ML_project.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "133410ed",
   "metadata": {},
   "source": [
    "# Проект speech emotion recognition on CREMA-D Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c42fea9",
   "metadata": {},
   "source": [
    "Определение эмоции по звуку - обширная задача, решение которой с развитием голосовых помощников и умных колонок вышло на первый план для лидеров индустрии. \n",
    "\n",
    "Более того, решение задачи SER однозначно необходимо в гуманитарной сфере (телефонах поддержки, выявлении фактов домашнего насилия и многое другое)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b9a959c",
   "metadata": {},
   "source": [
    "## Подготовка данных"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b358cbe0",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T16:11:10.891342Z",
     "start_time": "2023-06-27T16:11:10.835521Z"
    }
   },
   "source": [
    "CREMA-D Dataset — это мультимодальный (видео+аудио) датасет из 7442 оригинальных клипов от 91 актера. Эти клипы были сняты 48 мужчинами и 43 женщинами-актёрами в возрасте от 20 до 74 лет, принадлежащими к разным расам и этническим группам (афроамериканцам, азиатам, кавказцам, латиноамериканцам и неуказанным).\n",
    "\n",
    "Актеры произносили 12 предложений с одной из шести эмоций (гнев, отвращение, страх, счастье, нейтральная и грустная) и четырех различных уровней эмоций (низкий, средний, высокий и неопределенный).\n",
    "\n",
    "В данном исследовании используется только аудио часть датасета."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "62d2959b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:35.131325Z",
     "start_time": "2023-06-27T15:50:35.128398Z"
    }
   },
   "outputs": [],
   "source": [
    "data_path = \"/Users/alexandradolidze/Desktop/ML/AudioWAV/\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b56dc60f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:35.137015Z",
     "start_time": "2023-06-27T15:50:35.133837Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# ! pip install matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e21dc32b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.337623Z",
     "start_time": "2023-06-27T15:50:35.141088Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import librosa\n",
    "import librosa.display\n",
    "import matplotlib.pyplot as plt\n",
    "# from matplotlib.pyplot import specgram\n",
    "from tqdm import tqdm\n",
    "import IPython.display as ipd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9b7eb57d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.352014Z",
     "start_time": "2023-06-27T15:50:36.339744Z"
    }
   },
   "outputs": [],
   "source": [
    "data_list = os.listdir(data_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c8f3e37",
   "metadata": {},
   "source": [
    "Из названия звуковой дорожки можно извлечь данные, которые лягут в основу датасета:\n",
    "\n",
    "1004_IEO_SAD_HI.wav\n",
    "\n",
    "* номер говорящего (в данном исследовании он важен для опредлеения пола);\n",
    "* кодировка фразы (так как исследование направлено на опредление эмоции только по звуковым характеристикам, то есть задача speech recognition не ставится, этим параметром можно пренебречь);\n",
    "* эмоция;\n",
    "* интесивность выражения эмоции."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8b7d8d97",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.378644Z",
     "start_time": "2023-06-27T15:50:36.354883Z"
    }
   },
   "outputs": [],
   "source": [
    "# раздербанивание дорожки на датасет \n",
    "gender = []\n",
    "emotion = []\n",
    "intensity = []\n",
    "path = []\n",
    "\n",
    "female = [1002,1003,1004,1006,1007,1008,1009,1010,1012,1013,1018,1020,1021,1024,1025,1028,1029,1030,1037,1043,1046,1047,1049,\n",
    "          1052,1053,1054,1055,1056,1058,1060,1061,1063,1072,1073,1074,1075,1076,1078,1079,1082,1084,1089,1091]\n",
    "\n",
    "\n",
    "for i in data_list:\n",
    "    part = i.split('_')\n",
    "    \n",
    "    if int(part[0]) in female:\n",
    "        temp = 1 # 1 - если женщина\n",
    "    else:\n",
    "        temp = 0 # 0 - если мужчина\n",
    "    gender.append(temp)\n",
    "    \n",
    "    emotion.append(part[2])\n",
    "    \n",
    "    intent, _ = part[3].split('.')\n",
    "    intensity.append(intent)\n",
    "    \n",
    "    path.append(data_path + i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8a9eb78e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.390175Z",
     "start_time": "2023-06-27T15:50:36.381481Z"
    }
   },
   "outputs": [],
   "source": [
    "CREMA_df = pd.DataFrame({\n",
    "    'emotion': emotion,\n",
    "    'gender': gender,\n",
    "    'intensity': intensity,\n",
    "    'path': path\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "29d8b451",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.414912Z",
     "start_time": "2023-06-27T15:50:36.392808Z"
    }
   },
   "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>emotion</th>\n",
       "      <th>gender</th>\n",
       "      <th>intensity</th>\n",
       "      <th>path</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ANG</td>\n",
       "      <td>0</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DIS</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7437</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7438</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7439</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7440</th>\n",
       "      <td>ANG</td>\n",
       "      <td>0</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7441</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7442 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     emotion  gender intensity  \\\n",
       "0        ANG       0        XX   \n",
       "1        ANG       1        XX   \n",
       "2        NEU       1        XX   \n",
       "3        NEU       1        XX   \n",
       "4        DIS       1        XX   \n",
       "...      ...     ...       ...   \n",
       "7437     ANG       1        XX   \n",
       "7438     ANG       1        XX   \n",
       "7439     ANG       1        XX   \n",
       "7440     ANG       0        XX   \n",
       "7441     NEU       1        XX   \n",
       "\n",
       "                                                   path  \n",
       "0     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "1     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "2     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "3     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "4     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "...                                                 ...  \n",
       "7437  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "7438  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "7439  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "7440  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "7441  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "\n",
       "[7442 rows x 4 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CREMA_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0cb458f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T13:59:25.377430Z",
     "start_time": "2023-06-27T13:59:25.351904Z"
    }
   },
   "source": [
    "## Data Understanding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b2262fd6",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.441783Z",
     "start_time": "2023-06-27T15:50:36.417926Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>emotion</th>\n",
       "      <th>gender</th>\n",
       "      <th>intensity</th>\n",
       "      <th>path</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>7442</td>\n",
       "      <td>7442.000000</td>\n",
       "      <td>7442</td>\n",
       "      <td>7442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "      <td>7442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>ANG</td>\n",
       "      <td>NaN</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>1271</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6076</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.471916</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.499244</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       emotion       gender intensity  \\\n",
       "count     7442  7442.000000      7442   \n",
