normal-pathology analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "d35558c3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import seaborn as sns\n",
    "import math as mat\n",
    "import statistics as stat\n",
    "\n",
    "from scipy.stats import chisquare\n",
    "from scipy.stats import mannwhitneyu\n",
    "from scipy.stats import ttest_ind\n",
    "\n",
    "from tqdm.notebook import tqdm\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.ensemble import GradientBoostingClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8ef89a49",
   "metadata": {},
   "outputs": [],
   "source": [
    "norm = pd.read_csv('test_data_analysis_healthy.csv')\n",
    "pathology = pd.read_csv('test_data_analysis_ill.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "24f75d22",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>angle</th>\n",
       "      <th>area XY</th>\n",
       "      <th>area ZY</th>\n",
       "      <th>area XZ</th>\n",
       "      <th>class</th>\n",
       "      <th>file</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>107.944086</td>\n",
       "      <td>2.564592e-07</td>\n",
       "      <td>2.900698e-07</td>\n",
       "      <td>6.207406e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>1.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>41.084692</td>\n",
       "      <td>6.271042e-07</td>\n",
       "      <td>2.981814e-07</td>\n",
       "      <td>1.390327e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>10.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>90.869783</td>\n",
       "      <td>2.103518e-07</td>\n",
       "      <td>7.744710e-08</td>\n",
       "      <td>2.363872e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>11.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>135.336736</td>\n",
       "      <td>5.141226e-07</td>\n",
       "      <td>5.578624e-07</td>\n",
       "      <td>1.405886e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>12.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>53.498848</td>\n",
       "      <td>3.335500e-08</td>\n",
       "      <td>1.620247e-07</td>\n",
       "      <td>3.173419e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>13.edf</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        angle       area XY       area ZY       area XZ  class    file\n",
       "0  107.944086  2.564592e-07  2.900698e-07  6.207406e-07      0   1.edf\n",
       "1   41.084692  6.271042e-07  2.981814e-07  1.390327e-06      0  10.edf\n",
       "2   90.869783  2.103518e-07  7.744710e-08  2.363872e-07      0  11.edf\n",
       "3  135.336736  5.141226e-07  5.578624e-07  1.405886e-06      0  12.edf\n",
       "4   53.498848  3.335500e-08  1.620247e-07  3.173419e-07      0  13.edf"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "norm.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "16e131d5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>angle</th>\n",
       "      <th>area XY</th>\n",
       "      <th>area ZY</th>\n",
       "      <th>area XZ</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>22.000000</td>\n",
       "      <td>2.200000e+01</td>\n",
       "      <td>2.200000e+01</td>\n",
       "      <td>2.200000e+01</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>74.755332</td>\n",
       "      <td>4.372251e-07</td>\n",
       "      <td>3.641214e-07</td>\n",
       "      <td>9.668481e-07</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>41.009816</td>\n",
       "      <td>3.630782e-07</td>\n",
       "      <td>3.219467e-07</td>\n",
       "      <td>5.198191e-07</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>7.292270</td>\n",
       "      <td>4.312402e-09</td>\n",
       "      <td>9.767185e-09</td>\n",
       "      <td>2.363872e-07</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>42.380854</td>\n",
       "      <td>1.890571e-07</td>\n",
       "      <td>1.086582e-07</td>\n",
       "      <td>5.797152e-07</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>67.783209</td>\n",
       "      <td>3.533064e-07</td>\n",
       "      <td>2.941256e-07</td>\n",
       "      <td>8.524447e-07</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>107.729650</td>\n",
       "      <td>6.225803e-07</td>\n",
       "      <td>5.575291e-07</td>\n",
       "      <td>1.352128e-06</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>135.336736</td>\n",
       "      <td>1.214741e-06</td>\n",
       "      <td>1.208213e-06</td>\n",
       "      <td>2.132315e-06</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            angle       area XY       area ZY       area XZ  class\n",
       "count   22.000000  2.200000e+01  2.200000e+01  2.200000e+01   22.0\n",
       "mean    74.755332  4.372251e-07  3.641214e-07  9.668481e-07    0.0\n",
       "std     41.009816  3.630782e-07  3.219467e-07  5.198191e-07    0.0\n",
       "min      7.292270  4.312402e-09  9.767185e-09  2.363872e-07    0.0\n",
       "25%     42.380854  1.890571e-07  1.086582e-07  5.797152e-07    0.0\n",
       "50%     67.783209  3.533064e-07  2.941256e-07  8.524447e-07    0.0\n",
       "75%    107.729650  6.225803e-07  5.575291e-07  1.352128e-06    0.0\n",
       "max    135.336736  1.214741e-06  1.208213e-06  2.132315e-06    0.0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "norm.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "964b7bd5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>angle</th>\n",
       "      <th>area XY</th>\n",
       "      <th>area ZY</th>\n",
       "      <th>area XZ</th>\n",
       "      <th>class</th>\n",
       "      <th>file</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>90.235582</td>\n",
       "      <td>2.424185e-07</td>\n",
       "      <td>6.401535e-07</td>\n",
       "      <td>1.104737e-08</td>\n",
       "      <td>1</td>\n",
