033</td>\n",
       "      <td>2.978566</td>\n",
       "      <td>168.620776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Xdl7t</td>\n",
       "      <td>1.988431</td>\n",
       "      <td>0.155413</td>\n",
       "      <td>4.751769</td>\n",
       "      <td>154.036647</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id        f0        f1        f2     product\n",
       "0  txEyH  0.705745 -0.497823  1.221170  105.280062\n",
       "1  2acmU  1.334711 -0.340164  4.365080   73.037750\n",
       "2  409Wp  1.022732  0.151990  1.419926   85.265647\n",
       "3  iJLyR -0.032172  0.139033  2.978566  168.620776\n",
       "4  Xdl7t  1.988431  0.155413  4.751769  154.036647"
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_0.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 100000 entries, 0 to 99999\n",
      "Data columns (total 5 columns):\n",
      "id         100000 non-null object\n",
      "f0         100000 non-null float64\n",
      "f1         100000 non-null float64\n",
      "f2         100000 non-null float64\n",
      "product    100000 non-null float64\n",
      "dtypes: float64(4), object(1)\n",
      "memory usage: 3.8+ MB\n"
     ]
    }
   ],
   "source": [
    "geo_0.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id          99990\n",
       "f0         100000\n",
       "f1         100000\n",
       "f2         100000\n",
       "product    100000\n",
       "dtype: int64"
      ]
     },
     "execution_count": 179,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_0.nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "В первом регионе на нескольких объектах (10), скорее всего, были произведены повторные замеры, что должно уточнить наши данные.\n",
    "\n",
    "## Регион 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>f0</th>\n",
       "      <th>f1</th>\n",
       "      <th>f2</th>\n",
       "      <th>product</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>kBEdx</td>\n",
       "      <td>-15.001348</td>\n",
       "      <td>-8.276000</td>\n",
       "      <td>-0.005876</td>\n",
       "      <td>3.179103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>62mP7</td>\n",
       "      <td>14.272088</td>\n",
       "      <td>-3.475083</td>\n",
       "      <td>0.999183</td>\n",
       "      <td>26.953261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>vyE1P</td>\n",
       "      <td>6.263187</td>\n",
       "      <td>-5.948386</td>\n",
       "      <td>5.001160</td>\n",
       "      <td>134.766305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>KcrkZ</td>\n",
       "      <td>-13.081196</td>\n",
       "      <td>-11.506057</td>\n",
       "      <td>4.999415</td>\n",
       "      <td>137.945408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>AHL4O</td>\n",
       "      <td>12.702195</td>\n",
       "      <td>-8.147433</td>\n",
       "      <td>5.004363</td>\n",
       "      <td>134.766305</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id         f0         f1        f2     product\n",
       "0  kBEdx -15.001348  -8.276000 -0.005876    3.179103\n",
       "1  62mP7  14.272088  -3.475083  0.999183   26.953261\n",
       "2  vyE1P   6.263187  -5.948386  5.001160  134.766305\n",
       "3  KcrkZ -13.081196 -11.506057  4.999415  137.945408\n",
       "4  AHL4O  12.702195  -8.147433  5.004363  134.766305"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_1 = pd.read_csv('/datasets/geo_data_1.csv')\n",
    "geo_1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 100000 entries, 0 to 99999\n",
      "Data columns (total 5 columns):\n",
      "id         100000 non-null object\n",
      "f0         100000 non-null float64\n",
      "f1         100000 non-null float64\n",
      "f2         100000 non-null float64\n",
      "product    100000 non-null float64\n",
      "dtypes: float64(4), object(1)\n",
      "memory usage: 3.8+ MB\n"
     ]
    }
   ],
   "source": [
    "geo_1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id          99996\n",
       "f0         100000\n",
       "f1         100000\n",
       "f2         100000\n",
       "product        12\n",
       "dtype: int64"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_1.nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Оценочные значения объемов продукта в данном регионе не уникальны - всего 12. Это выглядит подозрительно, однако возможно просто методология расчета аппроксимировала значения по классам и приближала их к среднему в классе \n",
    "\n",
    "## Регион 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>f0</th>\n",
       "      <th>f1</th>\n",
       "      <th>f2</th>\n",
       "      <th>product</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>fwXo0</td>\n",
       "      <td>-1.146987</td>\n",
       "      <td>0.963328</td>\n",
       "      <td>-0.828965</td>\n",
       "      <td>27.758673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>WJtFt</td>\n",
       "      <td>0.262778</td>\n",
       "      <td>0.269839</td>\n",
       "      <td>-2.530187</td>\n",
       "      <td>56.069697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>ovLUW</td>\n",
       "      <td>0.194587</td>\n",
       "      <td>0.289035</td>\n",
       "      <td>-5.586433</td>\n",
       "      <td>62.871910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>q6cA6</td>\n",
       "      <td>2.236060</td>\n",
       "      <td>-0.553760</td>\n",
       "      <td>0.930038</td>\n",
       "      <td>114.572842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>WPMUX</td>\n",
       "      <td>-0.515993</td>\n",
       "      <td>1.716266</td>\n",
       "      <td>5.899011</td>\n",
       "      <td>149.600746</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id        f0        f1        f2     product\n",
       "0  fwXo0 -1.146987  0.963328 -0.828965   27.758673\n",
       "1  WJtFt  0.262778  0.269839 -2.530187   56.069697\n",
       "2  ovLUW  0.194587  0.289035 -5.586433   62.871910\n",
       "3  q6cA6  2.236060 -0.553760  0.930038  114.572842\n",
       "4  WPMUX -0.515993  1.716266  5.899011  149.600746"