       "unique       6          NaN         5   \n",
       "top        ANG          NaN        XX   \n",
       "freq      1271          NaN      6076   \n",
       "mean       NaN     0.471916       NaN   \n",
       "std        NaN     0.499244       NaN   \n",
       "min        NaN     0.000000       NaN   \n",
       "25%        NaN     0.000000       NaN   \n",
       "50%        NaN     0.000000       NaN   \n",
       "75%        NaN     1.000000       NaN   \n",
       "max        NaN     1.000000       NaN   \n",
       "\n",
       "                                                     path  \n",
       "count                                                7442  \n",
       "unique                                               7442  \n",
       "top     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "freq                                                    1  \n",
       "mean                                                  NaN  \n",
       "std                                                   NaN  \n",
       "min                                                   NaN  \n",
       "25%                                                   NaN  \n",
       "50%                                                   NaN  \n",
       "75%                                                   NaN  \n",
       "max                                                   NaN  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CREMA_df.describe(include='all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "612428b9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.452329Z",
     "start_time": "2023-06-27T15:50:36.448168Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7442, 4)\n"
     ]
    }
   ],
   "source": [
    "print(CREMA_df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "30b05446",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.459529Z",
     "start_time": "2023-06-27T15:50:36.454892Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    3930\n",
      "1    3512\n",
      "Name: gender, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(CREMA_df['gender'].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e8a050b7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.782071Z",
     "start_time": "2023-06-27T15:50:36.462026Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "AttributeError",
     "evalue": "module 'matplotlib' has no attribute 'pyplot'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib_inline/backend_inline.py:99\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(close, block)\u001b[0m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;66;03m# only call close('all') if any to close\u001b[39;00m\n\u001b[1;32m     97\u001b[0m \u001b[38;5;66;03m# close triggers gc.collect, which can be slow\u001b[39;00m\n\u001b[1;32m     98\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m close \u001b[38;5;129;01mand\u001b[39;00m Gcf\u001b[38;5;241m.\u001b[39mget_all_fig_managers():\n\u001b[0;32m---> 99\u001b[0m     \u001b[43mmatplotlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpyplot\u001b[49m\u001b[38;5;241m.\u001b[39mclose(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mall\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/_api/__init__.py:224\u001b[0m, in \u001b[0;36mcaching_module_getattr.<locals>.__getattr__\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m props:\n\u001b[1;32m    223\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m props[name]\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__get__\u001b[39m(instance)\n\u001b[0;32m--> 224\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\n\u001b[1;32m    225\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__module__\u001b[39m\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'matplotlib' has no attribute 'pyplot'"
     ]
    }
   ],
   "source": [
    "CREMA_df['gender'].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "508d8321",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.787860Z",
     "start_time": "2023-06-27T15:50:36.784534Z"
    }
   },
   "outputs": [],
   "source": [
    "plt.close('all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "8032d876",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.796804Z",
     "start_time": "2023-06-27T15:50:36.790995Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ANG    1271\n",
      "DIS    1271\n",
      "SAD    1271\n",
      "FEA    1271\n",
      "HAP    1271\n",
      "NEU    1087\n",
      "Name: emotion, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(CREMA_df['emotion'].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "54f66ad8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.981183Z",
     "start_time": "2023-06-27T15:50:36.800057Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "AttributeError",
     "evalue": "module 'matplotlib' has no attribute 'pyplot'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib_inline/backend_inline.py:99\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(close, block)\u001b[0m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;66;03m# only call close('all') if any to close\u001b[39;00m\n\u001b[1;32m     97\u001b[0m \u001b[38;5;66;03m# close triggers gc.collect, which can be slow\u001b[39;00m\n\u001b[1;32m     98\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m close \u001b[38;5;129;01mand\u001b[39;00m Gcf\u001b[38;5;241m.\u001b[39mget_all_fig_managers():\n\u001b[0;32m---> 99\u001b[0m     \u001b[43mmatplotlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpyplot\u001b[49m\u001b[38;5;241m.\u001b[39mclose(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mall\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/_api/__init__.py:224\u001b[0m, in \u001b[0;36mcaching_module_getattr.<locals>.__getattr__\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m props:\n\u001b[1;32m    223\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m props[name]\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__get__\u001b[39m(instance)\n\u001b[0;32m--> 224\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\n\u001b[1;32m    225\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__module__\u001b[39m\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'matplotlib' has no attribute 'pyplot'"
     ]
    }
   ],
   "source": [
    "CREMA_df['emotion'].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "cf38ebb3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.986709Z",
     "start_time": "2023-06-27T15:50:36.983778Z"
    }
   },
   "outputs": [],
   "source": [
    "plt.close('all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "dcad8f07",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:36.994232Z",
     "start_time": "2023-06-27T15:50:36.989319Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XX    6076\n",
      "MD     455\n",
      "HI     455\n",
      "LO     455\n",
      "X        1\n",
      "Name: intensity, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(CREMA_df['intensity'].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e6be2fd9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:37.166190Z",
     "start_time": "2023-06-27T15:50:36.996996Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "AttributeError",
     "evalue": "module 'matplotlib' has no attribute 'pyplot'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib_inline/backend_inline.py:99\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(close, block)\u001b[0m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;66;03m# only call close('all') if any to close\u001b[39;00m\n\u001b[1;32m     97\u001b[0m \u001b[38;5;66;03m# close triggers gc.collect, which can be slow\u001b[39;00m\n\u001b[1;32m     98\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m close \u001b[38;5;129;01mand\u001b[39;00m Gcf\u001b[38;5;241m.\u001b[39mget_all_fig_managers():\n\u001b[0;32m---> 99\u001b[0m     \u001b[43mmatplotlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpyplot\u001b[49m\u001b[38;5;241m.\u001b[39mclose(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mall\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/_api/__init__.py:224\u001b[0m, in \u001b[0;36mcaching_module_getattr.<locals>.__getattr__\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m props:\n\u001b[1;32m    223\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m props[name]\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__get__\u001b[39m(instance)\n\u001b[0;32m--> 224\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\n\u001b[1;32m    225\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__module__\u001b[39m\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'matplotlib' has no attribute 'pyplot'"
