decoding_vin_codes.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4ebc8a6a-af22-4e7d-a814-c25685c4cf2f",
   "metadata": {},
   "source": [
    "# Расшифровка VIN-кодов автомобилей при помощи регулярных выражений\n",
    "\n",
    "**Описание проекта**\n",
    "Изучение информации, зашифрованной в VIN-кодах с целью прогнозирования цены.\n",
    "\n",
    "**Цель проекта**\n",
    "- Составление прогноза цен для автомобилей FORD на основании информации скрытой в VIN-кодах.\n",
    "\n",
    "**Задачи**\n",
    "- изучить структуру VIN-кода: какие информационные компоненты он содержит (регион, производитель, характеристики автомобиля и т. д.);\n",
    "- составлять регулярные выражения для поиска и извлечения конкретных частей VIN-кода, таких как производитель, модель и год выпуска;\n",
    "- распарсить текстовый массив данных с VIN-кодами и ценами;\n",
    "- решить задачу предсказания цены по VIN-коду.\n",
    "\n",
    "**Метрики**\n",
    "- `MSE`, `MAE`, `MAPE`, `R2`\n",
    "\n",
    "**Ход исследования**\n",
    "- Ознакомление с данными;\n",
    "- ознакомление со способами кодирования данных в VIN-кодах;\n",
    "- составление паттерна для парсинга VIN-кодов;\n",
    "- проверка работоспособности паттерна;\n",
    "- парсинг VIN-кодов;\n",
    "- подготовка данных для модели;\n",
    "- обучение baseline модели;\n",
    "- обучение других моделей;\n",
    "- выбор лучшей модели;\n",
    "- проверка важности признаков;\n",
    "- заключение по исследованию.\n",
    "\n",
    "**Справочная информация**\n",
    "- https://docs.python.org/3/library/re.html\n",
    "- https://regex101.com\n",
    "- https://en.wikibooks.org/wiki/Vehicle_Identification_Numbers_(VIN_codes)#Table_of_Contents\n",
    "- https://en.wikibooks.org/wiki/Vehicle_Identification_Numbers_(VIN_codes)/World_Manufacturer_Identifier_(WMI)\n",
    "- https://en.wikibooks.org/wiki/Vehicle_Identification_Numbers_(VIN_codes)/Printable_version#Model#\n",
    "- https://en.wikipedia.org/wiki/List_of_Ford_factories\n",
    "\n",
    "**Форматы VIN-кодов**\n",
    "\n",
    "![варианты vin кодов](data/photo_5269222301897973406_x.jpg)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4cab216-63c3-4f01-bfd8-097a849f314b",
   "metadata": {},
   "source": [
    "Настройка среды"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "5f2d06d6-8e7b-4e6d-ae02-4137e3579c20",
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install pandas\n",
    "# !pip install seaborn\n",
    "# !pip install mlflow\n",
    "# !pip install psycopg2-binary\n",
    "# !pip install python-dotenv\n",
    "# !pip install phik\n",
    "# !pip install catboost\n",
    "# !pip install shap\n",
    "!pip freeze > requirements.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "45a8c5a6-44c9-40c5-9c39-09c6e5fb5aa3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "import os\n",
    "import pandas as pd\n",
    "import psycopg2 as psycopg\n",
    "\n",
    "from dotenv import load_dotenv"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "97e21132-ffa0-49b0-8c22-d3c434966b87",
   "metadata": {},
   "source": [
    "Установка глобальных переменных"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6638043a-51af-4ecd-bd0c-bfdfce84bbe2",
   "metadata": {},
   "outputs": [],
   "source": [
    "RANDOM_STATE = 42\n",
    "ASSETS_DIR = 'assets'\n",
    "DATA_PATH = 'data/vin_ford_train.txt'\n",
    "TRACKING_SERVER_PORT = 5000\n",
    "TRACKING_SERVER_HOST = \"127.0.0.1\"\n",
    "EXPERIMENT_NAME = 'vin_codes'\n",
    "RUN_NAME = ['baseline_0_registry', 'lr_0_registry', 'rf_0_registry', 'cb_0_registry']\n",
    "REGISTRY_MODEL_NAME = ['baseline', 'linear_regression', 'rendom_forest', 'catboost']\n",
    "\n",
    "load_dotenv()\n",
    "connection = {\"sslmode\": \"require\",\n",
    "              \"target_session_attrs\": \"read-write\"}\n",
    "postgres_credentials = {\"host\": os.getenv(\"DB_HOST\"),\n",
    "                        \"port\": os.getenv(\"DB_PORT\"),\n",
    "                        \"dbname\": os.getenv(\"DB_NAME\"),\n",
    "                        \"user\": os.getenv(\"DB_USERNAME\"),\n",
    "                        \"password\": os.getenv(\"DB_PASSWORD\")}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "843d6222-0599-425f-ab79-e6e77001ed38",
   "metadata": {},
   "source": [
    "Импорт класса для проведения исследования"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "07b2ab2c-3705-475d-9af9-431f58f59269",
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "from research_class import DatasetExplorer"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6fda7911-a9b3-4bf5-bbb6-3b30ae1460d4",
   "metadata": {},
   "source": [
    "## Знакомство с данными"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3b43a6c9-af48-4c4c-b709-73f54f3ce6cc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2FMDK3JC4BBA41556:12500]мой[3FA6P0H75ER208976:14500]дядя [3FAHP08Z17R268380:1300][1FMCU0GX3DUC59421:15200]самых [1FAHP2E89DG119368:11600] верных[1FM5K8D82DGA16143:26400][1FTFW1CFXCFC23663:14000][2FMDK3JC6CBA90002:19500][1FTFW1CT8DKD06233:24400][1FMZU64W13UC28949:2900][2FMDK3JC9DBB30736:23500][1FMCU9D76CKC49193:12700][1FMCU9EG2CKA90624:20700][1FMCU0JX7EUA28208:21300][2FMGK5D85EBD31213:26700][2FMDK\n"
     ]
    }
   ],
   "source": [
    "with open (DATA_PATH, 'r') as f:\n",
    "    data = f.read()\n",
    "print(data[:400])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b393b99b-8f26-4273-99f9-389425152c08",
   "metadata": {},
   "source": [
    "### Выводы и наблюдения\n",
    "- исходя из условий задачи, необходимый вариант VIN-кодов - `American VIN format`;\n",
    "- в предоставленном файле VIN-коды находятся в неудобном для исследования формате:\n",
    "  - необходима очистка и преобразование;\n",
    "- помимо VIN-кодов в файле предоставлены цены."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6730ee6-ef55-491a-809d-fc32c81ed1a8",
   "metadata": {},
   "source": [
    "## Подготовка паттерна\n",
    "В соответствии с правилами составления VIN-кодов для американских автомобилей:\n",
    "- современные VIN-коды состоят из 17 символов, которые не содержат букв I, O или Q;\n",
    "- `check digit` - 0-9 or X;\n",
    "- буквы U и Z и цифра 0 не используются для кода года;\n",
    "- позиции 13–17 должны быть цифрами, тогда как позиция 12 может быть буквой или цифрой"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "721eee61-f15e-460b-8f31-88c799048d25",
   "metadata": {},
   "source": [
    "### Разрешённые в VIN-кодах символы"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d10056f4-aea0-4387-83e3-3ff10eb069c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'ABCDEFGHJKLMNPRSTUVWXYZ1234567890'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "allowed = r'ABCDEFGHJKLMNPRSTUVWXYZ1234567890'\n",
    "allowed"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d17dfb3-a2b4-4ef5-af8e-456c749c749e",
   "metadata": {},
   "source": [