      ]
     },
     "execution_count": 207,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_2 = pd.read_csv('/datasets/geo_data_2.csv')\n",
    "\n",
    "geo_2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 100000 entries, 0 to 99999\n",
      "Data columns (total 5 columns):\n",
      "id         100000 non-null object\n",
      "f0         100000 non-null float64\n",
      "f1         100000 non-null float64\n",
      "f2         100000 non-null float64\n",
      "product    100000 non-null float64\n",
      "dtypes: float64(4), object(1)\n",
      "memory usage: 3.8+ MB\n"
     ]
    }
   ],
   "source": [
    "geo_2.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id          99996\n",
       "f0         100000\n",
       "f1         100000\n",
       "f2         100000\n",
       "product    100000\n",
       "dtype: int64"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_2.nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font color='blue'> \n",
    "### <u>КОММЕНТАРИЙ РЕВЬЮЕРА</u>\n",
    "</font>\n",
    "<font color='green'>\n",
    "данные на базе)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "На 4 скважинах в данном регионе были произведены повторные замеры\n",
    "\n",
    "### Выведем общую статистику по регионам"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>f0</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>0.500419</td>\n",
       "      <td>0.871832</td>\n",
       "      <td>-1.408605</td>\n",
       "      <td>-0.072580</td>\n",
       "      <td>0.502360</td>\n",
       "      <td>1.073581</td>\n",
       "      <td>2.362331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>f1</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>0.250143</td>\n",
       "      <td>0.504433</td>\n",
       "      <td>-0.848218</td>\n",
       "      <td>-0.200881</td>\n",
       "      <td>0.250252</td>\n",
       "      <td>0.700646</td>\n",
       "      <td>1.343769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>f2</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>2.502647</td>\n",
       "      <td>3.248248</td>\n",
       "      <td>-12.088328</td>\n",
       "      <td>0.287748</td>\n",
       "      <td>2.515969</td>\n",
       "      <td>4.715088</td>\n",
       "      <td>16.003790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>product</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>92.500000</td>\n",
       "      <td>44.288691</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>56.497507</td>\n",
       "      <td>91.849972</td>\n",
       "      <td>128.564089</td>\n",
       "      <td>185.364347</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            count       mean        std        min        25%        50%  \\\n",
       "f0       100000.0   0.500419   0.871832  -1.408605  -0.072580   0.502360   \n",
       "f1       100000.0   0.250143   0.504433  -0.848218  -0.200881   0.250252   \n",
       "f2       100000.0   2.502647   3.248248 -12.088328   0.287748   2.515969   \n",
       "product  100000.0  92.500000  44.288691   0.000000  56.497507  91.849972   \n",
       "\n",
       "                75%         max  \n",
       "f0         1.073581    2.362331  \n",
       "f1         0.700646    1.343769  \n",
       "f2         4.715088   16.003790  \n",
       "product  128.564089  185.364347  "
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_0.describe().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>f0</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>1.141296</td>\n",
       "      <td>8.965932</td>\n",
       "      <td>-31.609576</td>\n",
       "      <td>-6.298551</td>\n",
       "      <td>1.153055</td>\n",
       "      <td>8.621015</td>\n",
       "      <td>29.421755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>f1</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>-4.796579</td>\n",
       "      <td>5.119872</td>\n",
       "      <td>-26.358598</td>\n",
       "      <td>-8.267985</td>\n",
       "      <td>-4.813172</td>\n",
       "      <td>-1.332816</td>\n",
       "      <td>18.734063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>f2</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>2.494541</td>\n",
       "      <td>1.703572</td>\n",
       "      <td>-0.018144</td>\n",
       "      <td>1.000021</td>\n",
       "      <td>2.011479</td>\n",
       "      <td>3.999904</td>\n",
       "      <td>5.019721</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>product</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>68.825000</td>\n",
       "      <td>45.944423</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>26.953261</td>\n",
       "      <td>57.085625</td>\n",
       "      <td>107.813044</td>\n",
       "      <td>137.945408</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            count       mean        std        min        25%        50%  \\\n",
       "f0       100000.0   1.141296   8.965932 -31.609576  -6.298551   1.153055   \n",
       "f1       100000.0  -4.796579   5.119872 -26.358598  -8.267985  -4.813172   \n",
       "f2       100000.0   2.494541   1.703572  -0.018144   1.000021   2.011479   \n",
       "product  100000.0  68.825000  45.944423   0.000000  26.953261  57.085625   \n",
       "\n",
       "                75%         max  \n",
       "f0         8.621015   29.421755  \n",
       "f1        -1.332816   18.734063  \n",
       "f2         3.999904    5.019721  \n",
       "product  107.813044  137.945408  "
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_1.describe().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>f0</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>0.002023</td>\n",
       "      <td>1.732045</td>\n",
       "      <td>-8.760004</td>\n",
       "      <td>-1.162288</td>\n",
       "      <td>0.009424</td>\n",
       "      <td>1.158535</td>\n",
       "      <td>7.238262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>f1</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>-0.002081</td>\n",
       "      <td>1.730417</td>\n",