     ]
    }
   ],
   "source": [
    "CREMA_df['intensity'].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "88869617",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:37.171974Z",
     "start_time": "2023-06-27T15:50:37.168976Z"
    }
   },
   "outputs": [],
   "source": [
    "plt.close('all')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82f80dc9",
   "metadata": {},
   "source": [
    "## Извлечение фичей из аудио"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "5dba3420",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:50:37.900775Z",
     "start_time": "2023-06-27T15:50:37.174518Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.decomposition import PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "6dd6a07a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:54.167264Z",
     "start_time": "2023-06-27T15:50:37.903050Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "4653it [01:58, 43.93it/s]/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/sklearn/decomposition/_pca.py:527: RuntimeWarning: invalid value encountered in divide\n",
      "  explained_variance_ratio_ = explained_variance_ / total_var\n",
      "7442it [03:16, 37.94it/s]\n"
     ]
    }
   ],
   "source": [
    "mfccs = np.zeros((len(CREMA_df.path), 2))\n",
    "melspecs = np.zeros((len(CREMA_df.path), 2))\n",
    "# zero_cross_rates = np.zeros((len(CREMA_df.path), 2))\n",
    "# rolloffs = np.zeros((len(CREMA_df.path), 2))\n",
    "\n",
    "for i, path in tqdm(enumerate(CREMA_df.path)):\n",
    "    X, sample_rate = librosa.load(path)  \n",
    "    \n",
    "    # MFCC\n",
    "    mfcc = librosa.feature.mfcc(y=X, sr=sample_rate, n_mels = 128)\n",
    "    \n",
    "    pca_mfcc = PCA(n_components=2, random_state=322)\n",
    "    principalComponents_mfcc = pca_mfcc.fit_transform(mfcc)\n",
    "    mfccs[i] = np.mean(principalComponents_mfcc, axis=0)\n",
    "      \n",
    "    # Log Mel-spectogram\n",
    "    melspec = librosa.feature.melspectrogram(y=X, sr=sample_rate, n_mels = 128)\n",
    "    # Конвертируем в логарифмическую шкалу\n",
    "    melspec_log_S = librosa.amplitude_to_db(melspec)\n",
    "    \n",
    "    pca_melspec = PCA(n_components=2, random_state=322)\n",
    "    principalComponents_melspec = pca_melspec.fit_transform(melspec_log_S)\n",
    "    melspecs[i] = np.mean(principalComponents_melspec, axis=0)\n",
    "    \n",
    "#     # Zero-crossing rate\n",
    "#     zero_cross_rate = np.mean(librosa.feature.zero_crossing_rate(X), axis=0)\n",
    "#     print(zero_cross_rate)\n",
    "#     print(np.squeeze(zero_cross_rate).shape)\n",
    "    \n",
    "    \n",
    "#     np.vstack((tp, fp)).T\n",
    "#     if len(zero_cross_rate) < 216:\n",
    "#         zero_cross_rate = np.pad(zero_cross_rate, (0, 216-len(zero_cross_rate)), 'constant', constant_values=(0))\n",
    "#     zero_cross_rates[i] = zero_cross_rate  \n",
    "\n",
    "#     # Sectral roll-off\n",
    "#     zero_cross_rate = np.mean(librosa.feature.zero_crossing_rate(X), axis = 0)\n",
    "#     if len(zero_cross_rate) < 216:\n",
    "#         zero_cross_rate = np.pad(zero_cross_rate, (0, 216-len(zero_cross_rate)), 'constant', constant_values=(0))\n",
    "#     zero_cross_rates[i] = zero_cross_rate  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "80d35adc",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:54.266891Z",
     "start_time": "2023-06-27T15:53:54.216170Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "                <audio  controls=\"controls\" >\n",
       "                    <source src=\"data:audio/x-wav;base64,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\" type=\"audio/x-wav\" />\n",
       "                    Your browser does not support the audio element.\n",
       "                </audio>\n",
       "              "
      ],
      "text/plain": [
       "<IPython.lib.display.Audio object>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ipd.Audio(CREMA_df.path[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "1df84389",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:54.788883Z",
     "start_time": "2023-06-27T15:53:54.270722Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Audio sampled at 22050 hrz')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "AttributeError",
     "evalue": "module 'matplotlib' has no attribute 'pyplot'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib_inline/backend_inline.py:99\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(close, block)\u001b[0m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;66;03m# only call close('all') if any to close\u001b[39;00m\n\u001b[1;32m     97\u001b[0m \u001b[38;5;66;03m# close triggers gc.collect, which can be slow\u001b[39;00m\n\u001b[1;32m     98\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m close \u001b[38;5;129;01mand\u001b[39;00m Gcf\u001b[38;5;241m.\u001b[39mget_all_fig_managers():\n\u001b[0;32m---> 99\u001b[0m     \u001b[43mmatplotlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpyplot\u001b[49m\u001b[38;5;241m.\u001b[39mclose(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mall\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/_api/__init__.py:224\u001b[0m, in \u001b[0;36mcaching_module_getattr.<locals>.__getattr__\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m props:\n\u001b[1;32m    223\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m props[name]\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__get__\u001b[39m(instance)\n\u001b[0;32m--> 224\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\n\u001b[1;32m    225\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__module__\u001b[39m\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'matplotlib' has no attribute 'pyplot'"
     ]
    }
   ],
   "source": [
    "# визуализация фичей\n",
    "\n",
    "X, sample_rate = librosa.load(CREMA_df.path[0]) \n",
    "\n",
    "# \n",
    "\n",
    "plt.figure(figsize=(12, 3))\n",
    "librosa.display.waveshow(X, sr=sample_rate)\n",
    "plt.title(f'Audio sampled at {sample_rate} hrz')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "b14eb671",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:54.795274Z",
     "start_time": "2023-06-27T15:53:54.791865Z"
    }
   },
   "outputs": [],
   "source": [
    "plt.close('all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "884572a3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.025721Z",
     "start_time": "2023-06-27T15:53:54.797728Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7f7fac5279d0>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "AttributeError",
     "evalue": "module 'matplotlib' has no attribute 'pyplot'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib_inline/backend_inline.py:99\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(close, block)\u001b[0m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;66;03m# only call close('all') if any to close\u001b[39;00m\n\u001b[1;32m     97\u001b[0m \u001b[38;5;66;03m# close triggers gc.collect, which can be slow\u001b[39;00m\n\u001b[1;32m     98\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m close \u001b[38;5;129;01mand\u001b[39;00m Gcf\u001b[38;5;241m.\u001b[39mget_all_fig_managers():\n\u001b[0;32m---> 99\u001b[0m     \u001b[43mmatplotlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpyplot\u001b[49m\u001b[38;5;241m.\u001b[39mclose(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mall\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/_api/__init__.py:224\u001b[0m, in \u001b[0;36mcaching_module_getattr.<locals>.__getattr__\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m props:\n\u001b[1;32m    223\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m props[name]\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__get__\u001b[39m(instance)\n\u001b[0;32m--> 224\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\n\u001b[1;32m    225\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__module__\u001b[39m\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'matplotlib' has no attribute 'pyplot'"
     ]
    }
   ],
   "source": [