    "### Тестовый образец"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "df4444b6-f814-414e-8303-447251e95187",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1FT7X2B6XCEB34867 1FTNF1CF4EKD18628 1GNKRFED3CJ319802 5N1BA08D49N601792 5TFHY5F19CX236781 JN1CV6AR7BM356524 ZFF65LJA9A0170669'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vin_sample = '1FT7X2B6XCEB34867 1FTNF1CF4EKD18628 1GNKRFED3CJ319802 5N1BA08D49N601792 5TFHY5F19CX236781 JN1CV6AR7BM356524 ZFF65LJA9A0170669'\n",
    "vin_sample"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bfddfed-d313-419b-b12a-daa97a4fa3f0",
   "metadata": {},
   "source": [
    "### Северо-Американский формат VIN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c189ecf5-de5a-4abd-af4c-21d6f6d79adb",
   "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>wmi</th>\n",
       "      <th>brake</th>\n",
       "      <th>body</th>\n",
       "      <th>engine</th>\n",
       "      <th>check_digit</th>\n",
       "      <th>year</th>\n",
       "      <th>plant</th>\n",
       "      <th>snum</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1FT</td>\n",
       "      <td>7</td>\n",
       "      <td>X2B</td>\n",
       "      <td>6</td>\n",
       "      <td>X</td>\n",
       "      <td>C</td>\n",
       "      <td>E</td>\n",
       "      <td>B34867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1FT</td>\n",
       "      <td>N</td>\n",
       "      <td>F1C</td>\n",
       "      <td>F</td>\n",
       "      <td>4</td>\n",
       "      <td>E</td>\n",
       "      <td>K</td>\n",
       "      <td>D18628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1GN</td>\n",
       "      <td>K</td>\n",
       "      <td>RFE</td>\n",
       "      <td>D</td>\n",
       "      <td>3</td>\n",
       "      <td>C</td>\n",
       "      <td>J</td>\n",
       "      <td>319802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5N1</td>\n",
       "      <td>B</td>\n",
       "      <td>A08</td>\n",
       "      <td>D</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>N</td>\n",
       "      <td>601792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5TF</td>\n",
       "      <td>H</td>\n",
       "      <td>Y5F</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>C</td>\n",
       "      <td>X</td>\n",
       "      <td>236781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>JN1</td>\n",
       "      <td>C</td>\n",
       "      <td>V6A</td>\n",
       "      <td>R</td>\n",
       "      <td>7</td>\n",
       "      <td>B</td>\n",
       "      <td>M</td>\n",
       "      <td>356524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ZFF</td>\n",
       "      <td>6</td>\n",
       "      <td>5LJ</td>\n",
       "      <td>A</td>\n",
       "      <td>9</td>\n",
       "      <td>A</td>\n",
       "      <td>0</td>\n",
       "      <td>170669</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   wmi brake body engine check_digit year plant    snum\n",
       "0  1FT     7  X2B      6           X    C     E  B34867\n",
       "1  1FT     N  F1C      F           4    E     K  D18628\n",
       "2  1GN     K  RFE      D           3    C     J  319802\n",
       "3  5N1     B  A08      D           4    9     N  601792\n",
       "4  5TF     H  Y5F      1           9    C     X  236781\n",
       "5  JN1     C  V6A      R           7    B     M  356524\n",
       "6  ZFF     6  5LJ      A           9    A     0  170669"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wmi_pattern = r'(?P<wmi>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{3})'  # pos. 1-3\n",
    "brake_pattern = r'(?P<brake>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])' # pos. 4\n",
    "body_pattern = r'(?P<body>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{3})'  # pos. 5-7\n",
    "engine_pattern = r'(?P<engine>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])'  # pos. 8\n",
    "check_digit_pattern = r'(?P<check_digit>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])'  # pos. 9\n",
    "year_pattern = r'(?P<year>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])'.replace('U', '').replace('Z', '').replace('0', '')  # pos. 10\n",
    "plant_pattern = r'(?P<plant>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])'  # pos. 11\n",
    "snum_pattern = r'(?P<snum>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]\\d{5})'  # pos. 12-17\n",
    "pattern = f'{wmi_pattern}{brake_pattern}{body_pattern}{engine_pattern}{check_digit_pattern}{year_pattern}{plant_pattern}{snum_pattern}'\n",
    "pd.DataFrame([x.groupdict() for x in re.compile(pattern).finditer(vin_sample)])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d487366-13e6-4542-8a18-f665cdc35f2e",
   "metadata": {},
   "source": [
    "### Извлечение производителя из VIN-кода"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c3c8fbdd-3300-4d83-af03-70cf33187970",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['1FT', '1FT', '1GN', '5N1', '5TF', 'JN1', 'ZFF']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pattern = r'''\n",
    "    \\b(?P<wmi>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{3})  # три первые элемента слова, принадлежащие такому набору\n",
    "    [ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{14}\\b          # исключение 14-ти последующих элементов слова, принадлежащих набору\n",
    "'''\n",
    "re.compile(pattern, re.X).findall(vin_sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4c55afd-37bf-4c5f-9c3e-933862eb8c9b",
   "metadata": {},
   "source": [
    "### Поиск всех VIN-кодов определенной марки автомобиля на примере Ferrari\n",
    "- Вин-коды Феррари ZDF, ZFF, ZSG"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2d800abd-8814-425c-a163-f71e18ebd33e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['ZFF65LJA9A0170669']"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ferrari_wmi = ['ZDF', 'ZFF', 'ZSG']\n",
    "wmi_pattern = '|'.join(ferrari_wmi)\n",
    "pattern = re.compile(f'\\\\b(?:{wmi_pattern})[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{{14}}\\\\b')\n",
    "pattern.findall(vin_sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f04b4f6-d374-4580-84f2-97d8f6fb8265",
   "metadata": {},
   "source": [
    "### Поиск VIN-кодов с определенным регионом\n",
    "Северная Америка как регион (США, Канада, Мексика)\n",
    "\n",
    "    1, 4, 5 = United States\n",
    "    2 = Canada\n",
    "    3 = Mexico\n",
    "    7F-70 = United States"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c196773-0de9-4a1b-b9ee-bda57f41fa58",
   "metadata": {},
   "source": [
    "Североамериканские машины"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3f57886c-3865-47be-84a8-3e6f246a5a2c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['1FT7X2B6XCEB34867',\n",
       " '1FTNF1CF4EKD18628',\n",
       " '1GNKRFED3CJ319802',\n",
       " '5N1BA08D49N601792',\n",
       " '5TFHY5F19CX236781']"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pattern = r'''\n",
    "    \\b(?:1|4|5|7F|70)[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{16}\\b\n",