       "      <td>-7.084020</td>\n",
       "      <td>-1.174820</td>\n",
       "      <td>-0.009482</td>\n",
       "      <td>1.163678</td>\n",
       "      <td>7.844801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>f2</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>2.495128</td>\n",
       "      <td>3.473445</td>\n",
       "      <td>-11.970335</td>\n",
       "      <td>0.130359</td>\n",
       "      <td>2.484236</td>\n",
       "      <td>4.858794</td>\n",
       "      <td>16.739402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>product</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>44.749921</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>59.450441</td>\n",
       "      <td>94.925613</td>\n",
       "      <td>130.595027</td>\n",
       "      <td>190.029838</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            count       mean        std        min        25%        50%  \\\n",
       "f0       100000.0   0.002023   1.732045  -8.760004  -1.162288   0.009424   \n",
       "f1       100000.0  -0.002081   1.730417  -7.084020  -1.174820  -0.009482   \n",
       "f2       100000.0   2.495128   3.473445 -11.970335   0.130359   2.484236   \n",
       "product  100000.0  95.000000  44.749921   0.000000  59.450441  94.925613   \n",
       "\n",
       "                75%         max  \n",
       "f0         1.158535    7.238262  \n",
       "f1         1.163678    7.844801  \n",
       "f2         4.858794   16.739402  \n",
       "product  130.595027  190.029838  "
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geo_2.describe().T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Выводы по этапу:\n",
    "\n",
    "- Данные успешно загружены и соотносимы между собой\n",
    "\n",
    "- Пропуски отсутствуют, полных дубликатов также не обнаружено (как минимум 3 столбца - уникальны)\n",
    "\n",
    "- Некоторые id повторяются, однако это не является проблемой, так как они предоставляют нам дополнительные сведения для моделей, хоть и взяты с одного источника. Более того, он не понадобится нам в дальнейшем.\n",
    "\n",
    "- Статистика информативна только по продукту: самое маленькое среднее в выборках - в регионе с индексом 1."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Обучение и проверка модели"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.linear_model import LinearRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "def grid_search(model, param_grid, cv, x, y):\n",
    "        grid_model = GridSearchCV(model, param_grid=param_grid, cv=cv, verbose=1, n_jobs=-1)\n",
    "        grid_model.fit(x, y)\n",
    "        best_estimator = grid_model.best_estimator_\n",
    "        return best_estimator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualize(data,title,x ='product', y ='predicted'):\n",
    "    plt.figure(figsize=(15,10))\n",
    "    ax = sns.regplot(x ='product',y = 'predicted', x_bins=100, data = data).set_title(title)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualize_dist(data1, data2, title, bins=100):\n",
    "    _,ax = plt.subplots(figsize=(10,10))\n",
    "    \n",
    "    ax.hist(data1, bins = bins, color = 'red', alpha = 0.65, label = 'real')\n",
    "    ax.hist(data2, bins = bins, color = 'blue', alpha = 0.65, label = 'data2_name')\n",
    "        \n",
    "    ax.axvline(data1.mean(),linestyle = '--', color = 'lime', label = 'Real')\n",
    "                \n",
    "    ax.axvline(data2.mean(),linestyle = '--', color = 'coral', label = 'Predicted')\n",
    "                \n",
    "    #ax.set_ylabel(y_label)\n",
    "    #ax.set_xlabel(x_label)\n",
    "    ax.set_title(title)\n",
    "    ax.legend(loc='best')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = {'fit_intercept':[True, False], 'normalize':[True, False], 'copy_X':[True, False]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [],
   "source": [
    "lin = LinearRegression()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Регион 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_1, X_test_1, y_train_1, y_test_1 = train_test_split(\\\n",
    "                                geo_0.iloc[:,1:-1], geo_0['product'], test_size=0.25, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((75000, 3), (75000,))"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_1.shape, y_train_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 4 folds for each of 8 candidates, totalling 32 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  32 out of  32 | elapsed:    3.1s finished\n"
     ]
    }
   ],
   "source": [
    "lin_1 = grid_search(lin, param_grid, 4, X_train_1, y_train_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predict_1 = lin_1.predict(X_test_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [],
   "source": [
    "region_1_report = pd.DataFrame(y_test_1)\n",
    "region_1_report['predicted'] = y_predict_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>product</th>\n",
       "      <th>predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>46784</td>\n",
       "      <td>153.639837</td>\n",
       "      <td>176.536104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27658</td>\n",
       "      <td>140.631646</td>\n",
       "      <td>176.274510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6496</td>\n",
       "      <td>178.879516</td>\n",
       "      <td>173.249504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>65743</td>\n",
       "      <td>176.807828</td>\n",
       "      <td>172.802708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>93716</td>\n",
       "      <td>130.985681</td>\n",
       "      <td>172.744977</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          product   predicted\n",
       "46784  153.639837  176.536104\n",
       "27658  140.631646  176.274510\n",
       "6496   178.879516  173.249504\n",
       "65743  176.807828  172.802708\n",