       "      <td>61-1.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>58.311132</td>\n",
       "      <td>2.236326e-07</td>\n",
       "      <td>2.542808e-07</td>\n",
       "      <td>4.138347e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>62.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>49.263116</td>\n",
       "      <td>1.375380e-08</td>\n",
       "      <td>2.058379e-07</td>\n",
       "      <td>1.606429e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>63.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>35.716467</td>\n",
       "      <td>5.450050e-08</td>\n",
       "      <td>2.565739e-07</td>\n",
       "      <td>7.212155e-08</td>\n",
       "      <td>1</td>\n",
       "      <td>64.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>62.086588</td>\n",
       "      <td>1.317732e-06</td>\n",
       "      <td>8.439604e-07</td>\n",
       "      <td>2.596829e-06</td>\n",
       "      <td>1</td>\n",
       "      <td>65.edf</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       angle       area XY       area ZY       area XZ  class      file\n",
       "0  90.235582  2.424185e-07  6.401535e-07  1.104737e-08      1  61-1.edf\n",
       "1  58.311132  2.236326e-07  2.542808e-07  4.138347e-07      1    62.edf\n",
       "2  49.263116  1.375380e-08  2.058379e-07  1.606429e-07      1    63.edf\n",
       "3  35.716467  5.450050e-08  2.565739e-07  7.212155e-08      1    64.edf\n",
       "4  62.086588  1.317732e-06  8.439604e-07  2.596829e-06      1    65.edf"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pathology.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f2faffa7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>angle</th>\n",
       "      <th>area XY</th>\n",
       "      <th>area ZY</th>\n",
       "      <th>area XZ</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>15.000000</td>\n",
       "      <td>1.500000e+01</td>\n",
       "      <td>1.500000e+01</td>\n",
       "      <td>1.500000e+01</td>\n",
       "      <td>15.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>66.838697</td>\n",
       "      <td>3.671075e-07</td>\n",
       "      <td>2.525860e-07</td>\n",
       "      <td>5.772142e-07</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>51.705755</td>\n",
       "      <td>4.186123e-07</td>\n",
       "      <td>2.269778e-07</td>\n",
       "      <td>6.757654e-07</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2.491115</td>\n",
       "      <td>1.375380e-08</td>\n",
       "      <td>1.967466e-08</td>\n",
       "      <td>1.104737e-08</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>30.227751</td>\n",
       "      <td>6.313175e-08</td>\n",
       "      <td>9.536807e-08</td>\n",
       "      <td>1.277747e-07</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>52.491435</td>\n",
       "      <td>1.704104e-07</td>\n",
       "      <td>2.052165e-07</td>\n",
       "      <td>3.859385e-07</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>97.181040</td>\n",
       "      <td>6.761639e-07</td>\n",
       "      <td>2.975455e-07</td>\n",
       "      <td>7.220510e-07</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>153.584378</td>\n",
       "      <td>1.317732e-06</td>\n",
       "      <td>8.439604e-07</td>\n",
       "      <td>2.596829e-06</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            angle       area XY       area ZY       area XZ  class\n",
       "count   15.000000  1.500000e+01  1.500000e+01  1.500000e+01   15.0\n",
       "mean    66.838697  3.671075e-07  2.525860e-07  5.772142e-07    1.0\n",
       "std     51.705755  4.186123e-07  2.269778e-07  6.757654e-07    0.0\n",
       "min      2.491115  1.375380e-08  1.967466e-08  1.104737e-08    1.0\n",
       "25%     30.227751  6.313175e-08  9.536807e-08  1.277747e-07    1.0\n",
       "50%     52.491435  1.704104e-07  2.052165e-07  3.859385e-07    1.0\n",
       "75%     97.181040  6.761639e-07  2.975455e-07  7.220510e-07    1.0\n",
       "max    153.584378  1.317732e-06  8.439604e-07  2.596829e-06    1.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pathology.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "23029c3c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "statistic = 472.447, p=0.000\n",
      "angle normal : probably not gaussian\n"
     ]
    }
   ],
   "source": [
    "#normal distribution test\n",
    "stat,pvalue=chisquare(norm.angle)\n",
    "print('statistic = %.3f, p=%.3f'% (stat,pvalue))\n",
    "if pvalue>0.05:\n",
    "    print('angle normal : probably gaussian')\n",
    "else:\n",
    "     print('angle normal : probably not gaussian')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "70ea953b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "statistic = 559.987, p=0.000\n",
      "angle pathology : probably not gaussian\n"
     ]
    }
   ],
   "source": [
    "stat,pvalue=chisquare(pathology.angle)\n",
    "print('statistic = %.3f, p=%.3f'% (stat,pvalue))\n",
    "if pvalue>0.05:\n",
    "    print('angle pathology : probably gaussian')\n",
    "else:\n",
    "     print('angle pathology : probably not gaussian')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ba0868cf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "U1 = 187.0, p = 0.50599\n",
      "отличия отсутствуют\n"
     ]
    }
   ],
   "source": [
    "U1, p = mannwhitneyu(norm.angle, pathology.angle)\n",
    "print(f'U1 = {U1}, p = {round(p, 5)}')\n",
    "if p>0.05:\n",
    "    print('отличия отсутствуют')\n",
    "else:\n",
    "     print('существуют отличия')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5183528c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "statistic = 0.000, p=1.000\n",
      "area xOy normal : probably gaussian\n"
     ]
    }
   ],
   "source": [
    "#normal distribution test\n",
    "stat,pvalue=chisquare(norm['area XY'])\n",