    "mfcc = librosa.feature.mfcc(y=X, sr=sample_rate, n_mfcc = 5)\n",
    "\n",
    "plt.figure(figsize=(12, 6))\n",
    "librosa.display.specshow(mfcc)\n",
    "plt.ylabel('MFCC')\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "b7d518fe",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.031665Z",
     "start_time": "2023-06-27T15:53:55.028789Z"
    }
   },
   "outputs": [],
   "source": [
    "plt.close('all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "e14cf653",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.368274Z",
     "start_time": "2023-06-27T15:53:55.033982Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7f7faac72c10>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "AttributeError",
     "evalue": "module 'matplotlib' has no attribute 'pyplot'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib_inline/backend_inline.py:99\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(close, block)\u001b[0m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;66;03m# only call close('all') if any to close\u001b[39;00m\n\u001b[1;32m     97\u001b[0m \u001b[38;5;66;03m# close triggers gc.collect, which can be slow\u001b[39;00m\n\u001b[1;32m     98\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m close \u001b[38;5;129;01mand\u001b[39;00m Gcf\u001b[38;5;241m.\u001b[39mget_all_fig_managers():\n\u001b[0;32m---> 99\u001b[0m     \u001b[43mmatplotlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpyplot\u001b[49m\u001b[38;5;241m.\u001b[39mclose(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mall\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/_api/__init__.py:224\u001b[0m, in \u001b[0;36mcaching_module_getattr.<locals>.__getattr__\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m props:\n\u001b[1;32m    223\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m props[name]\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__get__\u001b[39m(instance)\n\u001b[0;32m--> 224\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\n\u001b[1;32m    225\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__module__\u001b[39m\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'matplotlib' has no attribute 'pyplot'"
     ]
    }
   ],
   "source": [
    "melspec = librosa.feature.melspectrogram(y=X, sr=sample_rate, n_mels=128)\n",
    "\n",
    "log_S = librosa.amplitude_to_db(melspec)\n",
    "\n",
    "plt.figure(figsize=(12,4))\n",
    "librosa.display.specshow(log_S, sr=sample_rate, x_axis='time', y_axis='mel')\n",
    "plt.title('Log mel spectrogram')\n",
    "plt.colorbar(format='%+02.0f dB')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "1c9ff9fb",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.386525Z",
     "start_time": "2023-06-27T15:53:55.383973Z"
    }
   },
   "outputs": [],
   "source": [
    "plt.close('all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "c7d85770",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.395206Z",
     "start_time": "2023-06-27T15:53:55.389165Z"
    }
   },
   "outputs": [],
   "source": [
    "principalComponents_mfcc_df = pd.DataFrame(data = mfccs) \n",
    "principalComponents_melspec_df = pd.DataFrame(data = melspecs) \n",
    "# principalComponents_zero_crossing_df = pd.DataFrame(data = zero_cross_rates)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "7d559c8f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.413658Z",
     "start_time": "2023-06-27T15:53:55.397491Z"
    }
   },
   "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>emotion</th>\n",
       "      <th>gender</th>\n",
       "      <th>intensity</th>\n",
       "      <th>path</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ANG</td>\n",
       "      <td>0</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DIS</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7437</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7438</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7439</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7440</th>\n",
       "      <td>ANG</td>\n",
       "      <td>0</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7441</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7442 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     emotion  gender intensity  \\\n",
       "0        ANG       0        XX   \n",
       "1        ANG       1        XX   \n",
       "2        NEU       1        XX   \n",
       "3        NEU       1        XX   \n",
       "4        DIS       1        XX   \n",
       "...      ...     ...       ...   \n",
       "7437     ANG       1        XX   \n",
       "7438     ANG       1        XX   \n",
       "7439     ANG       1        XX   \n",
       "7440     ANG       0        XX   \n",
       "7441     NEU       1        XX   \n",
       "\n",
       "                                                   path  \n",
       "0     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "1     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "2     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "3     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "4     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "...                                                 ...  \n",
       "7437  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "7438  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "7439  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "7440  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "7441  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  \n",
       "\n",
       "[7442 rows x 4 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CREMA_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "da5c4ef6",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.441574Z",
     "start_time": "2023-06-27T15:53:55.416294Z"
    }
   },
   "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>emotion</th>\n",
       "      <th>gender</th>\n",
       "      <th>intensity</th>\n",
       "      <th>path</th>\n",
       "      <th>0_MFCC</th>\n",
       "      <th>1_MFCC</th>\n",
       "      <th>0_Melspec</th>\n",
       "      <th>1_Melspec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ANG</td>\n",
       "      <td>0</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000007</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>-0.000011</td>\n",
       "      <td>0.000004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000211</td>\n",
       "      <td>0.000082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000008</td>\n",
       "      <td>-0.000001</td>\n",
       "      <td>-0.000307</td>\n",
       "      <td>0.000053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000008</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000433</td>\n",
       "      <td>0.000165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DIS</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000179</td>\n",
       "      <td>-0.000169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7437</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000006</td>\n",
       "      <td>0.000050</td>\n",
       "      <td>0.000019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7438</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000024</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.000060</td>\n",
       "      <td>0.000081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7439</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000016</td>\n",
       "      <td>-0.000005</td>\n",
       "      <td>-0.000044</td>\n",
       "      <td>-0.000032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7440</th>\n",
       "      <td>ANG</td>\n",
       "      <td>0</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000063</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>-0.000088</td>\n",
       "      <td>-0.000178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7441</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>XX</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000015</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000345</td>\n",
       "      <td>0.000324</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7442 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     emotion  gender intensity  \\\n",
       "0        ANG       0        XX   \n",
       "1        ANG       1        XX   \n",
       "2        NEU       1        XX   \n",