       "93716  130.985681  172.744977"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "region_1_report.sort_values('predicted', ascending=False).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize(region_1_report, 'Регрессия для 1-го региона')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Расчет метрик и среднего значения"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Метрики для региона 1:\n",
      "Корень среднеквадратичной ошибки - 37.76\n",
      "Абсолютная средняя ошибка  - 31.05\n",
      "R2-Score - 0.27\n",
      "\n",
      "Среднее значение для региона - 92.40, настоящее среднее - 92.33\n"
     ]
    }
   ],
   "source": [
    "mse_1 = mean_squared_error(y_test_1, y_predict_1 )\n",
    "rmse_1 = mse_1 ** 0.5\n",
    "\n",
    "mae_1 = mean_absolute_error(y_test_1, y_predict_1)\n",
    "\n",
    "r2_1 = r2_score(y_test_1, y_predict_1 )\n",
    "\n",
    "mean_1 = np.mean(region_1_report['predicted'])\n",
    "mean_1_true = np.mean(region_1_report['product'])\n",
    "\n",
    "print(f'Метрики для региона 1:\\nКорень среднеквадратичной ошибки - \\\n",
    "{rmse_1:.2f}\\nАбсолютная средняя ошибка  - {mae_1:.2f}\\nR2-Score - {r2_1:.2f}\\\n",
    "\\n\\nСреднее значение для региона - {mean_1:.2f}, настоящее среднее - {mean_1_true:.2f}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['region_1',\n",
       "  37.75660035026169,\n",
       "  31.047681538788183,\n",
       "  0.2728287328173471,\n",
       "  92.3987999065777,\n",
       "  92.32595637084387]]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores.append(['region_1',  rmse_1, mae_1, r2_1, mean_1, mean_1_true])\n",
    "scores"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Вывод по первому региону:\n",
    "\n",
    "- Модель показала спорную эффективность в предсказании значении скважины – по сути она предсказывает аппроксимировала к среднему значению показатели по разным месторождениям. Отсюда логичный вывод, что у нас недостаток \n",
    "\n",
    "- Средний запас сырья предсказанный моделью - 92,4 тыс. баррелей\n",
    "\n",
    "- Корень из среднеквадратической ошибки (RMSE) - 37,75 тыс. баррелей\n",
    "\n",
    "- Абсолютная ошибка - 31 тыс. баррелей\n",
    "\n",
    "Таким образом, мы получили достаточно широкий разброс значений относительно действительного для данного региона. Проделаем ту же операцию с остальными."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Регион 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_2, X_test_2, y_train_2, y_test_2 = train_test_split(\\\n",
    "                                geo_1.iloc[:,1:-1], geo_1['product'], test_size=0.25, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 4 folds for each of 8 candidates, totalling 32 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  32 out of  32 | elapsed:    3.6s finished\n"
     ]
    }
   ],
   "source": [
    "lin_2 = grid_search(lin, param_grid, 4, X_train_2, y_train_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predict_2 = lin_2.predict(X_test_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "region_2_report = pd.DataFrame(y_test_2)\n",
    "region_2_report['predicted'] = y_predict_2\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize(region_2_report, 'Регрессия для 2-го региона')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Метрики для региона 2:\n",
      "Корень среднеквадратичной ошибки - 0.89\n",
      "Абсолютная средняя ошибка  - 0.72\n",
      "R2-Score - 1.00\n",
      "\n",
      "Среднее значение для региона - 68.71, настоящее среднее - 68.73\n"
     ]
    }
   ],
   "source": [
    "mse_2 = mean_squared_error(y_test_2, y_predict_2 )\n",
    "rmse_2 = mse_2 ** 0.5\n",
    "\n",
    "mae_2 = mean_absolute_error(y_test_2, y_predict_2)\n",
    "\n",
    "r2_2 = r2_score(y_test_2, y_predict_2 )\n",
    "\n",
    "mean_2 = np.mean(region_2_report['predicted'])\n",
    "mean_2_true = np.mean(region_2_report['product'])\n",
    "\n",
    "print(f'Метрики для региона 2:\\nКорень среднеквадратичной ошибки - \\\n",
    "{rmse_2:.2f}\\nАбсолютная средняя ошибка  - {mae_2:.2f}\\nR2-Score - {r2_2:.2f}\\\n",
    "\\n\\nСреднее значение для региона - {mean_2:.2f}, настоящее среднее - {mean_2_true:.2f}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores.append(['region_2',  rmse_2, mae_2, r2_2, mean_2, mean_2_true])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Вывод по второму региону:\n",
    "\n",
    "- Модель показала спорную эффективность в предсказании значении скважины – по сути она предсказывает аппроксимировала к среднему значению показатели по разным месторождениям. Отсюда логичный вывод, что у нас недостаток \n",
    "\n",
    "- Средний запас сырья предсказанный моделью - 68.71 тыс. баррелей, реальный - 68.73\n",
    "\n",
    "- Корень из среднеквадратической ошибки (RMSE) - 0.89 тыс. баррелей\n",
    "\n",
    "- Абсолютная ошибка - 0.72 тыс. баррелей\n",
    "\n",
    "- R2-Score - 1\n",
    "\n",
    "Таким образом, мы получили очень хорошие оценки, следовательно низкая вариация объектом по целевому признаку в регионе или связь между факторными переменными и целевой – более явная. Перейдем к третьему региону"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Регион 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_3, X_test_3, y_train_3, y_test_3 = train_test_split(\\\n",
    "                                geo_2.iloc[:,1:-1], geo_2['product'], test_size=0.25, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 4 folds for each of 8 candidates, totalling 32 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  32 out of  32 | elapsed:    3.3s finished\n"
     ]
    }
   ],
   "source": [
    "lin_3 = grid_search(lin, param_grid, 4, X_train_3, y_train_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predict_3 = lin_3.predict(X_test_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "region_3_report = pd.DataFrame(y_test_3)\n",
    "region_3_report['predicted'] = y_predict_3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize(region_3_report, 'Регрессия для 3-го региона')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Метрики для региона 3:\n",
      "Корень среднеквадратичной ошибки - 40.15\n",
      "Абсолютная средняя ошибка  - 32.90\n",
      "R2-Score - 0.20\n",
      "\n",