    "print('statistic = %.3f, p=%.3f'% (stat,pvalue))\n",
    "if pvalue>0.05:\n",
    "    print('area xOy normal : probably gaussian')\n",
    "else:\n",
    "     print('area xOy normal : probably not gaussian')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a0ecc4a1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "statistic = 0.000, p=1.000\n",
      "area xOy pathology : probably gaussian\n"
     ]
    }
   ],
   "source": [
    "stat,pvalue=chisquare(pathology['area XY'])\n",
    "print('statistic = %.3f, p=%.3f'% (stat,pvalue))\n",
    "if pvalue>0.05:\n",
    "    print('area xOy pathology : probably gaussian')\n",
    "else:\n",
    "     print('area xOy pathology : probably not gaussian')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "1f815ee8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "p = 0.59116\n",
      "отличия отсутствуют\n"
     ]
    }
   ],
   "source": [
    "_, p = ttest_ind(norm['area XY'], pathology['area XY'])\n",
    "print(f'p = {round(p, 5)}')\n",
    "if p>0.05:\n",
    "    print('отличия отсутствуют')\n",
    "else:\n",
    "     print('существуют отличия')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "aff834f1",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "ce2f9514",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>angle</th>\n",
       "      <th>area XY</th>\n",
       "      <th>area ZY</th>\n",
       "      <th>area XZ</th>\n",
       "      <th>class</th>\n",
       "      <th>file</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>107.944086</td>\n",
       "      <td>2.564592e-07</td>\n",
       "      <td>2.900698e-07</td>\n",
       "      <td>6.207406e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>1.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>41.084692</td>\n",
       "      <td>6.271042e-07</td>\n",
       "      <td>2.981814e-07</td>\n",
       "      <td>1.390327e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>10.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>90.869783</td>\n",
       "      <td>2.103518e-07</td>\n",
       "      <td>7.744710e-08</td>\n",
       "      <td>2.363872e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>11.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>135.336736</td>\n",
       "      <td>5.141226e-07</td>\n",
       "      <td>5.578624e-07</td>\n",
       "      <td>1.405886e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>12.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>53.498848</td>\n",
       "      <td>3.335500e-08</td>\n",
       "      <td>1.620247e-07</td>\n",
       "      <td>3.173419e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>13.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>61.639357</td>\n",
       "      <td>1.214741e-06</td>\n",
       "      <td>4.389717e-07</td>\n",
       "      <td>1.223365e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>14.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>51.209468</td>\n",
       "      <td>6.125123e-08</td>\n",
       "      <td>1.167028e-07</td>\n",
       "      <td>6.508041e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>15.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>92.805624</td>\n",
       "      <td>1.991401e-07</td>\n",
       "      <td>7.858660e-07</td>\n",
       "      <td>2.132315e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>16.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>72.723865</td>\n",
       "      <td>4.984853e-07</td>\n",
       "      <td>6.408894e-07</td>\n",
       "      <td>6.388547e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>17.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>124.932783</td>\n",
       "      <td>1.102874e-06</td>\n",
       "      <td>1.208213e-06</td>\n",
       "      <td>1.560849e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>19.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>62.842552</td>\n",
       "      <td>5.520333e-08</td>\n",
       "      <td>1.059767e-07</td>\n",
       "      <td>1.797727e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>20.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>107.086343</td>\n",
       "      <td>1.066235e-06</td>\n",
       "      <td>8.220356e-07</td>\n",
       "      <td>3.836831e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>21.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>99.086402</td>\n",
       "      <td>1.856961e-07</td>\n",
       "      <td>9.767185e-09</td>\n",
       "      <td>1.226803e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>22.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>46.269338</td>\n",
       "      <td>2.342741e-07</td>\n",
       "      <td>1.021741e-07</td>\n",
       "      <td>4.217848e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>23.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>14.673529</td>\n",
       "      <td>4.720966e-08</td>\n",
       "      <td>1.171263e-07</td>\n",
       "      <td>8.398457e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>24.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>134.916963</td>\n",
       "      <td>4.062827e-07</td>\n",
       "      <td>4.167434e-08</td>\n",
       "      <td>5.043590e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>3.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>20.588012</td>\n",
       "      <td>4.312402e-09</td>\n",
       "      <td>1.842030e-08</td>\n",
       "      <td>6.483387e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>4.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>120.705387</td>\n",
       "      <td>3.003301e-07</td>\n",
       "      <td>4.259191e-07</td>\n",
       "      <td>1.456431e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>5.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>126.911062</td>\n",
       "      <td>6.090085e-07</td>\n",
       "      <td>7.395871e-07</td>\n",