    "    |\\b2[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{16}\\b\n",
    "    |\\b3[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{16}\\b\n",
    "    '''\n",
    "re.compile(pattern, re.X).findall(vin_sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e376116f-8700-412b-a963-f79f148afcbc",
   "metadata": {},
   "source": [
    "Машины из любых других регионов (не Сев.Америка)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2bb717e2-6f42-4d45-b438-994498699bd5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['JN1CV6AR7BM356524', 'ZFF65LJA9A0170669']"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pattern = re.compile(r'\\b(?!(?:1|4|5|7F|70|2|3)[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{16})[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{17}\\b')\n",
    "pattern.findall(vin_sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2d6d5a5-7026-4262-a0c8-c82a5ae2c8d8",
   "metadata": {},
   "source": [
    "### Извлечение года из VIN-кода только североамериканских машин\n",
    "- В североамериканском вине год 10й знак\n",
    "- в кодировке года не используются знаки U и Z в дополнение к запрещенным I, O, Q"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "8d6a4c37-17f3-4dd1-9e73-15f3941f8890",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['C', 'E', 'C', '9', 'C']"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pattern = r'''\n",
    "    \\b(?:1|4|5|7F|70)[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{16}\\b\n",
    "    |\\b2[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{16}\\b\n",
    "    |\\b3[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{16}\\b\n",
    "'''\n",
    "[x[9] for x in re.compile(pattern, re.X).findall(vin_sample)]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "937c30ea-9fdf-4d81-a323-5ff920d48d9f",
   "metadata": {},
   "source": [
    "Для проверки выведем весь датафрейм компонентов VIN кода"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "84b53da7-3597-4f47-a9b6-bdd9549fdc55",
   "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>wmi</th>\n",
       "      <th>brake</th>\n",
       "      <th>body</th>\n",
       "      <th>engine</th>\n",
       "      <th>check_digit</th>\n",
       "      <th>year</th>\n",
       "      <th>plant</th>\n",
       "      <th>snum</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1FT</td>\n",
       "      <td>7</td>\n",
       "      <td>X2B</td>\n",
       "      <td>6</td>\n",
       "      <td>X</td>\n",
       "      <td>C</td>\n",
       "      <td>E</td>\n",
       "      <td>B34867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1FT</td>\n",
       "      <td>N</td>\n",
       "      <td>F1C</td>\n",
       "      <td>F</td>\n",
       "      <td>4</td>\n",
       "      <td>E</td>\n",
       "      <td>K</td>\n",
       "      <td>D18628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1GN</td>\n",
       "      <td>K</td>\n",
       "      <td>RFE</td>\n",
       "      <td>D</td>\n",
       "      <td>3</td>\n",
       "      <td>C</td>\n",
       "      <td>J</td>\n",
       "      <td>319802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5N1</td>\n",
       "      <td>B</td>\n",
       "      <td>A08</td>\n",
       "      <td>D</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>N</td>\n",
       "      <td>601792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5TF</td>\n",
       "      <td>H</td>\n",
       "      <td>Y5F</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>C</td>\n",
       "      <td>X</td>\n",
       "      <td>236781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>JN1</td>\n",
       "      <td>C</td>\n",
       "      <td>V6A</td>\n",
       "      <td>R</td>\n",
       "      <td>7</td>\n",
       "      <td>B</td>\n",
       "      <td>M</td>\n",
       "      <td>356524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ZFF</td>\n",
       "      <td>6</td>\n",
       "      <td>5LJ</td>\n",
       "      <td>A</td>\n",
       "      <td>9</td>\n",
       "      <td>A</td>\n",
       "      <td>0</td>\n",
       "      <td>170669</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   wmi brake body engine check_digit year plant    snum\n",
       "0  1FT     7  X2B      6           X    C     E  B34867\n",
       "1  1FT     N  F1C      F           4    E     K  D18628\n",
       "2  1GN     K  RFE      D           3    C     J  319802\n",
       "3  5N1     B  A08      D           4    9     N  601792\n",
       "4  5TF     H  Y5F      1           9    C     X  236781\n",
       "5  JN1     C  V6A      R           7    B     M  356524\n",
       "6  ZFF     6  5LJ      A           9    A     0  170669"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wmi_pattern = r'(?P<wmi>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{3})'  # pos. 1-3\n",
    "brake_pattern = r'(?P<brake>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])' # pos. 4\n",
    "body_pattern = r'(?P<body>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{3})'  # pos. 5-7\n",
    "engine_pattern = r'(?P<engine>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])'  # pos. 8\n",
    "check_digit_pattern = r'(?P<check_digit>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])'  # pos. 9\n",
    "year_pattern = r'(?P<year>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])'.replace('U', '').replace('Z', '').replace('0', '')  # pos. 10\n",
    "plant_pattern = r'(?P<plant>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890])'  # pos. 11\n",
    "snum_pattern = r'(?P<snum>[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]\\d{5})'  # pos. 12-17\n",
    "pattern = f'{wmi_pattern}{brake_pattern}{body_pattern}{engine_pattern}{check_digit_pattern}{year_pattern}{plant_pattern}{snum_pattern}'\n",
    "pd.DataFrame([x.groupdict() for x in re.compile(pattern).finditer(vin_sample)])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90051d8b-3af5-42a0-9d0b-ae64a9ecf03a",
   "metadata": {},
   "source": [
    "### Проверка корректности формата VIN-кода\n",
    "\n",
    "- Вин-код состоит из 17 знаков – цифр и букв, кроме 'I', 'O', 'Q'\n",
    "- Последние три знака – всегда цифры\n",
    "\n",
    "Создадим набор винов, где только один вин валидный и остальные ошибочные\n",
    "- отрежем один знак\n",
    "- добавим лишний знак\n",
    "- заменим буквой последнюю цифру в номере\n",
    "- заменим разрешенный знак на неразрешенный"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "3e4f3d3b-b3ca-47df-a218-399a6e797067",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1FT7X2B6XCEB34867 1FTNF1CF4EKD18628 1GNKRFED3CJ319802 5N1BA08D49N601792 5TFHY5F19CX236781 JN1CV6AR7BM356524 ZFF65LJA9A0170669 3LN6L2GKXDR811732 BAAM3334XKC59086 1N4BL11D84C1859519 1C4RJEBG5EC48287Q QN1AB7AP5EL654699'"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "error_sample = vin_sample + ' 3LN6L2GKXDR811732 BAAM3334XKC59086 1N4BL11D84C1859519 1C4RJEBG5EC48287Q QN1AB7AP5EL654699'\n",