      "Среднее значение для региона - 94.77, настоящее среднее - 95.15\n"
     ]
    }
   ],
   "source": [
    "mse_3 = mean_squared_error(y_test_3, y_predict_3 )\n",
    "rmse_3 = mse_3 ** 0.5\n",
    "\n",
    "mae_3 = mean_absolute_error(y_test_3, y_predict_3)\n",
    "\n",
    "r2_3 = r2_score(y_test_3, y_predict_3)\n",
    "\n",
    "mean_3 = np.mean(region_3_report['predicted'])\n",
    "mean_3_true = np.mean(region_3_report['product'])\n",
    "\n",
    "print(f'Метрики для региона 3:\\nКорень среднеквадратичной ошибки - \\\n",
    "{rmse_3:.2f}\\nАбсолютная средняя ошибка  - {mae_3:.2f}\\nR2-Score - {r2_3:.2f}\\\n",
    "\\n\\nСреднее значение для региона - {mean_3:.2f}, настоящее среднее - {mean_3_true:.2f}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores.append(['region_3',  rmse_3, mae_3, r2_3, mean_3, mean_3_true])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Вывод по второму региону:\n",
    "\n",
    "- Модель показала похожие результирующие показатели по сравнению с 1-ым регионом, нежели, чем со второым\n",
    "\n",
    "- Средний запас сырья предсказанный моделью - 94.77 тыс. баррелей, реальный - 95.15\n",
    "\n",
    "- Корень из среднеквадратической ошибки (RMSE) - 40,15 тыс. баррелей\n",
    "\n",
    "- Абсолютная ошибка - 32.90 тыс. баррелей\n",
    "\n",
    "- R2-Score - 0.2\n",
    "\n",
    "**Таким образом, мы получили схожие результаты, как в первом регионе**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Посмотрим на распределение ответов в каждом из регионов\n",
    "\n",
    "### Регион 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_dist(region_1_report['product'],region_1_report['predicted'],'Распределение ответов и реальных значений (1)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Регион 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_dist(region_2_report['product'],region_2_report['predicted'],'Распределение ответов и реальных значений (2)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Регион 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_dist(region_3_report['product'],region_3_report['predicted'],'Распределение ответов и реальных значений (3)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Таким образом, мы явном можем увидеть то, что\n",
    " \n",
    " - **в регионе 1 и 3 модель линейной регрессии \"усреднила запасы\", плотность её оценок гораздо выше, чем реальные значения запасов.**\n",
    "\n",
    "\n",
    " - **в регионе 2 - месторождения разбились на группы по параметрам, что заставляет задуматься о том, что факторные и целевая переменные были приведены к некоему усредненному значению и лишь затем переданы в работу.**\n",
    "\n",
    "### Результаты работы модели в разных регионах"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scores = pd.DataFrame(data = scores, columns =['region','RMSE', 'MAE', 'R2-Score', 'Predicted_mean', 'Real_mean'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row1_col3 {\n",
       "            background-color:  lightblue;\n",
       "        }    #T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col1 {\n",
       "            background-color:  lightblue;\n",
       "        }    #T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col2 {\n",
       "            background-color:  lightblue;\n",
       "        }    #T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col4 {\n",
       "            background-color:  lightblue;\n",
       "        }    #T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col5 {\n",
       "            background-color:  lightblue;\n",
       "        }</style><table id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >region</th>        <th class=\"col_heading level0 col1\" >RMSE</th>        <th class=\"col_heading level0 col2\" >MAE</th>        <th class=\"col_heading level0 col3\" >R2-Score</th>        <th class=\"col_heading level0 col4\" >Predicted_mean</th>        <th class=\"col_heading level0 col5\" >Real_mean</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row0_col0\" class=\"data row0 col0\" >region_1</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row0_col1\" class=\"data row0 col1\" >37.7566</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row0_col2\" class=\"data row0 col2\" >31.0477</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row0_col3\" class=\"data row0 col3\" >0.272829</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row0_col4\" class=\"data row0 col4\" >92.3988</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row0_col5\" class=\"data row0 col5\" >92.326</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row1_col0\" class=\"data row1 col0\" >region_2</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row1_col1\" class=\"data row1 col1\" >0.89028</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row1_col2\" class=\"data row1 col2\" >0.717012</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row1_col3\" class=\"data row1 col3\" >0.999625</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row1_col4\" class=\"data row1 col4\" >68.7129</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row1_col5\" class=\"data row1 col5\" >68.7254</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col0\" class=\"data row2 col0\" >region_3</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col1\" class=\"data row2 col1\" >40.1459</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col2\" class=\"data row2 col2\" >32.897</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col3\" class=\"data row2 col3\" >0.196347</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col4\" class=\"data row2 col4\" >94.771</td>\n",
       "                        <td id=\"T_3e4e2d78_ffdc_11ea_9dbe_02420a393713row2_col5\" class=\"data row2 col5\" >95.151</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fae3db06850>"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_scores.style.highlight_max(color='lightblue', axis = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Выводы по данной части:\n",
    "\n",
    "1. В каждом регионе были построены модели с соответствующим RMSE (корень среднеквадратичной ошибки) :\n",