       "3        NEU       1        XX   \n",
       "4        DIS       1        XX   \n",
       "...      ...     ...       ...   \n",
       "7437     ANG       1        XX   \n",
       "7438     ANG       1        XX   \n",
       "7439     ANG       1        XX   \n",
       "7440     ANG       0        XX   \n",
       "7441     NEU       1        XX   \n",
       "\n",
       "                                                   path    0_MFCC    1_MFCC  \\\n",
       "0     /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000007  0.000017   \n",
       "1     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  0.000045  0.000005   \n",
       "2     /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000008 -0.000001   \n",
       "3     /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000008  0.000003   \n",
       "4     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  0.000015  0.000002   \n",
       "...                                                 ...       ...       ...   \n",
       "7437  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  0.000003 -0.000006   \n",
       "7438  /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000024  0.000023   \n",
       "7439  /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000016 -0.000005   \n",
       "7440  /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000063  0.000005   \n",
       "7441  /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000015  0.000006   \n",
       "\n",
       "      0_Melspec  1_Melspec  \n",
       "0     -0.000011   0.000004  \n",
       "1      0.000211   0.000082  \n",
       "2     -0.000307   0.000053  \n",
       "3      0.000433   0.000165  \n",
       "4      0.000179  -0.000169  \n",
       "...         ...        ...  \n",
       "7437   0.000050   0.000019  \n",
       "7438   0.000060   0.000081  \n",
       "7439  -0.000044  -0.000032  \n",
       "7440  -0.000088  -0.000178  \n",
       "7441   0.000345   0.000324  \n",
       "\n",
       "[7442 rows x 8 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "principalComponents_mfcc_df = principalComponents_mfcc_df.add_suffix('_MFCC')\n",
    "principalComponents_melspec_df = principalComponents_melspec_df.add_suffix('_Melspec')\n",
    "# principalComponents_zero_crossing_df = principalComponents_zero_crossing_df.add_suffix('_Zero-Crossing')\n",
    "\n",
    "CREMA_df = pd.concat([CREMA_df, principalComponents_mfcc_df, principalComponents_melspec_df], axis=1)\n",
    "\n",
    "CREMA_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "c11870dd",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.449485Z",
     "start_time": "2023-06-27T15:53:55.444899Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7442, 8)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CREMA_df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55d4f5c4",
   "metadata": {},
   "source": [
    "## Классификация"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "39df3687",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.463770Z",
     "start_time": "2023-06-27T15:53:55.451796Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['HI' 'LO' 'MD' 'X' 'XX']\n"
     ]
    }
   ],
   "source": [
    "# Перекодируем intensity\n",
    "\n",
    "from sklearn import preprocessing\n",
    "le = preprocessing.LabelEncoder()\n",
    "\n",
    "intesity_value = CREMA_df['intensity'].values\n",
    "\n",
    "le.fit(intesity_value)\n",
    "print(le.classes_)\n",
    "CREMA_df['intensity'] = list(le.transform(intesity_value))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "bf75902e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.479679Z",
     "start_time": "2023-06-27T15:53:55.466021Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>emotion</th>\n",
       "      <th>gender</th>\n",
       "      <th>intensity</th>\n",
       "      <th>path</th>\n",
       "      <th>0_MFCC</th>\n",
       "      <th>1_MFCC</th>\n",
       "      <th>0_Melspec</th>\n",
       "      <th>1_Melspec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ANG</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000007</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>-0.000011</td>\n",
       "      <td>0.000004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000211</td>\n",
       "      <td>0.000082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000008</td>\n",
       "      <td>-0.000001</td>\n",
       "      <td>-0.000307</td>\n",
       "      <td>0.000053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000008</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000433</td>\n",
       "      <td>0.000165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DIS</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000179</td>\n",
       "      <td>-0.000169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7437</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-0.000006</td>\n",
       "      <td>0.000050</td>\n",
       "      <td>0.000019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7438</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000024</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.000060</td>\n",
       "      <td>0.000081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7439</th>\n",
       "      <td>ANG</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000016</td>\n",
       "      <td>-0.000005</td>\n",
       "      <td>-0.000044</td>\n",
       "      <td>-0.000032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7440</th>\n",
       "      <td>ANG</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000063</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>-0.000088</td>\n",
       "      <td>-0.000178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7441</th>\n",
       "      <td>NEU</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>/Users/alexandradolidze/Desktop/ML/AudioWAV/10...</td>\n",
       "      <td>-0.000015</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000345</td>\n",
       "      <td>0.000324</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7442 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     emotion  gender  intensity  \\\n",
       "0        ANG       0          4   \n",
       "1        ANG       1          4   \n",
       "2        NEU       1          4   \n",
       "3        NEU       1          4   \n",
       "4        DIS       1          4   \n",
       "...      ...     ...        ...   \n",
       "7437     ANG       1          4   \n",
       "7438     ANG       1          4   \n",
       "7439     ANG       1          4   \n",
       "7440     ANG       0          4   \n",
       "7441     NEU       1          4   \n",
       "\n",
       "                                                   path    0_MFCC    1_MFCC  \\\n",
       "0     /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000007  0.000017   \n",
       "1     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  0.000045  0.000005   \n",
       "2     /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000008 -0.000001   \n",
       "3     /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000008  0.000003   \n",
       "4     /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  0.000015  0.000002   \n",
       "...                                                 ...       ...       ...   \n",
       "7437  /Users/alexandradolidze/Desktop/ML/AudioWAV/10...  0.000003 -0.000006   \n",
       "7438  /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000024  0.000023   \n",
       "7439  /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000016 -0.000005   \n",
       "7440  /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000063  0.000005   \n",
       "7441  /Users/alexandradolidze/Desktop/ML/AudioWAV/10... -0.000015  0.000006   \n",
       "\n",
       "      0_Melspec  1_Melspec  \n",
       "0     -0.000011   0.000004  \n",
       "1      0.000211   0.000082  \n",
       "2     -0.000307   0.000053  \n",
       "3      0.000433   0.000165  \n",
       "4      0.000179  -0.000169  \n",
       "...         ...        ...  \n",
       "7437   0.000050   0.000019  \n",
       "7438   0.000060   0.000081  \n",
       "7439  -0.000044  -0.000032  \n",
       "7440  -0.000088  -0.000178  \n",
       "7441   0.000345   0.000324  \n",
       "\n",