    "error_sample"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2d35aae8-a1e2-4748-b4c4-163dcda81fc5",
   "metadata": {},
   "source": [
    "Валидные VIN-коды"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "317c0ef3-28a4-41cb-8066-79d12a6d6c60",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['1FT7X2B6XCEB34867',\n",
       " '1FTNF1CF4EKD18628',\n",
       " '1GNKRFED3CJ319802',\n",
       " '5N1BA08D49N601792',\n",
       " '5TFHY5F19CX236781',\n",
       " 'JN1CV6AR7BM356524',\n",
       " 'ZFF65LJA9A0170669',\n",
       " '3LN6L2GKXDR811732']"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pattern = re.compile(r'\\b[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{14}\\d{3}\\b')\n",
    "pattern.findall(error_sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd657409-b572-4714-b30d-4fff4c92126e",
   "metadata": {},
   "source": [
    "Не валидные VIN-коды"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b4e0e067-c2ca-477b-a6a8-46b47e00c5db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['BAAM3334XKC59086',\n",
       " '1N4BL11D84C1859519',\n",
       " '1C4RJEBG5EC48287Q',\n",
       " 'QN1AB7AP5EL654699']"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pattern = re.compile(r'(?!\\b[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{14}\\d{3}\\b)\\b\\w+\\b')\n",
    "\n",
    "pattern.findall(error_sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "719b94dc-e3f6-4ef5-8dfe-ac23ad3d6668",
   "metadata": {},
   "source": [
    "### Автомобили Ford"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "43989b59-65ef-48b3-a792-c52b69a09201",
   "metadata": {},
   "outputs": [],
   "source": [
    "ford_wmi = {\n",
    "    'AFA': 'Ford Motor Company of Southern Africa & Samcor',\n",
    "    'AFB': 'Mazda BT-50 made by Ford Motor Company of Southern Africa',\n",
    "    'JC0': 'Ford brand cars made by Mazda',\n",
    "    'JC2': 'Ford Courier made by Mazda',\n",
    "    'KNJ': 'Ford Festiva & Aspire made by Kia',\n",
    "    'LJX': 'JMC Ford',\n",
    "    'LVS': 'Changan Ford & Changan Ford Mazda',\n",
    "    'MAJ': 'Ford India',\n",
    "    'MNB': 'Ford Thailand',\n",
    "    'NM0': 'Ford Otosan',\n",
    "    'PE1': 'Ford Motor Company Philippines',\n",
    "    'PE3': 'Mazda Philippines made by Ford Motor Company Philippines',\n",
    "    'PR8': 'Ford',\n",
    "    'LFA': 'Ford Lio Ho Motor Co Ltd. old designation',\n",
    "    'RHA': 'Ford Lio Ho Motor Co Ltd. new designation',\n",
    "    'RL0': 'Ford Vietnam',\n",
    "    'SBC': 'Iveco Ford Truck',\n",
    "    'SFA': 'Ford UK',\n",
    "    'VSK': 'Nissan Motor Iberica SA, Nissan passenger car/MPV/van/SUV/pickup & Ford Maverick 1993–1999',\n",
    "    'VS6': 'Ford Spain',\n",
    "    'WF0': 'Ford Germany',\n",
    "    'X9F': 'Ford Motor Company ZAO',\n",
    "    'Z6F': 'Ford Sollers (Russia)',\n",
    "    '1FA': 'Ford car',\n",
    "    '1FB': 'Ford \"bus\" (van with more than 3 rows of seats)',\n",
    "    '1FC': 'Ford stripped chassis made by Ford',\n",
    "    '1FD': 'Ford incomplete vehicle',\n",
    "    '1FM': 'Ford MPV/SUV',\n",
    "    '1FT': 'Ford truck',\n",
    "    '1F1': 'Ford SUV - Limousine (through 2009)',\n",
    "    '1F6': 'Ford stripped chassis made by Detroit Chassis LLC',\n",
    "    '1ZV': 'Ford made by AutoAlliance International',\n",
    "    '2FA': 'Ford car',\n",
    "    '2FM': 'Ford MPV/SUV',\n",
    "    '2FT': 'Ford truck',\n",
    "    '3FA': 'Ford car',\n",
    "    '3FC': 'Ford stripped chassis made by Ford & IMMSA',\n",
    "    '3FE': 'Ford Mexico',\n",
    "    '3FM': 'Ford MPV/SUV',\n",
    "    '3FN': 'Ford F-650/F-750 made by Blue Diamond Truck Co. (truck)',\n",
    "    '3FR': 'Ford F-650/F-750 made by Blue Diamond Truck Co. (incomplete vehicle)',\n",
    "    '3FT': 'Ford truck',\n",
    "    '4F2': 'Mazda SUV made by Ford',\n",
    "    '4F4': 'Mazda truck made by Ford',\n",
    "    '4N2': 'Nissan Quest made by Ford',\n",
    "    '5LD': 'Ford & Lincoln incomplete vehicle – limousine (2010–2014)',\n",
    "    '6F1': 'Ford',\n",
    "    '6FP': 'Ford Australia',\n",
    "    '7A5': 'Ford New Zealand',\n",
    "    '8AF': 'Ford Argentina',\n",
    "    '9BF': 'Ford Brazil',\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7c0bfe1d-8546-4b11-af1a-0046ccc79224",
   "metadata": {},
   "outputs": [],
   "source": [
    "ford_groups = {\n",
    "    'P': 'All Ford, Mercury, Merkur, and Lincoln passenger cars 1981-1986, Ford-brand passenger cars made in North America at a factory owned 100% by Ford Motor Co. 1987-2009',\n",
    "    'M': 'Mercury & Lincoln passenger cars made in North America at a factory owned 100% by Ford Motor Co. 1987-2009. Mercury passenger cars 2010-2011',\n",
    "    'T': 'Ford, Mercury, and Merkur passenger cars made outside North America or made in North America at a factory not owned 100% by Ford Motor Co. 1987-2009 (Ford Festiva, Aspire, Probe, & \\'05-\\'09 Mustang, Mercury Capri \\'91-\\'94, & Cougar \\'99-\\'02, Merkur XR4Ti \\'87-\\'89 and Scorpio \\'88-\\'89)',\n",
    "    'P': 'Ford-brand passenger cars 2010-',\n",
    "    'L': 'Lincoln passenger cars 2010-2020',\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14c94dfe-dd24-47d9-850e-8193db71d671",
   "metadata": {},
   "source": [
    "Отбор автомобилей FORD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "0692d93c-4fa8-4632-96eb-613e764f9065",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['1FT7X2B6XCEB34867', '1FTNF1CF4EKD18628']"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ford_wmi_pattern = '|'.join(ford_wmi.keys())\n",
    "pattern = f'\\\\b(?:{ford_wmi_pattern})[ABCDEFGHJKLMNPRSTUVWXYZ1234567890]{{14}}\\\\b'\n",
    "re.compile(pattern).findall(vin_sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afa5e888-3dec-41c5-8cee-cbb0d168d69f",
   "metadata": {},
   "source": [
    "### Выводы и наблюдения\n",
    "- проведено исследование для подготовки скрипта, извлекающего необходимые VIN-коды;\n",
    "- подготовлен и протестирован код для парсинга."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a75f280f-8459-4d8f-88b6-c6e79df937b7",
   "metadata": {},
   "source": [
    "## Подготовка датафрейма и первичный анализ данных"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b5e5709-a4e2-4e31-8833-d58a4d55d56e",
   "metadata": {},
   "source": [
    "Создание экземпляра исследовательского класса"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "32cbca7c-1235-451d-9503-f567c8cdec4a",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_explorer = DatasetExplorer(DATA_PATH=DATA_PATH)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8655ea96-1e03-4450-8dc2-6c9b545ec31e",
   "metadata": {},
   "source": [
    "Создание датафрейма, проведение анализа"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e2274e3f-07e2-4265-9dac-41c1a326e32a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Верхние пять строк датафрейма:\n",
      "\n"
     ]
    },
    {
     "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>wmi</th>\n",
       "      <th>brake</th>\n",