       "      <td>1.237531e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>6.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>35.212431</td>\n",
       "      <td>7.494674e-07</td>\n",
       "      <td>5.565292e-07</td>\n",
       "      <td>8.650437e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>7.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>7.292270</td>\n",
       "      <td>5.000934e-07</td>\n",
       "      <td>1.823594e-07</td>\n",
       "      <td>1.146202e-06</td>\n",
       "      <td>0</td>\n",
       "      <td>8.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>36.987782</td>\n",
       "      <td>7.429546e-07</td>\n",
       "      <td>3.128738e-07</td>\n",
       "      <td>5.660401e-07</td>\n",
       "      <td>0</td>\n",
       "      <td>9.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>90.235582</td>\n",
       "      <td>2.424185e-07</td>\n",
       "      <td>6.401535e-07</td>\n",
       "      <td>1.104737e-08</td>\n",
       "      <td>1</td>\n",
       "      <td>61-1.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>58.311132</td>\n",
       "      <td>2.236326e-07</td>\n",
       "      <td>2.542808e-07</td>\n",
       "      <td>4.138347e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>62.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>49.263116</td>\n",
       "      <td>1.375380e-08</td>\n",
       "      <td>2.058379e-07</td>\n",
       "      <td>1.606429e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>63.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>35.716467</td>\n",
       "      <td>5.450050e-08</td>\n",
       "      <td>2.565739e-07</td>\n",
       "      <td>7.212155e-08</td>\n",
       "      <td>1</td>\n",
       "      <td>64.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>62.086588</td>\n",
       "      <td>1.317732e-06</td>\n",
       "      <td>8.439604e-07</td>\n",
       "      <td>2.596829e-06</td>\n",
       "      <td>1</td>\n",
       "      <td>65.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>150.136713</td>\n",
       "      <td>1.005861e-06</td>\n",
       "      <td>3.385170e-07</td>\n",
       "      <td>1.284320e-06</td>\n",
       "      <td>1</td>\n",
       "      <td>66.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2.491115</td>\n",
       "      <td>6.886565e-08</td>\n",
       "      <td>8.690495e-08</td>\n",
       "      <td>4.498437e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>67.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>10.355798</td>\n",
       "      <td>7.762331e-07</td>\n",
       "      <td>1.038312e-07</td>\n",
       "      <td>9.297623e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>68.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>16.444659</td>\n",
       "      <td>5.966210e-08</td>\n",
       "      <td>2.006909e-07</td>\n",
       "      <td>9.490650e-08</td>\n",
       "      <td>1</td>\n",
       "      <td>69.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>52.491435</td>\n",
       "      <td>5.760947e-07</td>\n",
       "      <td>2.052165e-07</td>\n",
       "      <td>5.143396e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>70.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>151.391161</td>\n",
       "      <td>1.060914e-07</td>\n",
       "      <td>5.337871e-08</td>\n",
       "      <td>6.344864e-08</td>\n",
       "      <td>1</td>\n",
       "      <td>71.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>41.206780</td>\n",
       "      <td>3.346671e-08</td>\n",
       "      <td>1.967466e-08</td>\n",
       "      <td>3.218536e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>72.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>104.126498</td>\n",
       "      <td>1.704104e-07</td>\n",
       "      <td>1.452480e-07</td>\n",
       "      <td>3.090789e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>73.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>24.739035</td>\n",
       "      <td>6.660140e-08</td>\n",
       "      <td>6.028857e-08</td>\n",
       "      <td>3.859385e-07</td>\n",
       "      <td>1</td>\n",
       "      <td>74.edf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>153.584378</td>\n",
       "      <td>7.912891e-07</td>\n",
       "      <td>3.742325e-07</td>\n",
       "      <td>1.050246e-06</td>\n",
       "      <td>1</td>\n",
       "      <td>75.edf</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         angle       area XY       area ZY       area XZ  class      file\n",
       "0   107.944086  2.564592e-07  2.900698e-07  6.207406e-07      0     1.edf\n",
       "1    41.084692  6.271042e-07  2.981814e-07  1.390327e-06      0    10.edf\n",
       "2    90.869783  2.103518e-07  7.744710e-08  2.363872e-07      0    11.edf\n",
       "3   135.336736  5.141226e-07  5.578624e-07  1.405886e-06      0    12.edf\n",
       "4    53.498848  3.335500e-08  1.620247e-07  3.173419e-07      0    13.edf\n",
       "5    61.639357  1.214741e-06  4.389717e-07  1.223365e-06      0    14.edf\n",
       "6    51.209468  6.125123e-08  1.167028e-07  6.508041e-07      0    15.edf\n",
       "7    92.805624  1.991401e-07  7.858660e-07  2.132315e-06      0    16.edf\n",
       "8    72.723865  4.984853e-07  6.408894e-07  6.388547e-07      0    17.edf\n",
       "9   124.932783  1.102874e-06  1.208213e-06  1.560849e-06      0    19.edf\n",
       "10   62.842552  5.520333e-08  1.059767e-07  1.797727e-06      0    20.edf\n",
       "11  107.086343  1.066235e-06  8.220356e-07  3.836831e-07      0    21.edf\n",
       "12   99.086402  1.856961e-07  9.767185e-09  1.226803e-06      0    22.edf\n",
       "13   46.269338  2.342741e-07  1.021741e-07  4.217848e-07      0    23.edf\n",
       "14   14.673529  4.720966e-08  1.171263e-07  8.398457e-07      0    24.edf\n",
       "15  134.916963  4.062827e-07  4.167434e-08  5.043590e-07      0     3.edf\n",