    "    - Регион 1 - 37.7566\n",
    "    - Регион 2 - 0.89028\n",
    "    - Регион 3 - 40.1459\n",
    "\n",
    "\n",
    "2. MAE (абсолютная ошибка):\n",
    "    - Регион 1 - 31.05\n",
    "    - Регион 2 - 0.71\n",
    "    - Регион 3 - 32.89\n",
    "\n",
    "\n",
    "3. R2-Score (коэффициент детерминации):\n",
    "    - Регион 1 - 0.27\n",
    "    - Регион 2 - 0.99\n",
    "    - Регион 3 - 0.19\n",
    " \n",
    " \n",
    "4. Среднее (тыс. баррелей)\n",
    "    - Предсказанное среднее в 1-м регионе - 92.4, реальное - 92.3 \n",
    "    - Предсказанное среднее в 2-м регионе - 68.7, реальное - 68.7\n",
    "    - Предсказанное среднее в 3-м регионе - 94.7, реальное - 95.15\n",
    "    \n",
    "    \n",
    "    \n",
    "5. Основываясь на графическом представлении распределений реальных и предсказанных значений, мы можем заключить, что регион 1 и 2 были \"усреднены\" по показаниям запасов в связи с недостаточной полной признаков (распределение предсказаний намного более плотное, чем ответов). То есть, они были существенно аппроксимированы к среднему значению в регионе. В регионе 2 значения факторных и целевой переменной, скорее всего, были существенным образом стандартизированны (Предположим, что нам дали усредненные значения в группах)\n",
    "\n",
    "\n",
    "6. Таким образом, на полученных данных модели оказались малопригодны. Для 1 и 3-го по предсказанием на условно \"сырых\" данных - необходимо нарастить количество признаков, чтобы добиться лучшего результата, для 2-го представить полные данные"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Подготовка к расчёту прибыли\n",
    "\n",
    "Сохраним ключевые значения констант для расчетов:\n",
    "\n",
    "- Общий бюджет - 10 млрд. рублей\n",
    "\n",
    "- Доход с единицы признака - 450 рублей * 10^3\n",
    "\n",
    "- Исследуемые скважины - 500\n",
    "\n",
    "- Отбор по 200 лучших\n",
    "\n",
    "**Константа бюджета (рубли)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10000000000"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "budget = 10**10\n",
    "budget"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Константа дохода с единицы `product` (рубли/тыс.баррелей)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "450000"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "revenue_per_well = 450000\n",
    "\n",
    "revenue_per_well"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Посчитаем минимально необходимое количество продукции на каждой скважине в разрабатываемом регионе:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "На каждой скважине требуется как минимум 111 тыс. баррелей\n",
      "Для окупаемости бюджета в регионе требуется освоить всего 22222 тыс.баррелей\n"
     ]
    }
   ],
   "source": [
    "min_product_single = int(budget/(revenue_per_well*200))\n",
    "min_product_sum = int(budget/(revenue_per_well))\n",
    "\n",
    "print(f'На каждой скважине требуется как минимум {min_product_single}\\\n",
    " тыс. баррелей\\nДля окупаемости бюджета в регионе требуется освоить всего {min_product_sum} тыс.баррелей')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Получили, что для окупаемости бюджета в 10 млрд. рублей нужно, чтобы в среднем на скважинах было как минимум 111 тыс. баррелей запаса.\n",
    "\n",
    "Таким образом, наша задача сводится к поиску такого региона, в котором наша модель наиболее удовлетворительно предсказывает запасы на скважине. \n",
    "\n",
    "Функция расчет прибыли должна по предсказаниям по шахте давать прогноз прибыли:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "def profit_(data, col_1 ='product', col_2 = 'predicted'):\n",
    "    \n",
    "    wells = data.sort_values(by = col_2 , ascending=False)\n",
    "    good_wells = wells.iloc[:200]\n",
    "    total_revenue = np.sum(good_wells[col_1]) * revenue_per_well \n",
    "    profit = total_revenue - budget\n",
    "    return profit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "profit() missing 1 required positional argument: 'predictions'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-191-2a21842720a1>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m print(f'Максимальная прибыль первого региона - ', profit(region_1_report)\\\n\u001b[0m\u001b[1;32m      2\u001b[0m       \u001b[0;34m,\u001b[0m\u001b[0;34m'\\nМаксимальная прибыль второго региона - '\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mprofit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mregion_2_report\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;31m\\\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m      ,'\\nМаксимальная прибыль третьего региона - ',profit(region_3_report))\n",
      "\u001b[0;31mTypeError\u001b[0m: profit() missing 1 required positional argument: 'predictions'"
     ]
    }
   ],
   "source": [
    "print(f'Максимальная прибыль первого региона - ', profit(region_1_report)\\\n",
    "      ,'\\nМаксимальная прибыль второго региона - ',profit(region_2_report)\\\n",
    "     ,'\\nМаксимальная прибыль третьего региона - ',profit(region_3_report))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Данные показатели иллюстрируют поведение в \"тепличных\" условиях, когда мы можем разведать все месторождения и без дополнительной процедуры поиска точного значения запасов.\n",
    "\n",
    "**Алгоритм проверки регионов:**\n",
    "\n",
    "1) Извлечь 500 наблюдений из тестовой выборки\n",
    "\n",
    "2) Выбрать лучшие 200 по оценке модели\n",
    "\n",
    "3) Сохранить результат прибыльности выборки\n",
    "\n",
    "4) Повторить для 1000 выборок\n",
    "\n",
    "5) Сохранить результат в массив"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bootstrap(data, size = 500, iterations = 1000):\n",
    "    \n",
    "    profit_results = []\n",
    "    state = np.random.RandomState(42)\n",
    "    \n",
    "    for i in range(0, iterations):\n",
    "        comp_sample = data.sample(size, replace = True, random_state = state)\n",
    "        sample_profit = profit(comp_sample)\n",
    "        \n",