       "      <th>body</th>\n",
       "      <th>engine</th>\n",
       "      <th>check_digit</th>\n",
       "      <th>year</th>\n",
       "      <th>plant</th>\n",
       "      <th>snum</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2FM</td>\n",
       "      <td>D</td>\n",
       "      <td>K3J</td>\n",
       "      <td>C</td>\n",
       "      <td>4</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>A41556</td>\n",
       "      <td>12500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3FA</td>\n",
       "      <td>6</td>\n",
       "      <td>P0H</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>E</td>\n",
       "      <td>R</td>\n",
       "      <td>208976</td>\n",
       "      <td>14500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3FA</td>\n",
       "      <td>H</td>\n",
       "      <td>P08</td>\n",
       "      <td>Z</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>R</td>\n",
       "      <td>268380</td>\n",
       "      <td>1300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1FM</td>\n",
       "      <td>C</td>\n",
       "      <td>U0G</td>\n",
       "      <td>X</td>\n",
       "      <td>3</td>\n",
       "      <td>D</td>\n",
       "      <td>U</td>\n",
       "      <td>C59421</td>\n",
       "      <td>15200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1FA</td>\n",
       "      <td>H</td>\n",
       "      <td>P2E</td>\n",
       "      <td>8</td>\n",
       "      <td>9</td>\n",
       "      <td>D</td>\n",
       "      <td>G</td>\n",
       "      <td>119368</td>\n",
       "      <td>11600</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   wmi brake body engine check_digit year plant    snum  price\n",
       "0  2FM     D  K3J      C           4    B     B  A41556  12500\n",
       "1  3FA     6  P0H      7           5    E     R  208976  14500\n",
       "2  3FA     H  P08      Z           1    7     R  268380   1300\n",
       "3  1FM     C  U0G      X           3    D     U  C59421  15200\n",
       "4  1FA     H  P2E      8           9    D     G  119368  11600"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Общая информация по датафрейму:\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 71884 entries, 0 to 71883\n",
      "Data columns (total 9 columns):\n",
      " #   Column       Non-Null Count  Dtype \n",
      "---  ------       --------------  ----- \n",
      " 0   wmi          71884 non-null  object\n",
      " 1   brake        71884 non-null  object\n",
      " 2   body         71884 non-null  object\n",
      " 3   engine       71884 non-null  object\n",
      " 4   check_digit  71884 non-null  object\n",
      " 5   year         71884 non-null  object\n",
      " 6   plant        71884 non-null  object\n",
      " 7   snum         71884 non-null  object\n",
      " 8   price        71884 non-null  int64 \n",
      "dtypes: int64(1), object(8)\n",
      "memory usage: 4.9+ MB\n",
      "\n",
      "Количество дубликатов: 0\n",
      "\n",
      "Количество уникальных значений в каждом признаке:\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "wmi               14\n",
       "brake             26\n",
       "body             365\n",
       "engine            32\n",
       "check_digit       11\n",
       "year              25\n",
       "plant             21\n",
       "snum           68215\n",
       "price            643\n",
       "dtype: int64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Анализ целевой переменной:\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "count    71884.000000\n",
       "mean     14321.449377\n",
       "std       7827.057238\n",
       "min       1000.000000\n",
       "25%       8800.000000\n",
       "50%      13500.000000\n",
       "75%      19500.000000\n",
       "max      36500.000000\n",
       "Name: price, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = data_explorer.explore_dataset(assets_dir=ASSETS_DIR)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84a53ac7-876c-4e7d-b629-9daad1dba913",
   "metadata": {},
   "source": [
    "### Выводы и наблюдения\n",
    "- на основании предоставленного файла создан датафрейм,\n",
    "  <br>состоящий из частей VIN-кодов и цен автомобилей;\n",
    "- общее количество записей в датафрейме - 71884;\n",
    "- дубликатов не обнаружено;\n",
    "- почти все признаки относятся к категориальным;\n",
    "- цены на автомобили варьируются от 1000 до 36500 единиц,\n",
    "  <br>с большинством автомобилей имеющих цену в диапазоне от 8800 до 19500 единиц;\n",
    "- средняя цена находится близко к медианной цене, что может указывать на отсутствие сильных выбросов в данных -\n",
    "  <br>данное предположение подтверждается диаграммой \"ящик с усами\";\n",
    "- практически между всеми признаками имеются зависимости, возможно, над этим придётся поработать."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "20d451e5-2dab-4462-a87c-2fc842b26a3d",
   "metadata": {},
   "source": [
    "## Подготовка данных для обучения модели"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c96ed9d6-9282-4ba2-933e-dd2ea06fef09",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Размер обучающего набора данных: (57507, 8)\n",
      "Размер тестового набора данных: (14377, 8)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = data_explorer.data_preparing(dataset=df, target='price', test_size=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25bc2a70-cd75-44a8-b517-86846f1d4f35",
   "metadata": {},
   "source": [
    "### Выводы и наблюдения\n",
    "- произведено разделение данных на выборки;\n",
    "- произведено кодирование данных для моделей"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7b57f6c-26ba-4e8f-a00c-e834ef5380ce",
   "metadata": {},
   "source": [
    "## Обучение моделей"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dae39573-fa7e-452c-8dd0-d109456be3a2",
   "metadata": {},
   "source": [
    "### Baseline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "507c84d5-36ae-4f3f-b1d2-ca788d6f329f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Результаты кросс-валидации: {'fit_time': 0.021, 'score_time': 0.005, 'test_neg_mean_squared_error': 30475394.373, 'test_r2': 0.502, 'test_neg_mean_absolute_error': 4035.195, 'test_neg_mean_absolute_percentage_error': 0.581}\n"
     ]
    }
   ],
   "source": [
    "params = {'fit_intercept':True,\n",
    "          'copy_X':True,\n",
    "          'n_jobs':-1,\n",
    "          'positive':False}\n",
    "cv_res, model = data_explorer.model_fitting(model_name='Baseline', features=X_train, labels=y_train, params=params, cv=4)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e57c6618-5337-4604-9784-6679f98fe724",
   "metadata": {},
   "source": [
    "### Linear Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "af44e614-e921-4a99-923d-4834ca9e958a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Результаты кросс-валидации: {'test_neg_mean_squared_error': 136230189.195, 'test_r2': -1.228, 'test_neg_mean_absolute_error': 9386.676, 'test_neg_mean_absolute_percentage_error': 1.069}\n",