       "[7442 rows x 8 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CREMA_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "5eb15e15",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.494924Z",
     "start_time": "2023-06-27T15:53:55.482202Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score, classification_report, silhouette_score"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84423517",
   "metadata": {},
   "source": [
    "### Определение пола говорящего (бинарная классификация)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "9ab7f663",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.504985Z",
     "start_time": "2023-06-27T15:53:55.496911Z"
    }
   },
   "outputs": [],
   "source": [
    "binary_X = CREMA_df.drop(columns=['emotion', 'path', 'gender'], axis=1).values\n",
    "binary_y = CREMA_df['gender'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "7ed4cb51",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.516060Z",
     "start_time": "2023-06-27T15:53:55.508033Z"
    }
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(binary_X, binary_y, test_size=0.33, random_state=322)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b0cc12b",
   "metadata": {},
   "source": [
    "#### Рандомный бейзлайн"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "164d565d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.520938Z",
     "start_time": "2023-06-27T15:53:55.518178Z"
    }
   },
   "outputs": [],
   "source": [
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "78bf7cbe",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.528633Z",
     "start_time": "2023-06-27T15:53:55.523538Z"
    }
   },
   "outputs": [],
   "source": [
    "num_samples = CREMA_df.shape[0]  \n",
    "prior_class_counts = Counter(binary_y)  \n",
    "prior_class_probs = dict()\n",
    "for class_label in prior_class_counts:\n",
    "    prior_class_probs[class_label] = prior_class_counts[class_label] / num_samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "40e1014f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.548414Z",
     "start_time": "2023-06-27T15:53:55.530982Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.4853\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "        male       0.51      0.52      0.52      1298\n",
      "      female       0.45      0.45      0.45      1158\n",
      "\n",
      "    accuracy                           0.49      2456\n",
      "   macro avg       0.48      0.48      0.48      2456\n",
      "weighted avg       0.48      0.49      0.49      2456\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import random\n",
    "random.seed(322)\n",
    "\n",
    "random_pred = random.choices([0, 1], \n",
    "                             weights=(list(prior_class_probs.values())),\n",
    "                             k=len(X_test))\n",
    "\n",
    "print(\"Accuracy: {:.4}\".format(accuracy_score(y_test, random_pred)))\n",
    "\n",
    "print(classification_report(y_test, random_pred, target_names = ['male', 'female']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19048166",
   "metadata": {},
   "source": [
    "#### Логистическая регрессия"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "d2632cdd",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.554474Z",
     "start_time": "2023-06-27T15:53:55.550734Z"
    }
   },
   "outputs": [],
   "source": [
    "model_log_reg = Pipeline([\n",
    "    ('scaler', StandardScaler()),\n",
    "    ('log_reg', LogisticRegression(class_weight = 'balanced'))\n",
    "], verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "8a5fddd1",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.594733Z",
     "start_time": "2023-06-27T15:53:55.557282Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Pipeline] ............ (step 1 of 2) Processing scaler, total=   0.0s\n",
      "[Pipeline] ........... (step 2 of 2) Processing log_reg, total=   0.0s\n"
     ]
    }
   ],
   "source": [
    "model_log_reg.fit(X_train, y_train)\n",
    "y_pred_log_reg = model_log_reg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "79d52bce",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.611507Z",
     "start_time": "2023-06-27T15:53:55.597431Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.5106\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "        male       0.54      0.49      0.51      1298\n",
      "      female       0.48      0.54      0.51      1158\n",
      "\n",
      "    accuracy                           0.51      2456\n",
      "   macro avg       0.51      0.51      0.51      2456\n",
      "weighted avg       0.51      0.51      0.51      2456\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy: {:.4}\".format(accuracy_score(y_test, y_pred_log_reg)))\n",
    "\n",
    "print(classification_report(y_test, y_pred_log_reg, target_names = ['male', 'female']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e25652df",
   "metadata": {},
   "source": [
    "#### Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "c93f896e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.617576Z",
     "start_time": "2023-06-27T15:53:55.614314Z"
    }
   },
   "outputs": [],
   "source": [
    "model_naive = Pipeline([\n",
    "    ('scaler', StandardScaler()),\n",
    "    ('naive_bayes', GaussianNB())\n",
    "], verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "1a7e16e9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.634002Z",
     "start_time": "2023-06-27T15:53:55.620375Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Pipeline] ............ (step 1 of 2) Processing scaler, total=   0.0s\n",
      "[Pipeline] ....... (step 2 of 2) Processing naive_bayes, total=   0.0s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;scaler&#x27;, StandardScaler()), (&#x27;naive_bayes&#x27;, GaussianNB())],\n",
       "         verbose=True)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;scaler&#x27;, StandardScaler()), (&#x27;naive_bayes&#x27;, GaussianNB())],\n",
       "         verbose=True)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GaussianNB</label><div class=\"sk-toggleable__content\"><pre>GaussianNB()</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('scaler', StandardScaler()), ('naive_bayes', GaussianNB())],\n",
       "         verbose=True)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_naive.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "4d096a78",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.642304Z",
     "start_time": "2023-06-27T15:53:55.636666Z"
    }
   },
   "outputs": [],
   "source": [
    "y_pred_naive = model_naive.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "7597f7c7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.659302Z",
     "start_time": "2023-06-27T15:53:55.645617Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.5651\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "        male       0.56      0.81      0.66      1298\n",
      "      female       0.58      0.29      0.39      1158\n",
      "\n",
      "    accuracy                           0.57      2456\n",
      "   macro avg       0.57      0.55      0.53      2456\n",
      "weighted avg       0.57      0.57      0.53      2456\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(f'Accuracy: {accuracy_score(y_test, y_pred_naive):.4}')\n",
    "\n",
    "print(classification_report(y_test, y_pred_naive, target_names = ['male', 'female']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75ce0308",
   "metadata": {},
   "source": [
    "### Определение эмоции говорящего (мультиклассовая классификация)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "cbe490a1",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.669153Z",
     "start_time": "2023-06-27T15:53:55.662506Z"
    }
   },
   "outputs": [],
   "source": [
    "multiclass_X = CREMA_df.drop(columns=['path', 'emotion', 'gender'], axis=1).values\n",
    "multiclass_y = CREMA_df['emotion'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "8889ae47",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.734947Z",
     "start_time": "2023-06-27T15:53:55.672046Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.tree import DecisionTreeClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "a0d13cdb",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.743410Z",
     "start_time": "2023-06-27T15:53:55.737555Z"
    }
   },
   "outputs": [],
   "source": [