    "        profit_results.append(sample_profit)\n",
    "        \n",
    "    return profit_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Расчёт прибыли и рисков "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy.stats as st\n",
    "\n",
    "revenue_report = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_risk(array):\n",
    "    total = len(array)\n",
    "    negative = 0\n",
    "    for mean in array:\n",
    "        if mean < 0:\n",
    "            negative +=1\n",
    "    \n",
    "    return negative / total"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Регион 1\n",
    "\n",
    "**Бустрапим:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "region_1_bootstrap = bootstrap(region_1_report)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "region_1_profits = np.divide(region_1_bootstrap, 10**9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Вычисляем важные показатели:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_reg1 = np.mean(region_1_profits)\n",
    "median_reg1 = np.median(region_1_profits)\n",
    "lower_reg1 = np.quantile(region_1_profits, .025)\n",
    "conf_interval_reg1 = st.t.interval(0.95, len(region_1_profits)-1, \n",
    "                                    mean_reg1)\n",
    "confidence_interval_1 = (np.quantile(region_1_profits, 0.025), np.quantile(region_1_profits, 0.975))\n",
    "                         \n",
    "risks_1 = compute_risk(region_1_profits)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Визуализируем**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7fae3e056f90>"
      ]
     },
     "execution_count": 84,
     "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": [
    "ax = sns.distplot(region_1_bootstrap)\n",
    "ax.axvline(np.quantile(region_1_bootstrap, 0.025),linestyle = '--',color='r')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Добавляем в отчет**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['region_1',\n",
       "  0.3995754780542297,\n",
       "  0.40378087879384517,\n",
       "  -0.11046789533197227,\n",
       "  (-1.5627659830792189, 2.361916939187678),\n",
       "  (-0.11046789533197227, 0.8974603277178793),\n",
       "  0.06]]"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "revenue_report.append(['region_1', mean_reg1,median_reg1,\\\n",
    "                       lower_reg1,conf_interval_reg1,\\\n",
    "                       confidence_interval_1,risks_1])\n",
    "\n",
    "revenue_report"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Регион 2\n",
    "\n",
    "**Бутстрапим:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "region_2_bootstrap = bootstrap(region_2_report)\n",
    "\n",
    "region_2_profits = np.divide(region_2_bootstrap, 10**9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Вычисляем важные показатели:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_reg2 = np.mean(region_2_profits)\n",
    "median_reg2 = np.median(region_2_profits)\n",
    "lower_reg2 = np.quantile(region_2_profits, .025)\n",
    "conf_interval_reg2 = st.t.interval(0.95, len(region_2_profits)-1, \n",
    "                                    mean_reg2)\n",
    "\n",
    "confidence_interval_2 = (np.quantile(region_2_profits, 0.025), np.quantile(region_2_profits, 0.975))\n",
    "                         \n",
    "risks_2 = compute_risk(region_2_profits)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Визуализируем**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7fae46b770d0>"
      ]
     },
     "execution_count": 88,
     "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": [
    "ax = sns.distplot(region_2_bootstrap)\n",
    "ax.axvline(np.quantile(region_2_bootstrap, 0.025),linestyle = '--',color='r')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Добавляем в отчет**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "revenue_report.append(['region_2', mean_reg2,median_reg2,lower_reg2,conf_interval_reg2,confidence_interval_2,risks_2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Регион 3\n",
    "\n",
    "**Бутстрапим**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "region_3_bootstrap = bootstrap(region_3_report)\n",
    "\n",
    "region_3_profits = np.divide(region_3_bootstrap, 10**9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Вычисляем важные показатели:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_reg3 = np.mean(region_3_profits)\n",
    "lower_reg3 = np.quantile(region_3_profits, .025)\n",
    "median_reg3 = np.median(region_3_profits)\n",
    "conf_interval_reg3 = st.t.interval(0.95, len(region_3_profits)-1, \n",
    "                                    mean_reg2)\n",
    "\n",
    "#дополнение\n",
    "\n",
    "confidence_interval_3 = (np.quantile(region_3_profits, 0.025), np.quantile(region_3_profits, 0.975))\n",
    "                         \n",
    "risks_3 = compute_risk(region_3_profits)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Визуализируем**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7fae3e8d8b90>"
      ]
     },
     "execution_count": 92,
     "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": [
    "ax = sns.distplot(region_3_bootstrap)\n",
    "ax.axvline(np.quantile(region_3_bootstrap, 0.025),linestyle = '--',color='r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "revenue_report.append(['region_3', mean_reg3,median_reg3,lower_reg3,conf_interval_reg3,confidence_interval_3,risks_3])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Отчет о прибыльности и общий вывод"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_061b227e_ffde_11ea_9dbe_02420a393713row1_col1 {\n",
       "            background-color:  lightgreen;\n",
       "        }    #T_061b227e_ffde_11ea_9dbe_02420a393713row1_col2 {\n",
       "            background-color:  lightgreen;\n",
       "        }    #T_061b227e_ffde_11ea_9dbe_02420a393713row1_col3 {\n",
       "            background-color:  lightgreen;\n",
       "        }    #T_061b227e_ffde_11ea_9dbe_02420a393713row2_col6 {\n",
       "            background-color:  lightgreen;\n",