      "\n",
      "Лучшие гиперпараметры: {'copy_X': True, 'fit_intercept': True, 'n_jobs': -1, 'positive': False}\n"
     ]
    }
   ],
   "source": [
    "params = {'fit_intercept': [True, False],\n",
    "          'copy_X': [True, False],\n",
    "          'n_jobs': [-1],\n",
    "          'positive': [True, False]}\n",
    "cv_res, model, params = data_explorer.model_fitting(model_name='LinearRegression',\n",
    "                                                    features=X_train,\n",
    "                                                    labels=y_train,\n",
    "                                                    params=params,\n",
    "                                                    cv=4)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d1a02c9-4a0f-4fbd-8242-472b8b0483fd",
   "metadata": {},
   "source": [
    "### Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f55cdbcf-45ee-4fae-977b-390865077332",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Результаты кросс-валидации: {'test_neg_mean_squared_error': 16800978.901, 'test_r2': 0.725, 'test_neg_mean_absolute_error': 2907.04, 'test_neg_mean_absolute_percentage_error': 0.337}\n",
      "\n",
      "Лучшие гиперпараметры: {'bootstrap': True, 'criterion': 'friedman_mse', 'max_depth': None, 'n_estimators': 150, 'n_jobs': -1, 'random_state': 42}\n"
     ]
    }
   ],
   "source": [
    "params = {'n_estimators': [100, 150],\n",
    "          'criterion': ['squared_error', 'friedman_mse'],\n",
    "          'max_depth': [None, 3, 5],\n",
    "          'bootstrap': [True, False],\n",
    "          'n_jobs': [-1],\n",
    "          'random_state': [RANDOM_STATE]}\n",
    "cv_res, model, params = data_explorer.model_fitting(model_name='RandomForest',\n",
    "                                                    features=X_train,\n",
    "                                                    labels=y_train,\n",
    "                                                    params=params,\n",
    "                                                    cv=4)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "293da105-7e4d-4184-9504-d09eba640eb9",
   "metadata": {},
   "source": [
    "### CatBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "4cd11e26-940b-4538-8d81-9ace0f7f2bd3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "bestTest = 3863.351282\n",
      "bestIteration = 99\n",
      "\n",
      "0:\tloss: 3863.3512819\tbest: 3863.3512819 (0)\ttotal: 321ms\tremaining: 4.81s\n",
      "\n",
      "bestTest = 3232.654121\n",
      "bestIteration = 99\n",
      "\n",
      "1:\tloss: 3232.6541207\tbest: 3232.6541207 (1)\ttotal: 559ms\tremaining: 3.91s\n",
      "\n",
      "bestTest = 3864.033778\n",
      "bestIteration = 99\n",
      "\n",
      "2:\tloss: 3864.0337776\tbest: 3232.6541207 (1)\ttotal: 827ms\tremaining: 3.58s\n",
      "\n",
      "bestTest = 3243.236396\n",
      "bestIteration = 99\n",
      "\n",
      "3:\tloss: 3243.2363964\tbest: 3232.6541207 (1)\ttotal: 1.08s\tremaining: 3.24s\n",
      "\n",
      "bestTest = 3255.851674\n",
      "bestIteration = 299\n",
      "\n",
      "4:\tloss: 3255.8516745\tbest: 3232.6541207 (1)\ttotal: 1.78s\tremaining: 3.93s\n",
      "\n",
      "bestTest = 2958.849935\n",
      "bestIteration = 299\n",
      "\n",
      "5:\tloss: 2958.8499348\tbest: 2958.8499348 (5)\ttotal: 2.49s\tremaining: 4.14s\n",
      "\n",
      "bestTest = 3267.1275\n",
      "bestIteration = 299\n",
      "\n",
      "6:\tloss: 3267.1275000\tbest: 2958.8499348 (5)\ttotal: 3.2s\tremaining: 4.12s\n",
      "\n",
      "bestTest = 2974.545073\n",
      "bestIteration = 299\n",
      "\n",
      "7:\tloss: 2974.5450731\tbest: 2958.8499348 (5)\ttotal: 3.93s\tremaining: 3.93s\n",
      "\n",
      "bestTest = 3409.78264\n",
      "bestIteration = 99\n",
      "\n",
      "8:\tloss: 3409.7826401\tbest: 2958.8499348 (5)\ttotal: 4.41s\tremaining: 3.43s\n",
      "\n",
      "bestTest = 2953.55534\n",
      "bestIteration = 99\n",
      "\n",
      "9:\tloss: 2953.5553396\tbest: 2953.5553396 (9)\ttotal: 4.91s\tremaining: 2.94s\n",
      "\n",
      "bestTest = 3424.762783\n",
      "bestIteration = 99\n",
      "\n",
      "10:\tloss: 3424.7627831\tbest: 2953.5553396 (9)\ttotal: 5.39s\tremaining: 2.45s\n",
      "\n",
      "bestTest = 2968.779828\n",
      "bestIteration = 99\n",
      "\n",
      "11:\tloss: 2968.7798278\tbest: 2953.5553396 (9)\ttotal: 5.88s\tremaining: 1.96s\n",
      "\n",
      "bestTest = 2969.812674\n",
      "bestIteration = 299\n",
      "\n",
      "12:\tloss: 2969.8126739\tbest: 2953.5553396 (9)\ttotal: 7.3s\tremaining: 1.69s\n",
      "\n",
      "bestTest = 2784.250037\n",
      "bestIteration = 298\n",
      "\n",
      "13:\tloss: 2784.2500371\tbest: 2784.2500371 (13)\ttotal: 8.76s\tremaining: 1.25s\n",
      "\n",
      "bestTest = 2983.139445\n",
      "bestIteration = 299\n",
      "\n",
      "14:\tloss: 2983.1394452\tbest: 2784.2500371 (13)\ttotal: 10.3s\tremaining: 685ms\n",
      "\n",
      "bestTest = 2797.07418\n",
      "bestIteration = 299\n",
      "\n",
      "15:\tloss: 2797.0741804\tbest: 2784.2500371 (13)\ttotal: 11.8s\tremaining: 0us\n",
      "Estimating final quality...\n",
      "Training on fold [0/4]\n",
      "\n",
      "bestTest = 2776.215622\n",
      "bestIteration = 298\n",
      "\n",
      "Training on fold [1/4]\n",
      "\n",
      "bestTest = 2775.362134\n",
      "bestIteration = 299\n",
      "\n",
      "Training on fold [2/4]\n",
      "\n",
      "bestTest = 2838.801107\n",
      "bestIteration = 299\n",
      "\n",
      "Training on fold [3/4]\n",
      "\n",
      "bestTest = 2775.176168\n",
      "bestIteration = 295\n",
      "\n",
      "Лучший результат RMSE: 2791.434\n",
      "Лучшие гиперпараметры: {'depth': 8, 'learning_rate': 0.1, 'l2_leaf_reg': 1, 'iterations': 300}\n"
     ]
    }
   ],
   "source": [
    "params = {'iterations': [100, 300],\n",
    "          'learning_rate': [0.03, 0.1],\n",
    "          'depth': [4, 8],\n",
    "          'l2_leaf_reg': [1, 5]}\n",
    "cv_res, model, params = data_explorer.model_fitting(model_name='CatBoost',\n",
    "                                                    features=X_train,\n",
    "                                                    labels=y_train,\n",
    "                                                    params=params,\n",
    "                                                    cv=4)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4991d766-409d-46ad-827e-e578b9d4a819",
   "metadata": {},
   "source": [
    "Логирование модели и метрик"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "80b91743-6c33-4996-ae8b-2d3abacda657",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/denis/Documents/main_venv/vin_codes/.env_vin/lib/python3.10/site-packages/_distutils_hack/__init__.py:11: UserWarning: Distutils was imported before Setuptools, but importing Setuptools also replaces the `distutils` module in `sys.modules`. This may lead to undesirable behaviors or errors. To avoid these issues, avoid using distutils directly, ensure that setuptools is installed in the traditional way (e.g. not an editable install), and/or make sure that setuptools is always imported before distutils.\n",