       "16   20.588012  4.312402e-09  1.842030e-08  6.483387e-07      0     4.edf\n",
       "17  120.705387  3.003301e-07  4.259191e-07  1.456431e-06      0     5.edf\n",
       "18  126.911062  6.090085e-07  7.395871e-07  1.237531e-06      0     6.edf\n",
       "19   35.212431  7.494674e-07  5.565292e-07  8.650437e-07      0     7.edf\n",
       "20    7.292270  5.000934e-07  1.823594e-07  1.146202e-06      0     8.edf\n",
       "21   36.987782  7.429546e-07  3.128738e-07  5.660401e-07      0     9.edf\n",
       "0    90.235582  2.424185e-07  6.401535e-07  1.104737e-08      1  61-1.edf\n",
       "1    58.311132  2.236326e-07  2.542808e-07  4.138347e-07      1    62.edf\n",
       "2    49.263116  1.375380e-08  2.058379e-07  1.606429e-07      1    63.edf\n",
       "3    35.716467  5.450050e-08  2.565739e-07  7.212155e-08      1    64.edf\n",
       "4    62.086588  1.317732e-06  8.439604e-07  2.596829e-06      1    65.edf\n",
       "5   150.136713  1.005861e-06  3.385170e-07  1.284320e-06      1    66.edf\n",
       "6     2.491115  6.886565e-08  8.690495e-08  4.498437e-07      1    67.edf\n",
       "7    10.355798  7.762331e-07  1.038312e-07  9.297623e-07      1    68.edf\n",
       "8    16.444659  5.966210e-08  2.006909e-07  9.490650e-08      1    69.edf\n",
       "9    52.491435  5.760947e-07  2.052165e-07  5.143396e-07      1    70.edf\n",
       "10  151.391161  1.060914e-07  5.337871e-08  6.344864e-08      1    71.edf\n",
       "11   41.206780  3.346671e-08  1.967466e-08  3.218536e-07      1    72.edf\n",
       "12  104.126498  1.704104e-07  1.452480e-07  3.090789e-07      1    73.edf\n",
       "13   24.739035  6.660140e-08  6.028857e-08  3.859385e-07      1    74.edf\n",
       "14  153.584378  7.912891e-07  3.742325e-07  1.050246e-06      1    75.edf"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frames = [norm, pathology]\n",
    "result = pd.concat(frames)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "1e4d7707",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>angle</th>\n",
       "      <th>area XY</th>\n",
       "      <th>area ZY</th>\n",
       "      <th>area XZ</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>107.944086</td>\n",
       "      <td>2.564592e-07</td>\n",
       "      <td>2.900698e-07</td>\n",
       "      <td>6.207406e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>41.084692</td>\n",
       "      <td>6.271042e-07</td>\n",
       "      <td>2.981814e-07</td>\n",
       "      <td>1.390327e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>90.869783</td>\n",
       "      <td>2.103518e-07</td>\n",
       "      <td>7.744710e-08</td>\n",
       "      <td>2.363872e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>135.336736</td>\n",
       "      <td>5.141226e-07</td>\n",
       "      <td>5.578624e-07</td>\n",
       "      <td>1.405886e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>53.498848</td>\n",
       "      <td>3.335500e-08</td>\n",
       "      <td>1.620247e-07</td>\n",
       "      <td>3.173419e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>61.639357</td>\n",
       "      <td>1.214741e-06</td>\n",
       "      <td>4.389717e-07</td>\n",
       "      <td>1.223365e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>51.209468</td>\n",
       "      <td>6.125123e-08</td>\n",
       "      <td>1.167028e-07</td>\n",
       "      <td>6.508041e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>92.805624</td>\n",
       "      <td>1.991401e-07</td>\n",
       "      <td>7.858660e-07</td>\n",
       "      <td>2.132315e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>72.723865</td>\n",
       "      <td>4.984853e-07</td>\n",
       "      <td>6.408894e-07</td>\n",
       "      <td>6.388547e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>124.932783</td>\n",
       "      <td>1.102874e-06</td>\n",
       "      <td>1.208213e-06</td>\n",
       "      <td>1.560849e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>62.842552</td>\n",
       "      <td>5.520333e-08</td>\n",
       "      <td>1.059767e-07</td>\n",
       "      <td>1.797727e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>107.086343</td>\n",
       "      <td>1.066235e-06</td>\n",
       "      <td>8.220356e-07</td>\n",
       "      <td>3.836831e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>99.086402</td>\n",
       "      <td>1.856961e-07</td>\n",
       "      <td>9.767185e-09</td>\n",
       "      <td>1.226803e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>46.269338</td>\n",
       "      <td>2.342741e-07</td>\n",
       "      <td>1.021741e-07</td>\n",
       "      <td>4.217848e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>14.673529</td>\n",
       "      <td>4.720966e-08</td>\n",
       "      <td>1.171263e-07</td>\n",
       "      <td>8.398457e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>134.916963</td>\n",
       "      <td>4.062827e-07</td>\n",
       "      <td>4.167434e-08</td>\n",
       "      <td>5.043590e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>20.588012</td>\n",
       "      <td>4.312402e-09</td>\n",
       "      <td>1.842030e-08</td>\n",
       "      <td>6.483387e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>120.705387</td>\n",
       "      <td>3.003301e-07</td>\n",
       "      <td>4.259191e-07</td>\n",
       "      <td>1.456431e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>126.911062</td>\n",
       "      <td>6.090085e-07</td>\n",
       "      <td>7.395871e-07</td>\n",
       "      <td>1.237531e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>35.212431</td>\n",
       "      <td>7.494674e-07</td>\n",
       "      <td>5.565292e-07</td>\n",
       "      <td>8.650437e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>7.292270</td>\n",