       "        }</style><table id=\"T_061b227e_ffde_11ea_9dbe_02420a393713\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Регион</th>        <th class=\"col_heading level0 col1\" >Среднее</th>        <th class=\"col_heading level0 col2\" >Медиана</th>        <th class=\"col_heading level0 col3\" >Граница в 2,5%</th>        <th class=\"col_heading level0 col4\" >Доверительный интервал 95% - стат тест</th>        <th class=\"col_heading level0 col5\" >Доверительный интервал 95% - Bootstrap</th>        <th class=\"col_heading level0 col6\" >Риски, доля</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_061b227e_ffde_11ea_9dbe_02420a393713level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row0_col0\" class=\"data row0 col0\" >region_1</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row0_col1\" class=\"data row0 col1\" >0.399575</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row0_col2\" class=\"data row0 col2\" >0.403781</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row0_col3\" class=\"data row0 col3\" >-0.110468</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row0_col4\" class=\"data row0 col4\" >(-1.5627659830792189, 2.361916939187678)</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row0_col5\" class=\"data row0 col5\" >(-0.11046789533197227, 0.8974603277178793)</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row0_col6\" class=\"data row0 col6\" >0.06</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_061b227e_ffde_11ea_9dbe_02420a393713level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row1_col0\" class=\"data row1 col0\" >region_2</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row1_col1\" class=\"data row1 col1\" >0.452049</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row1_col2\" class=\"data row1 col2\" >0.448141</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row1_col3\" class=\"data row1 col3\" >0.0616845</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row1_col4\" class=\"data row1 col4\" >(-1.5102925702269903, 2.414390352039907)</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row1_col5\" class=\"data row1 col5\" >(0.061684479654226736, 0.8453401775073849)</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row1_col6\" class=\"data row1 col6\" >0.015</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_061b227e_ffde_11ea_9dbe_02420a393713level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row2_col0\" class=\"data row2 col0\" >region_3</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row2_col1\" class=\"data row2 col1\" >0.37501</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row2_col2\" class=\"data row2 col2\" >0.372647</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row2_col3\" class=\"data row2 col3\" >-0.144767</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row2_col4\" class=\"data row2 col4\" >(-1.5102925702269903, 2.414390352039907)</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row2_col5\" class=\"data row2 col5\" >(-0.14476672730489268, 0.8883904035342213)</td>\n",
       "                        <td id=\"T_061b227e_ffde_11ea_9dbe_02420a393713row2_col6\" class=\"data row2 col6\" >0.08</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fae8554f050>"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_profits = pd.DataFrame(data = revenue_report, columns =['Регион','Среднее', \\\n",
    "                'Медиана','Граница в 2,5%', 'Доверительный интервал 95% - стат тест',\\\n",
    "                'Доверительный интервал 95% - Bootstrap', 'Риски, доля'])\n",
    "\n",
    "df_profits.style.highlight_max(color='lightgreen', axis = 0)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Новый вывод\n",
    "\n",
    "**Исходя из представленных данных - наилучший регион для разработки: регион № 1, несмотря на то что лучшая средняя выручка и риски у региона № 2.**\n",
    "\n",
    "\n",
    "Результаты в отношении региона № 2 детерменированы характером представленных данных, поэтому не могут приниматься в расчет до валидации у поставщика данных. (Неявная категоризация/стандартизация)\n",
    "\n",
    "\n",
    "Принимая во внимание близость качества моделей, предсказавших запасы в регионе № 1 и 3: средняя выручка в регионе № 1 (0.40 млрд. рублей) больше выручки в регионе № 3 (0.38), равно как и 2.5%-квантиль (-0.11 млрд. рублей в первом реионе против -0.17 млрд. рублей в регионе № 3).\n",
    "\n",
    "\n",
    "1. Средняя выручка по регионам:\n",
    "\n",
    "    - регион № 1: 0.40 млрд. рублей,\n",
    "    - регион № 2: 0.45 млрд. рублей,\n",
    "    - регион № 3: 0.37 млрд. рублей.\n",
    "    \n",
    "    \n",
    "2. Медиана (практически совпадает со средним):\n",
    "\n",
    "    - регион № 1: 0.40 млрд. рублей,\n",
    "    - регион № 2: 0.45 млрд. рублей,\n",
    "    - регион № 3: 0.37 млрд. рублей.\n",
    "\n",
    "\n",
    "3. Исследование показало, что исходя из представленных данных, риск убытков отсутвует, 2.5%-квантиль для регионов:\n",
    "\n",
    "    - регион № 1: -0.11 млрд. рублей, верхний - 0.89 млрд. рублей\n",
    "    - регион № 2: 0.06 млрд. рублей, верхний - 0.84 млрд. рублей\n",
    "    - регион № 3: -0.14 млрд. рублей, верхний - 0.88 млрд. рублей\n",
    "    \n",
    "    \n",
    "    \n",
    "    \n",
    "4. Всего лишь с 5% вероятностью мы можем утверждать, что наши результаты были получены случайно и реальные показатели оценок прибыли отличаются серьезно от полученных на модели. Если это истинно, то уход в убытки не должен превысить 1,5 млрд (для всех регионов)\n",
    "\n",
    "\n",
    "5. Процент рисков отбора неудачных скважин:\n",
    "    - регион № 1: 6%\n",
    "    - регион № 2: 1.5%\n",
    "    - регион № 3: 8%\n",
    "    \n",
    "\n",
    "\n",
    "\n",
    "Исходя из результатов исследования - следует принять во внимание, что модель малопригодна для настоящего прогнозирования данных, а значит требуется уточнение факторных фичей и апробация других моделей (например, регрессионного дерева) [хотя CatBoost показал себя примерно также(((]\n"
   ]
  }
 ],
 "metadata": {
  "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}