      "  warnings.warn(\n",
      "/home/denis/Documents/main_venv/vin_codes/.env_vin/lib/python3.10/site-packages/_distutils_hack/__init__.py:26: UserWarning: Setuptools is replacing distutils.\n",
      "  warnings.warn(\"Setuptools is replacing distutils.\")\n",
      "Successfully registered model 'catboost'.\n",
      "2024/04/22 12:33:41 INFO mlflow.store.model_registry.abstract_store: Waiting up to 60 seconds for model version to finish creation. Model name: catboost, version 1\n",
      "Created version '1' of model 'catboost'.\n"
     ]
    }
   ],
   "source": [
    "metadata = {\n",
    "    'author': 'denis_zeit',\n",
    "    'training_date': '2024-04-22',\n",
    "    'data_used': 'wo_test_data',\n",
    "    'library_versions': {'mlflow': '2.11.3',\n",
    "                         'scikit-learn': '1.4.2',\n",
    "                         'catboost': '1.2.5'}\n",
    "}\n",
    "\n",
    "data_explorer.model_logging(experiment_name=EXPERIMENT_NAME,\n",
    "                            run_name=RUN_NAME[3],\n",
    "                            registry_model=REGISTRY_MODEL_NAME[3],\n",
    "                            params=params,\n",
    "                            metrics=cv_res,\n",
    "                            model=model,\n",
    "                            train_data=X_train,\n",
    "                            train_label=y_train,\n",
    "                            metadata=metadata,\n",
    "                            code_paths=['decoding_vin_codes.ipynb', 'research_class.py'],\n",
    "                            tsh=TRACKING_SERVER_HOST,\n",
    "                            tsp=TRACKING_SERVER_PORT,\n",
    "                            assets_dir=ASSETS_DIR)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6efe729-8daf-4519-bf39-66ed3050896a",
   "metadata": {},
   "source": [
    "### Выводы и наблюдения\n",
    "- в качестве `baseline` принята модель линейной регрессии с гиперпараметрами по умолчанию;\n",
    "- обучена модель линейной регрессии с подбором гиперпараметров по сетке;\n",
    "- обучена модель случайного леса с подбором гиперпараметров по сетке;\n",
    "- обучена модель градиентного бустинга с подбором гиперпараметров по сетке;\n",
    "- все модели с подобранными параметрами и метриками залогированы."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abfbe512-2a4c-4cab-a107-13e681201129",
   "metadata": {},
   "source": [
    "## Выбор лучшей модели"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "57178ce6-74d6-479c-948e-1ffe3cb1b4a8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Находящиеся в базе таблицы:\n"
     ]
    },
    {
     "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>table_name</th>\n",
       "      <th>number_of_rows</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>experiments</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>tags</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>params</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>metrics</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>runs</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>alembic_version</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>experiment_tags</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>latest_metrics</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>registered_models</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>registered_model_tags</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>model_version_tags</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>registered_model_aliases</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>datasets</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>inputs</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>input_tags</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>model_versions</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  table_name  number_of_rows\n",
       "0                experiments               2\n",
       "1                       tags              20\n",
       "2                     params              18\n",
       "3                    metrics              18\n",
       "4                       runs               4\n",
       "5            alembic_version               1\n",
       "6            experiment_tags               0\n",
       "7             latest_metrics              18\n",
       "8          registered_models               4\n",
       "9      registered_model_tags               0\n",
       "10        model_version_tags               0\n",
       "11  registered_model_aliases               0\n",
       "12                  datasets               0\n",
       "13                    inputs               0\n",
       "14                input_tags               0\n",
       "15            model_versions               4"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Модели и их метрики:\n"
     ]
    },
    {
     "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>model_name</th>\n",
       "      <th>version</th>\n",
       "      <th>mse</th>\n",
       "      <th>r2</th>\n",
       "      <th>mae</th>\n",
       "      <th>mape</th>\n",
       "      <th>run_uuid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>catboost</td>\n",
       "      <td>1</td>\n",
       "      <td>2.791434e+03</td>\n",
       "      <td>0.881</td>\n",
       "      <td>1898.950</td>\n",
       "      <td>0.200</td>\n",
       "      <td>7642fc590c9e4622873d6a39a4ec049d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>rendom_forest</td>\n",
       "      <td>1</td>\n",
       "      <td>1.680098e+07</td>\n",
       "      <td>0.725</td>\n",
       "      <td>2907.040</td>\n",
       "      <td>0.337</td>\n",
       "      <td>e6895207e058469fbcf64ed37c2bdce3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>baseline</td>\n",
       "      <td>1</td>\n",
       "      <td>3.047539e+07</td>\n",
       "      <td>0.502</td>\n",
       "      <td>4035.195</td>\n",
       "      <td>0.581</td>\n",
       "      <td>3175f099d12941c3b1185f9675a15939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>linear_regression</td>\n",
       "      <td>1</td>\n",
       "      <td>1.362302e+08</td>\n",
       "      <td>-1.228</td>\n",
       "      <td>9386.676</td>\n",
       "      <td>1.069</td>\n",
       "      <td>35f3d37e332e4b5791a5ad0ff3c0e919</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          model_name  version           mse     r2       mae   mape  \\\n",
       "0           catboost        1  2.791434e+03  0.881  1898.950  0.200   \n",
       "1      rendom_forest        1  1.680098e+07  0.725  2907.040  0.337   \n",
       "2           baseline        1  3.047539e+07  0.502  4035.195  0.581   \n",