       "      <td>5.000934e-07</td>\n",
       "      <td>1.823594e-07</td>\n",
       "      <td>1.146202e-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>36.987782</td>\n",
       "      <td>7.429546e-07</td>\n",
       "      <td>3.128738e-07</td>\n",
       "      <td>5.660401e-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>90.235582</td>\n",
       "      <td>2.424185e-07</td>\n",
       "      <td>6.401535e-07</td>\n",
       "      <td>1.104737e-08</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>58.311132</td>\n",
       "      <td>2.236326e-07</td>\n",
       "      <td>2.542808e-07</td>\n",
       "      <td>4.138347e-07</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>49.263116</td>\n",
       "      <td>1.375380e-08</td>\n",
       "      <td>2.058379e-07</td>\n",
       "      <td>1.606429e-07</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>35.716467</td>\n",
       "      <td>5.450050e-08</td>\n",
       "      <td>2.565739e-07</td>\n",
       "      <td>7.212155e-08</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>62.086588</td>\n",
       "      <td>1.317732e-06</td>\n",
       "      <td>8.439604e-07</td>\n",
       "      <td>2.596829e-06</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>150.136713</td>\n",
       "      <td>1.005861e-06</td>\n",
       "      <td>3.385170e-07</td>\n",
       "      <td>1.284320e-06</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2.491115</td>\n",
       "      <td>6.886565e-08</td>\n",
       "      <td>8.690495e-08</td>\n",
       "      <td>4.498437e-07</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>10.355798</td>\n",
       "      <td>7.762331e-07</td>\n",
       "      <td>1.038312e-07</td>\n",
       "      <td>9.297623e-07</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>16.444659</td>\n",
       "      <td>5.966210e-08</td>\n",
       "      <td>2.006909e-07</td>\n",
       "      <td>9.490650e-08</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>52.491435</td>\n",
       "      <td>5.760947e-07</td>\n",
       "      <td>2.052165e-07</td>\n",
       "      <td>5.143396e-07</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>151.391161</td>\n",
       "      <td>1.060914e-07</td>\n",
       "      <td>5.337871e-08</td>\n",
       "      <td>6.344864e-08</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>41.206780</td>\n",
       "      <td>3.346671e-08</td>\n",
       "      <td>1.967466e-08</td>\n",
       "      <td>3.218536e-07</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>104.126498</td>\n",
       "      <td>1.704104e-07</td>\n",
       "      <td>1.452480e-07</td>\n",
       "      <td>3.090789e-07</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>24.739035</td>\n",
       "      <td>6.660140e-08</td>\n",
       "      <td>6.028857e-08</td>\n",
       "      <td>3.859385e-07</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>153.584378</td>\n",
       "      <td>7.912891e-07</td>\n",
       "      <td>3.742325e-07</td>\n",
       "      <td>1.050246e-06</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         angle       area XY       area ZY       area XZ  class\n",
       "0   107.944086  2.564592e-07  2.900698e-07  6.207406e-07      0\n",
       "1    41.084692  6.271042e-07  2.981814e-07  1.390327e-06      0\n",
       "2    90.869783  2.103518e-07  7.744710e-08  2.363872e-07      0\n",
       "3   135.336736  5.141226e-07  5.578624e-07  1.405886e-06      0\n",
       "4    53.498848  3.335500e-08  1.620247e-07  3.173419e-07      0\n",
       "5    61.639357  1.214741e-06  4.389717e-07  1.223365e-06      0\n",
       "6    51.209468  6.125123e-08  1.167028e-07  6.508041e-07      0\n",
       "7    92.805624  1.991401e-07  7.858660e-07  2.132315e-06      0\n",
       "8    72.723865  4.984853e-07  6.408894e-07  6.388547e-07      0\n",
       "9   124.932783  1.102874e-06  1.208213e-06  1.560849e-06      0\n",
       "10   62.842552  5.520333e-08  1.059767e-07  1.797727e-06      0\n",
       "11  107.086343  1.066235e-06  8.220356e-07  3.836831e-07      0\n",
       "12   99.086402  1.856961e-07  9.767185e-09  1.226803e-06      0\n",
       "13   46.269338  2.342741e-07  1.021741e-07  4.217848e-07      0\n",
       "14   14.673529  4.720966e-08  1.171263e-07  8.398457e-07      0\n",
       "15  134.916963  4.062827e-07  4.167434e-08  5.043590e-07      0\n",
       "16   20.588012  4.312402e-09  1.842030e-08  6.483387e-07      0\n",
       "17  120.705387  3.003301e-07  4.259191e-07  1.456431e-06      0\n",
       "18  126.911062  6.090085e-07  7.395871e-07  1.237531e-06      0\n",
       "19   35.212431  7.494674e-07  5.565292e-07  8.650437e-07      0\n",
       "20    7.292270  5.000934e-07  1.823594e-07  1.146202e-06      0\n",
       "21   36.987782  7.429546e-07  3.128738e-07  5.660401e-07      0\n",
       "0    90.235582  2.424185e-07  6.401535e-07  1.104737e-08      1\n",
       "1    58.311132  2.236326e-07  2.542808e-07  4.138347e-07      1\n",
       "2    49.263116  1.375380e-08  2.058379e-07  1.606429e-07      1\n",
       "3    35.716467  5.450050e-08  2.565739e-07  7.212155e-08      1\n",
       "4    62.086588  1.317732e-06  8.439604e-07  2.596829e-06      1\n",
       "5   150.136713  1.005861e-06  3.385170e-07  1.284320e-06      1\n",
       "6     2.491115  6.886565e-08  8.690495e-08  4.498437e-07      1\n",
       "7    10.355798  7.762331e-07  1.038312e-07  9.297623e-07      1\n",
       "8    16.444659  5.966210e-08  2.006909e-07  9.490650e-08      1\n",
       "9    52.491435  5.760947e-07  2.052165e-07  5.143396e-07      1\n",
       "10  151.391161  1.060914e-07  5.337871e-08  6.344864e-08      1\n",
       "11   41.206780  3.346671e-08  1.967466e-08  3.218536e-07      1\n",
       "12  104.126498  1.704104e-07  1.452480e-07  3.090789e-07      1\n",
       "13   24.739035  6.660140e-08  6.028857e-08  3.859385e-07      1\n",