    "X_train_multiclass, X_test_multiclass, y_train_multiclass, y_test_multiclass = train_test_split(multiclass_X, multiclass_y, test_size=0.33, random_state=322)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9ad2471",
   "metadata": {},
   "source": [
    "#### MultinomialNB (вместо бейзлайна)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "d5c3e0ff",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.749778Z",
     "start_time": "2023-06-27T15:53:55.746407Z"
    }
   },
   "outputs": [],
   "source": [
    "model_multibayes = Pipeline([\n",
    "    ('minmax', MinMaxScaler()),\n",
    "    ('naive_bayes', MultinomialNB(alpha=1.0, class_prior=None, fit_prior=True))\n",
    "], verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "c333a794",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.779076Z",
     "start_time": "2023-06-27T15:53:55.752264Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Pipeline] ............ (step 1 of 2) Processing minmax, total=   0.0s\n",
      "[Pipeline] ....... (step 2 of 2) Processing naive_bayes, total=   0.0s\n"
     ]
    }
   ],
   "source": [
    "model_multibayes.fit(X_train_multiclass, y_train_multiclass)\n",
    "y_pred_multibayes = model_multibayes.predict(X_test_multiclass)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "864171bc",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:53:55.838869Z",
     "start_time": "2023-06-27T15:53:55.781048Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.1641\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         ANG       0.16      0.94      0.28       398\n",
      "         DIS       0.00      0.00      0.00       428\n",
      "         FEA       0.00      0.00      0.00       445\n",
      "         HAP       0.00      0.00      0.00       426\n",
      "         NEU       0.00      0.00      0.00       355\n",
      "         SAD       0.21      0.07      0.11       404\n",
      "\n",
      "    accuracy                           0.16      2456\n",
      "   macro avg       0.06      0.17      0.06      2456\n",
      "weighted avg       0.06      0.16      0.06      2456\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "print(f'Accuracy: {accuracy_score(y_test_multiclass, y_pred_multibayes):.4}')\n",
    "print(classification_report(y_test_multiclass, y_pred_multibayes))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5909a6a2",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T11:51:26.803135Z",
     "start_time": "2023-06-27T11:51:26.800007Z"
    }
   },
   "source": [
    "#### Дерево решений"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "f9839036",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:44.064011Z",
     "start_time": "2023-06-27T15:55:44.015015Z"
    }
   },
   "outputs": [],
   "source": [
    "decition_tree = DecisionTreeClassifier()\n",
    "decition_tree.fit(X_train_multiclass, y_train_multiclass)\n",
    "y_pred_decision_tree = decition_tree.predict(X_test_multiclass)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "1866d7bf",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:44.835403Z",
     "start_time": "2023-06-27T15:55:44.751151Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.2333\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         ANG       0.32      0.35      0.33       398\n",
      "         DIS       0.17      0.17      0.17       428\n",
      "         FEA       0.17      0.15      0.16       445\n",
      "         HAP       0.23      0.23      0.23       426\n",
      "         NEU       0.20      0.20      0.20       355\n",
      "         SAD       0.29      0.29      0.29       404\n",
      "\n",
      "    accuracy                           0.23      2456\n",
      "   macro avg       0.23      0.23      0.23      2456\n",
      "weighted avg       0.23      0.23      0.23      2456\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(f'Accuracy: {accuracy_score(y_test_multiclass, y_pred_decition_tree):.4}')\n",
    "\n",
    "print(classification_report(y_test_multiclass, y_pred_decision_tree))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "7819d552",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:46.343865Z",
     "start_time": "2023-06-27T15:55:46.340330Z"
    }
   },
   "outputs": [],
   "source": [
    "# улучшие decition tree"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05036305",
   "metadata": {},
   "source": [
    "#### K-Neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "409fd189",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:47.569199Z",
     "start_time": "2023-06-27T15:55:47.563824Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "59737862",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:48.259146Z",
     "start_time": "2023-06-27T15:55:48.144126Z"
    }
   },
   "outputs": [],
   "source": [
    "neigh = KNeighborsClassifier()\n",
    "neigh.fit(X_train_multiclass, y_train_multiclass)\n",
    "\n",
    "y_pred_neigh = neigh.predict(X_test_multiclass)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "1e5e0ace",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:48.776323Z",
     "start_time": "2023-06-27T15:55:48.679067Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.2801\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         ANG       0.35      0.55      0.43       398\n",
      "         DIS       0.21      0.22      0.22       428\n",
      "         FEA       0.20      0.15      0.17       445\n",
      "         HAP       0.26      0.21      0.23       426\n",
      "         NEU       0.27      0.27      0.27       355\n",
      "         SAD       0.34      0.30      0.32       404\n",
      "\n",
      "    accuracy                           0.28      2456\n",
      "   macro avg       0.27      0.28      0.27      2456\n",
      "weighted avg       0.27      0.28      0.27      2456\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(f'Accuracy: {accuracy_score(y_test_multiclass, y_pred_neigh):.4}')\n",
    "\n",
    "print(classification_report(y_test_multiclass, y_pred_neigh))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "8bc9c6de",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:50.410560Z",
     "start_time": "2023-06-27T15:55:49.037110Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.2801, step/iter : 5\n",
      "Accuracy: 0.2944, step/iter : 10\n",
      "Accuracy: 0.305, step/iter : 15\n",
      "Accuracy: 0.3119, step/iter : 20\n",
      "Accuracy: 0.3184, step/iter : 25\n",
      "Accuracy: 0.3241, step/iter : 30\n",
      "Accuracy: 0.3233, step/iter : 35\n",
      "Accuracy: 0.3294, step/iter : 40\n",
      "Accuracy: 0.327, step/iter : 45\n",
      "Accuracy: 0.3286, step/iter : 50\n",
      "Accuracy: 0.3249, step/iter : 55\n",
      "Accuracy: 0.3229, step/iter : 60\n"
     ]
    }
   ],
   "source": [
    "# поподбираем n_neighbors\n",
    "for i in range(5, 65, 5):\n",
    "    neigh = KNeighborsClassifier(n_neighbors = i)\n",
    "    neigh_model = neigh.fit(X_train_multiclass, y_train_multiclass)\n",
    "    y_pred_neigh = neigh_model.predict(X_test_multiclass)\n",
    "    accuracy = accuracy_score(y_test_multiclass, y_pred_neigh)\n",
    "    print(f'Accuracy: {accuracy:.4}, step/iter : {i}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "c985bfd9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:53.403282Z",
     "start_time": "2023-06-27T15:55:52.122398Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.327, step/iter : 45\n",
      "Accuracy: 0.3249, step/iter : 46\n",
      "Accuracy: 0.3257, step/iter : 47\n",
      "Accuracy: 0.3274, step/iter : 48\n",
      "Accuracy: 0.3265, step/iter : 49\n",
      "Accuracy: 0.3286, step/iter : 50\n",
      "Accuracy: 0.3278, step/iter : 51\n",
      "Accuracy: 0.3265, step/iter : 52\n",
      "Accuracy: 0.3282, step/iter : 53\n",
      "Accuracy: 0.3253, step/iter : 54\n"
     ]
    }
   ],
   "source": [
    "for i in range(45, 55):\n",
    "    neigh = KNeighborsClassifier(n_neighbors = i)\n",
    "    neigh_model = neigh.fit(X_train_multiclass, y_train_multiclass)\n",
    "    y_pred_neigh = neigh_model.predict(X_test_multiclass)\n",
    "    accuracy = accuracy_score(y_test_multiclass, y_pred_neigh)\n",
    "    print(f'Accuracy: {accuracy:.4}, step/iter : {i}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "06b0661f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:53.547974Z",
     "start_time": "2023-06-27T15:55:53.406369Z"
    }
   },