       "3  linear_regression        1  1.362302e+08 -1.228  9386.676  1.069   \n",
       "\n",
       "                           run_uuid  \n",
       "0  7642fc590c9e4622873d6a39a4ec049d  \n",
       "1  e6895207e058469fbcf64ed37c2bdce3  \n",
       "2  3175f099d12941c3b1185f9675a15939  \n",
       "3  35f3d37e332e4b5791a5ad0ff3c0e919  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "connection.update(postgres_credentials)\n",
    "with psycopg.connect(**connection) as conn:\n",
    "    with conn.cursor() as cur:\n",
    "        cur.execute(f\"SELECT table_name FROM information_schema.tables WHERE table_schema='public'\")\n",
    "        table_data = []\n",
    "        for table_name in cur.fetchall():\n",
    "            cur.execute(f\"SELECT COUNT(*) FROM {table_name[0]}\")\n",
    "            table_data.append((table_name[0], cur.fetchone()[0]))\n",
    "        print('Находящиеся в базе таблицы:')\n",
    "        display(pd.DataFrame(table_data, columns=['table_name', 'number_of_rows']))\n",
    "\n",
    "        cur.execute(\"\"\"\n",
    "                       SELECT \n",
    "                         COALESCE(model_versions.name, 'baseline') AS model_name,\n",
    "                         model_versions.version,\n",
    "                         MAX(CASE WHEN metrics.key = 'test_neg_mean_squared_error' THEN metrics.value END) AS mse,\n",
    "                         MAX(CASE WHEN metrics.key = 'test_r2' THEN metrics.value END) AS r2,\n",
    "                         MAX(CASE WHEN metrics.key = 'test_neg_mean_absolute_error' THEN metrics.value END) AS mae,\n",
    "                         MAX(CASE WHEN metrics.key = 'test_neg_mean_absolute_percentage_error' THEN metrics.value END) AS mape,\n",
    "                         metrics.run_uuid\n",
    "                       FROM metrics\n",
    "                         LEFT JOIN model_versions ON model_versions.run_id=metrics.run_uuid\n",
    "                       WHERE metrics.key NOT IN ('fit_time', 'score_time')\n",
    "                       GROUP BY COALESCE(model_versions.name, 'baseline'), model_versions.version, metrics.run_uuid\n",
    "                       ORDER BY mse\n",
    "                       \"\"\")\n",
    "        table_data = cur.fetchall()\n",
    "        table_columns = [desc[0] for desc in cur.description]\n",
    "        print('Модели и их метрики:')\n",
    "        display(pd.DataFrame(table_data, columns=table_columns))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fcb18578-1826-4230-8b72-2380c2461b04",
   "metadata": {},
   "source": [
    "### Выводы и наблюдения\n",
    "- лучшие результаты на валидации показала модель `CatBoostRegressor`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95dcf578-bc47-4c25-934e-d6460f21daa7",
   "metadata": {},
   "source": [
    "## Тестирование модели на отложенной выборке"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "8e26bc5c-3006-46c6-a81e-14bb282bee5f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE лучшей модели на отложенной выборке: 17631685.118\n",
      "MAE лучшей модели на отложенной выборке: 3028.507\n",
      "R2 лучшей модели на отложенной выборке: 0.714\n",
      "MAPE лучшей модели на отложенной выборке: 0.338\n"
     ]
    }
   ],
   "source": [
    "data_explorer.model_testing(model=model, test_features=X_test, test_labels=y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "942e6260-ed64-44b2-9b9c-d04c4d30124e",
   "metadata": {},
   "source": [
    "### Выводы и наблюдения\n",
    "- произведено тестирование лучшей модели на отложенной выборке;\n",
    "- по результатам тестирования можно предположить, что переобучения модели не произошло,\n",
    "  <br>дальнейших доработок модели на данном этапе не требуется."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0723e823-e9ea-4e22-8806-e8e13356a6a4",
   "metadata": {},
   "source": [
    "## Проверка важности признаков"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "940ad89e-50c0-4c10-82ed-d49cedc61226",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_explorer.feature_importance(model=model, features=X_test, assets_dir=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9af94a23-cc15-46bf-949b-98933303fe05",
   "metadata": {},
   "source": [
    "### Выводы и наблюдения\n",
    "- наиболее важными признаками для прогнозирования цены выделены:\n",
    "  - год производства автомобиля;\n",
    "  - завод-производитель автомобиля;\n",
    "  - кузов автомобиля"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "031fcbe6-e080-4c18-92e4-f9bfadee08cd",
   "metadata": {},
   "source": [
    "## Отчёт по проведённому исследованию\n",
    "**Загрузка данных и ознакомление с ними**\n",
    "\n",
    "Выполнено\n",
    "\n",
    "- данные загружены и изучены;\n",
    "- составлено общее представление о необходимых действяих для обработки.\n",
    "\n",
    "**Подготовка паттерна для парсинга данных**\n",
    "\n",
    "Выполнено\n",
    "\n",
    "- паттерн подготовлен и протестирован.\n",
    "\n",
    "**Исследовательский анализ данных**\n",
    "\n",
    "Выполнено\n",
    "\n",
    "- исследованы общие параметры датафрейма, наличие дубликатов, типы признаков;\n",
    "- исследована целевая переменная;\n",
    "- изучены зависимости между признаками.\n",
    "\n",
    "**Подготовка данных для обучения моделей**\n",
    "\n",
    "Выполнено\n",
    "\n",
    "- произведено разделение данных на обучающую и тестовую выборки;\n",
    "- признаки в выборках закодированы и масштабированы.\n",
    "\n",
    "**Обучение моделей**\n",
    "\n",
    "Выполнено\n",
    "\n",
    "Обучены и оценены модели:\n",
    "\n",
    "- `Linear Regression` без подбора гиперпараметров - в качестве `baseline`;\n",
    "- `Linear Regression` с подбором гиперпараметров `GridSearch`;\n",
    "- `Random Forest Regressor` с подбором гиперпараметров `GridSearch`;\n",
    "- `CatBoost Regressor` с подбором гиперпараметров `GridSearch`.\n",
    "\n",
    "**Выбор лучшей модели**\n",
    "\n",
    "Выполнено\n",
    "\n",
    "- по метрике `MSE` лучший результат на валидации показала модель `CatBoost Regressor`\n",
    "\n",
    "**Тестирование модели**\n",
    "\n",
    "Выполнено\n",
    "\n",
    "- проверка на тестовой выборке показала адекватность обучения модели\n",
    "\n",
    "**Анализ важности признаков лучшей модели**\n",
    "\n",
    "Выполнено\n",
    "\n",
    "Наиболее важными признаками для прогнозирования цены выделены:\n",
    "- год производства автомобиля;\n",
    "- завод-производитель автомобиля;\n",
    "- кузов автомобиля\n",
    "\n",
    "**Дополнительно**\n",
    "- поднята база PostgreSQL для хранения метрик и параметров моделей;\n",
    "- поднят MLflow server для логирования моделей, параметров и метрик.\n",
    "\n",
    "**Общий итог: для выполнения поставленной задачи наиболее подходящей моделью оказалась модель градиентного бустинга `CatBoost`**"
   ]
  }
 ],
 "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}