   "outputs": [],
   "source": [
    "neigh = KNeighborsClassifier(n_neighbors = 50)\n",
    "neigh.fit(X_train_multiclass, y_train_multiclass)\n",
    "\n",
    "y_pred_neigh = neigh.predict(X_test_multiclass)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "e723f83b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:55:54.001601Z",
     "start_time": "2023-06-27T15:55:53.923732Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.3286\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         ANG       0.36      0.71      0.48       398\n",
      "         DIS       0.20      0.07      0.11       428\n",
      "         FEA       0.27      0.03      0.06       445\n",
      "         HAP       0.31      0.29      0.30       426\n",
      "         NEU       0.27      0.41      0.33       355\n",
      "         SAD       0.39      0.52      0.44       404\n",
      "\n",
      "    accuracy                           0.33      2456\n",
      "   macro avg       0.30      0.34      0.29      2456\n",
      "weighted avg       0.30      0.33      0.28      2456\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(f'Accuracy: {accuracy_score(y_test_multiclass, y_pred_neigh):.4}')\n",
    "\n",
    "print(classification_report(y_test_multiclass, y_pred_neigh))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19f06980",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T12:59:41.924578Z",
     "start_time": "2023-06-27T12:59:41.921290Z"
    }
   },
   "source": [
    "## К-means (кластеризация)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "f3456dce",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:56:03.676726Z",
     "start_time": "2023-06-27T15:56:03.158636Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "8a11ff8c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:56:05.909655Z",
     "start_time": "2023-06-27T15:56:03.680531Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████| 5/5 [00:02<00:00,  2.25it/s]\n"
     ]
    }
   ],
   "source": [
    "inertia = []\n",
    "silhouette = {}\n",
    "\n",
    "for i in tqdm(range(2,7)): \n",
    "    \n",
    "    kmeans = KMeans(\n",
    "               n_clusters=i, \n",
    "               init='k-means++', \n",
    "               n_init=15,\n",
    "               max_iter=500, \n",
    "               random_state=322 \n",
    "    ) \n",
    "    \n",
    "    \n",
    "    kmeans.fit(X_train_multiclass)\n",
    "    \n",
    "    inertia.append(kmeans.inertia_)\n",
    "    silhouette[i] = silhouette_score(X_train_multiclass, kmeans.labels_, metric='euclidean')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "51e0edfc",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:56:07.118407Z",
     "start_time": "2023-06-27T15:56:05.912043Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "AttributeError",
     "evalue": "module 'matplotlib' has no attribute 'pyplot'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "Input \u001b[0;32mIn [67]\u001b[0m, in \u001b[0;36m<cell line: 18>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     14\u001b[0m plt\u001b[38;5;241m.\u001b[39mtitle(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mSilhouette Score\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m     15\u001b[0m plt\u001b[38;5;241m.\u001b[39mxlabel(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mЧисло кластеров\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m---> 18\u001b[0m \u001b[43mplt\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mshow\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/pyplot.py:421\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    377\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m    378\u001b[0m \u001b[38;5;124;03mDisplay all open figures.\u001b[39;00m\n\u001b[1;32m    379\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    418\u001b[0m \u001b[38;5;124;03mexplicitly there.\u001b[39;00m\n\u001b[1;32m    419\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m    420\u001b[0m _warn_if_gui_out_of_main_thread()\n\u001b[0;32m--> 421\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_get_backend_mod\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mshow\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib_inline/backend_inline.py:99\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(close, block)\u001b[0m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;66;03m# only call close('all') if any to close\u001b[39;00m\n\u001b[1;32m     97\u001b[0m \u001b[38;5;66;03m# close triggers gc.collect, which can be slow\u001b[39;00m\n\u001b[1;32m     98\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m close \u001b[38;5;129;01mand\u001b[39;00m Gcf\u001b[38;5;241m.\u001b[39mget_all_fig_managers():\n\u001b[0;32m---> 99\u001b[0m     \u001b[43mmatplotlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpyplot\u001b[49m\u001b[38;5;241m.\u001b[39mclose(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mall\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/_api/__init__.py:224\u001b[0m, in \u001b[0;36mcaching_module_getattr.<locals>.__getattr__\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m props:\n\u001b[1;32m    223\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m props[name]\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__get__\u001b[39m(instance)\n\u001b[0;32m--> 224\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\n\u001b[1;32m    225\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__module__\u001b[39m\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'matplotlib' has no attribute 'pyplot'"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "AttributeError",
     "evalue": "module 'matplotlib' has no attribute 'pyplot'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib_inline/backend_inline.py:99\u001b[0m, in \u001b[0;36mshow\u001b[0;34m(close, block)\u001b[0m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;66;03m# only call close('all') if any to close\u001b[39;00m\n\u001b[1;32m     97\u001b[0m \u001b[38;5;66;03m# close triggers gc.collect, which can be slow\u001b[39;00m\n\u001b[1;32m     98\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m close \u001b[38;5;129;01mand\u001b[39;00m Gcf\u001b[38;5;241m.\u001b[39mget_all_fig_managers():\n\u001b[0;32m---> 99\u001b[0m     \u001b[43mmatplotlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpyplot\u001b[49m\u001b[38;5;241m.\u001b[39mclose(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mall\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[0;32m/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/matplotlib/_api/__init__.py:224\u001b[0m, in \u001b[0;36mcaching_module_getattr.<locals>.__getattr__\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m props:\n\u001b[1;32m    223\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m props[name]\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__get__\u001b[39m(instance)\n\u001b[0;32m--> 224\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\n\u001b[1;32m    225\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__module__\u001b[39m\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m!r}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'matplotlib' has no attribute 'pyplot'"
     ]
    }
   ],
   "source": [
    "plt.subplot(1, 2, 1)\n",
    "\n",
    "plt.plot(range(2,len(inertia)+2), inertia) \n",
    "\n",
    "plt.xlabel('Число кластеров')\n",
    "plt.title('K-means Inertia')\n",
    "plt.grid(True)\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "\n",
    "plt.bar(range(len(silhouette)), list(silhouette.values()), align='center')\n",
    "plt.xticks(range(len(silhouette)), list(silhouette.keys()))\n",
    "plt.grid(True)\n",
    "plt.title('Silhouette Score')\n",
    "plt.xlabel('Число кластеров')\n",
    "\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "ad6b1f3a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-06-27T15:56:28.194366Z",
     "start_time": "2023-06-27T15:56:27.738551Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inertia: 0.0003709\n",
      "Silhouette Score: 0.5154\n"
     ]
    }
   ],
   "source": [
    "print(f'Inertia: {kmeans.inertia_:.4}')\n",
    "print(f'Silhouette Score: {silhouette_score(X_train_multiclass, kmeans.labels_):.4}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9227b8c0",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.8.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}