       "14  153.584378  7.912891e-07  3.742325e-07  1.050246e-06      1"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.pop('file')\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "9e7e3148",
   "metadata": {},
   "outputs": [],
   "source": [
    "def apply_lda(X_train, X_test, Y_train):\n",
    "     \n",
    "    \n",
    "    x_scaler = StandardScaler()\n",
    "    X_train = x_scaler.fit_transform(X_train)\n",
    "    X_test = x_scaler.transform(X_test)\n",
    "    \n",
    "    LDA = LinearDiscriminantAnalysis() \n",
    "    LDA.fit(X_train, Y_train)\n",
    "    \n",
    "    X_train_new = LDA.transform(X_train)\n",
    "    X_test_new = LDA.transform(X_test)\n",
    "    \n",
    "    return X_train_new, X_test_new, LDA, x_scaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "b9bafb63",
   "metadata": {},
   "outputs": [],
   "source": [
    "def apply_pca(X_train, X_test, Y_train=0):\n",
    "    \n",
    "    x_scaler = StandardScaler()\n",
    "    X_train_scaled = x_scaler.fit_transform(X_train)\n",
    "    X_test_scaled = x_scaler.transform(X_test)\n",
    "\n",
    "    pca = PCA(n_components=2)\n",
    "    pca.fit(X_train_scaled)\n",
    "\n",
    "    PC_train, PC_test = pca.transform(X_train_scaled), pca.transform(X_test_scaled)\n",
    "\n",
    "    return PC_train, PC_test, pca, x_scaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "45730c30",
   "metadata": {},
   "outputs": [],
   "source": [
    "Y = result['class']\n",
    "X = result.drop(['class'], axis=1)\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.3, random_state=10, stratify=Y, shuffle=True)\n",
    "X_train, X_test, DR, x_scaler = apply_pca(X_train.values, X_test.values, Y_train.values)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "a9a25571",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f1 score:0.667\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x27222932470>"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = SVC(kernel = 'linear') \n",
    "model.fit(X_train, Y_train)\n",
    "prediction = model.predict(X_test) \n",
    "cm = confusion_matrix(prediction, Y_test)\n",
    "\n",
    "#print(f'f1 score:{f1_score(Y_test, prediction, average=\"binary\", pos_label = \"center\"):.3f}')\n",
    "print(f'f1 score:{f1_score(Y_test, prediction, average=\"macro\"):.3f}')\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm.T, display_labels=model.classes_)\n",
    "disp.plot(cmap = 'Blues')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "979c21cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 8))\n",
    "plt.xlim(X_train[:, 0].min()-1, X_train[:, 0].max()+1)\n",
    "plt.ylim(X_train[:, 1].min()-1, X_train[:, 1].max()+1)\n",
    "\n",
    "\n",
    "colors = ['blue', 'red']\n",
    "\n",
    "for i, target in enumerate(pd.unique(Y_train)):\n",
    "    index = np.where(Y_train == target)\n",
    "    plt.scatter(X_train[index, 0], \n",
    "                X_train[index, 1], \n",
    "                c=colors[i], \n",
    "                label=target,\n",
    "                linewidths=1,\n",
    "                )\n",
    "plt.legend(fontsize='large')\n",
    "plt.xlabel('1-st PCA component')\n",
    "plt.ylabel('2-nd PCA component')\n",
    "\n",
    "\n",
    "\n",
    "for i, target in enumerate(pd.unique(Y_train)):\n",
    "    index = np.where(Y_test == target)\n",
    "    plt.scatter(X_test[index, 0], \n",
    "                X_test[index, 1], \n",
    "                c=colors[i], \n",
    "                label=target,\n",
    "                linewidths=8,\n",
    "                marker='^',\n",
    "                alpha=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "9b76d9c1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Точность модели на обучающей выборке:  0.72\n",
      "Точность модели на тестовой выборке:  0.67\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfYAAAG2CAYAAACNs6TQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAt/0lEQVR4nO3de3RU5dn38d8OmEkgmQBaEgIBsdEAAuGgYmIVaEFEXiSlD1qKJVDgfamgHMQDtpzF+EgRoSAHUSNWCgoSKx5oig0HiUqAuMBCWhBN0CRYkYQESSCz3z8w044hMMPMZJzZ3w9rr+Xcsw9XXCyuXNd9770N0zRNAQCAkBAW6AAAAIDvkNgBAAghJHYAAEIIiR0AgBBCYgcAIISQ2AEACCEkdgAAQgiJHQCAEEJiBwAghJDYAQAIISR2AAD87Mknn5RhGJo8efJF93vttdfUoUMHRUREqEuXLnr77bc9vhaJHQAAP9q9e7dWrlyprl27XnS/Xbt2afjw4RozZoz27duntLQ0paWl6cCBAx5dz+AlMAAA+EdFRYV69OihZ599Vo8//ri6deumZ5555oL73nPPPaqsrNTmzZudYzfffLO6deumFStWuH3Nxt4GHUgOh0NffvmloqOjZRhGoMMBAHjINE2dOnVK8fHxCgvzXxP5zJkzqq6u9vo8pmnWyTc2m002m+2C+0+YMEGDBg1Sv3799Pjjj1/03Lm5uZo6darL2IABA5SVleVRjEGd2L/88kslJCQEOgwAgJeKiorUpk0bv5z7zJkzioy+Ujp32utzRUVFqaKiwmVs1qxZmj17dp19161bp71792r37t1unbukpESxsbEuY7GxsSopKfEoxqBO7NHR0ZKk8E7pMhqFBzgawD86D00LdAiA39RUnVb+k8Oc/577Q3V1tXTutGyd0iVvckVNtSr+8ZKKiopkt9udwxeq1ouKijRp0iRlZ2crIiLi8q95GYI6sde2Q4xG4SR2hKzGEU0DHQLgdw0yndo4wqtcYRrnpwrsdrtLYr+QPXv26Pjx4+rRo4dzrKamRtu3b9fSpUtVVVWlRo0auRwTFxen0tJSl7HS0lLFxcV5FCer4gEA1mBIMgwvNvcv9bOf/Uz79+9Xfn6+c7vhhhs0YsQI5efn10nqkpSSkqKtW7e6jGVnZyslJcWjHzOoK3YAANxmhJ3fvDneTdHR0ercubPLWNOmTXXllVc6x0eOHKnWrVsrIyNDkjRp0iT17t1bCxcu1KBBg7Ru3Trl5eVp1apVHoVJxQ4AQAAUFhaquLjY+Tk1NVVr167VqlWrlJycrA0bNigrK6vOLwiXQsUOALCG2pa6N8d7IScn56KfJWnYsGEaNmyYV9chsQMArKEBW/GBFBxRAgAAt1CxAwCsIcCt+IZCYgcAWISXrfggaXIHR5QAAMAtVOwAAGugFQ8AQAhhVTwAAAg2VOwAAGugFQ8AQAixSCuexA4AsAaLVOzB8esHAABwCxU7AMAaaMUDABBCDMPLxE4rHgAANDAqdgCANYQZ5zdvjg8CJHYAgDVYZI49OKIEAABuoWIHAFiDRe5jJ7EDAKyBVjwAAAg2VOwAAGugFQ8AQAixSCuexA4AsAaLVOzB8esHAABwCxU7AMAaaMUDABBCaMUDAIBgQ8UOALAIL1vxQVILk9gBANZAKx4AAAQbKnYAgDUYhper4oOjYiexAwCswSK3uwVHlAAAwC1U7AAAa7DI4jkSOwDAGizSiiexAwCswSIVe3D8+gEAANxCxQ4AsAZa8QAAhBBa8QAAINhQsQMALMEwDBkWqNhJ7AAAS7BKYqcVDwBACKFiBwBYg/Hd5s3xQYDEDgCwBFrxAAAg6FCxAwAswSoVO4kdAGAJJHYAAEKIVRI7c+wAAIQQEjsAwBoMH2weWL58ubp27Sq73S673a6UlBS988479e6fmZnp7CrUbhERER7+kLTiAQAW0dCt+DZt2ujJJ5/UtddeK9M09dJLL2nIkCHat2+frr/++gseY7fbVVBQ4BKzp0jsAAD4weDBg10+z58/X8uXL9cHH3xQb2I3DENxcXFeXZdWPADAEs6/tdXwYjt/nvLycpetqqrqkteuqanRunXrVFlZqZSUlHr3q6ioULt27ZSQkKAhQ4bok08+8fjnJLEDACzBkDdJ3ZDx3SR7QkKCYmJinFtGRka919y/f7+ioqJks9k0fvx4bdq0SZ06dbrgvklJSXrhhRf0xhtv6E9/+pMcDodSU1N17Ngxj35OWvEAAHigqKhIdrvd+dlms9W7b1JSkvLz81VWVqYNGzYoPT1d27Ztu2ByT0lJcanmU1NT1bFjR61cuVLz5s1zOz4SOwDAEny1eK52lbs7wsPDlZiYKEnq2bOndu/ercWLF2vlypWXPPaKK65Q9+7ddfjwYY/CpBUPALCGBr7d7UIcDodbc/LS+Xn5/fv3q1WrVh5dg4odAAA/mD59ugYOHKi2bdvq1KlTWrt2rXJycrRlyxZJ0siRI9W6dWvnHP3cuXN18803KzExUSdPntSCBQv0+eefa+zYsR5dl8QOALAGL1vxpofHHj9+XCNHjlRxcbFiYmLUtWtXbdmyRf3795ckFRYWKizsP43zb775RuPGjVNJSYmaN2+unj17ateuXfUutqsPiR0AYAnezrF7euzzzz9/0e9zcnJcPi9atEiLFi3yNKw6SOwAAEto6MQeKCyeAwAghFCxAwCswduV7cFRsJPYAQDWQCseAAAEHSp2AIAlWKViJ7EDACzBKomdVjwAACGEih0AYAlWqdhJ7AAAa7DI7W604gEACCFU7AAAS6AVDwBACCGxAwAQQqyS2JljBwAghFCxAwCswSKr4knsAABLoBUPAACCDhU7PDY5vb9mTRyi5X/+ux57emOgwwG8dldyKw3pFq84e4Qk6bOvT+ul3M/10dETAY4MvkTF3oCWLVumq6++WhEREerVq5c++uijQIeEenTv1Fajfn6LDvzzWKBDAXzmq1PVWrX9qP7vy3v1//60V3sLv9H8tOt19ZVNAh0afMiQ4Uzul7UFySR7wBP7+vXrNXXqVM2aNUt79+5VcnKyBgwYoOPHjwc6NHxP08hwrZo7SpOe+LNOnvo20OEAPpP76df68OgJfXHyWx375ls9v/MzfVtdo06t7IEODfBYwBP7008/rXHjxmn06NHq1KmTVqxYoSZNmuiFF14IdGj4ngUP36O/vn9A2z4qCHQogN+EGdJPk36kiCsa6ZPi8kCHAx/yqlr3so3fkAI6x15dXa09e/Zo+vTpzrGwsDD169dPubm5AYwM3ze0f08ld0jQT9OfCnQogF+0v6qpnv1Vd4U3DtO31TWa8cYn+vzr04EOC77E7W7+9+9//1s1NTWKjY11GY+NjdWhQ4fq7F9VVaWqqirn5/JyfptuCK1jmynjwV9o6MSlqqo+F+hwAL8oOnFaY9fkqamtsXpf9yNNH5ikSes/Jrkj6ATVqviMjAzNmTMn0GFYTnKHtmp5pV05Lz/iHGvcuJFSu/9Y44bdpthbJsvhMAMYIeC9cw5TX5w8I0n6Z2mFOsRF6xc9Wuvp7H8FODL4ilVWxQc0sV911VVq1KiRSktLXcZLS0sVFxdXZ//p06dr6tSpzs/l5eVKSEjwe5xWt313gVJ/Od9lbOnMe/Wvz0q1eE02SR0hyTCk8EYBX4YEHyKxN4Dw8HD17NlTW7duVVpamiTJ4XBo69atmjhxYp39bTabbDZbA0eJitNVOnik2GXs9LfVOlFWWWccCEbjbm2vD4+e0PHyM4oMb6x+HVuqW0IzPbRhf6BDgw8ZxvnNm+ODQcBb8VOnTlV6erpuuOEG3XTTTXrmmWdUWVmp0aNHBzo0ABbRrMkVemxgB7VoGq7K6nP69KtKPbRhv/Z8/k2gQwM8FvDEfs899+irr77SzJkzVVJSom7duundd9+ts6AOPyyDxy8OdAiAzyzY8s9Ah4AGcL5i96YV78Ng/CjgiV2SJk6ceMHWOwAAPuNlKz5YbndjZQgAACHkB1GxAwDgb6yKBwAghFhlVTyteAAAQggVOwDAEsLCDIWFXX7ZbXpxbEMisQMALIFWPAAACDpU7AAAS2BVPAAAIcQqrXgSOwDAEqxSsTPHDgBACKFiBwBYglUqdhI7AMASrDLHTiseAIAQQsUOALAEQ1624oPkva0kdgCAJdCKBwAAQYeKHQBgCayKBwAghNCKBwAAQYfEDgCwhNpWvDebJ5YvX66uXbvKbrfLbrcrJSVF77zzzkWPee2119ShQwdFRESoS5cuevvttz3+OUnsAABLqG3Fe7N5ok2bNnryySe1Z88e5eXl6ac//amGDBmiTz755IL779q1S8OHD9eYMWO0b98+paWlKS0tTQcOHPDouiR2AIAlNHTFPnjwYN1555269tprdd1112n+/PmKiorSBx98cMH9Fy9erDvuuEMPPfSQOnbsqHnz5qlHjx5aunSpR9clsQMA4IHy8nKXraqq6pLH1NTUaN26daqsrFRKSsoF98nNzVW/fv1cxgYMGKDc3FyP4iOxAwCswds2/HcFe0JCgmJiYpxbRkZGvZfcv3+/oqKiZLPZNH78eG3atEmdOnW64L4lJSWKjY11GYuNjVVJSYlHPya3uwEALMFX97EXFRXJbrc7x202W73HJCUlKT8/X2VlZdqwYYPS09O1bdu2epO7L5DYAQDwQO0qd3eEh4crMTFRktSzZ0/t3r1bixcv1sqVK+vsGxcXp9LSUpex0tJSxcXFeRQfrXgAgCU09Kr4C3E4HPXOyaekpGjr1q0uY9nZ2fXOydeHih0AYAkN/UjZ6dOna+DAgWrbtq1OnTqltWvXKicnR1u2bJEkjRw5Uq1bt3bO0U+aNEm9e/fWwoULNWjQIK1bt055eXlatWqVR9clsQMA4AfHjx/XyJEjVVxcrJiYGHXt2lVbtmxR//79JUmFhYUKC/tP4zw1NVVr167V73//ez322GO69tprlZWVpc6dO3t0XRI7AMASGvpZ8c8///xFv8/JyakzNmzYMA0bNsyzC30PiR0AYAlWebsbi+cAAAghVOwAAEuwSsVOYgcAWIJV3sdOYgcAWIJVKnbm2AEACCFU7AAAS6AVDwBACKEVDwAAgg4VOwDAEgx52Yr3WST+RWIHAFhCmGEozIvM7s2xDYlWPAAAIYSKHQBgCayKBwAghFhlVTyJHQBgCWHG+c2b44MBc+wAAIQQKnYAgDUYXrbTg6RiJ7EDACzBKovnaMUDABBCqNgBAJZgfPfHm+ODAYkdAGAJrIoHAABBh4odAGAJPKDmv/zlL39x+4R33XXXZQcDAIC/WGVVvFuJPS0tza2TGYahmpoab+IBAABecCuxOxwOf8cBAIBfWeW1rV7NsZ85c0YRERG+igUAAL+xSive41XxNTU1mjdvnlq3bq2oqCh9+umnkqQZM2bo+eef93mAAAD4Qu3iOW+2YOBxYp8/f74yMzP11FNPKTw83DneuXNnrV692qfBAQAAz3ic2NesWaNVq1ZpxIgRatSokXM8OTlZhw4d8mlwAAD4Sm0r3pstGHg8x/7FF18oMTGxzrjD4dDZs2d9EhQAAL5mlcVzHlfsnTp10o4dO+qMb9iwQd27d/dJUAAA4PJ4XLHPnDlT6enp+uKLL+RwOPT666+roKBAa9as0ebNm/0RIwAAXjPk3SvVg6Nev4yKfciQIXrzzTf1t7/9TU2bNtXMmTN18OBBvfnmm+rfv78/YgQAwGtWWRV/Wfex33rrrcrOzvZ1LAAAwEuX/YCavLw8HTx4UNL5efeePXv6LCgAAHzNKq9t9TixHzt2TMOHD9f777+vZs2aSZJOnjyp1NRUrVu3Tm3atPF1jAAAeM0qb3fzeI597NixOnv2rA4ePKgTJ07oxIkTOnjwoBwOh8aOHeuPGAEAgJs8rti3bdumXbt2KSkpyTmWlJSkP/7xj7r11lt9GhwAAL4UJEW3VzxO7AkJCRd8EE1NTY3i4+N9EhQAAL5GK74eCxYs0P3336+8vDznWF5eniZNmqQ//OEPPg0OAABfqV08580WDNyq2Js3b+7ym0plZaV69eqlxo3PH37u3Dk1btxYv/nNb5SWluaXQAEAwKW5ldifeeYZP4cBAIB/WaUV71ZiT09P93ccAAD4lVUeKXvZD6iRpDNnzqi6utplzG63exUQAAC4fB4n9srKSj3yyCN69dVX9fXXX9f5vqamxieBAQDgS7y2tR4PP/yw3nvvPS1fvlw2m02rV6/WnDlzFB8frzVr1vgjRgAAvGYY3m/BwOOK/c0339SaNWvUp08fjR49WrfeeqsSExPVrl07vfLKKxoxYoQ/4gQAAG7wuGI/ceKErrnmGknn59NPnDghSfrJT36i7du3+zY6AAB8xCqvbfU4sV9zzTU6evSoJKlDhw569dVXJZ2v5GtfCgMAwA+NVVrxHif20aNH6+OPP5YkPfroo1q2bJkiIiI0ZcoUPfTQQz4PEAAAuM/jOfYpU6Y4/7tfv346dOiQ9uzZo8TERHXt2tWnwQEA4CsNvSo+IyNDr7/+ug4dOqTIyEilpqbqf//3f11eovZ9mZmZGj16tMuYzWbTmTNn3L6uV/exS1K7du3Url07b08DAIBfedtO9/TYbdu2acKECbrxxht17tw5PfbYY7r99tv1j3/8Q02bNq33OLvdroKCgv+6rmcXdiuxL1myxO0TPvDAAx4FAABAQ2joR8q+++67Lp8zMzPVsmVL7dmzR7fddttFrxMXF3dZMUpuJvZFixa5dTLDMEjsAICQVl5e7vLZZrPJZrNd8riysjJJUosWLS66X0VFhdq1ayeHw6EePXroiSee0PXXX+92fG4l9tpV8D9UhTl/4FG2CFnNb5wY6BAAvzFrqi+9k4+E6TJWjH/veElKSEhwGZ81a5Zmz5590WMdDocmT56sW265RZ07d653v6SkJL3wwgvq2rWrysrK9Ic//EGpqan65JNP1KZNG7fi9HqOHQCAYOCrVnxRUZFLMelOtT5hwgQdOHBAO3fuvOh+KSkpSklJcX5OTU1Vx44dtXLlSs2bN8+tOEnsAAB4wG63e9QlnjhxojZv3qzt27e7XXXXuuKKK9S9e3cdPnzY7WO86UoAABA0DEMK82LztNg3TVMTJ07Upk2b9N5776l9+/Yex1xTU6P9+/erVatWbh9DxQ4AsITaBO3N8Z6YMGGC1q5dqzfeeEPR0dEqKSmRJMXExCgyMlKSNHLkSLVu3VoZGRmSpLlz5+rmm29WYmKiTp48qQULFujzzz/X2LFj3b4uiR0AAD9Yvny5JKlPnz4u4y+++KJGjRolSSosLFRY2H+a5998843GjRunkpISNW/eXD179tSuXbvUqVMnt697WYl9x44dWrlypY4cOaINGzaodevWevnll9W+fXv95Cc/uZxTAgDgVw19H7tpmpfcJycnx+XzokWL3L7FvD4ez7Fv3LhRAwYMUGRkpPbt26eqqipJ5+/Pe+KJJ7wKBgAAf/Fmft3bNn5D8jixP/7441qxYoWee+45XXHFFc7xW265RXv37vVpcAAAwDMet+ILCgou+Ci8mJgYnTx50hcxAQDgcw39rPhA8bhij4uLu+D9dDt37tQ111zjk6AAAPC12re7ebMFA48T+7hx4zRp0iR9+OGHMgxDX375pV555RVNmzZNv/3tb/0RIwAAXgvzwRYMPG7FP/roo3I4HPrZz36m06dP67bbbpPNZtO0adN0//33+yNGAADgJo8Tu2EY+t3vfqeHHnpIhw8fVkVFhTp16qSoqCh/xAcAgE9YZY79sh9QEx4e7tEN8wAABFKYvJsnD1NwZHaPE3vfvn0vepP+e++951VAAADg8nmc2Lt16+by+ezZs8rPz9eBAweUnp7uq7gAAPApWvH1qO9Rd7Nnz1ZFRYXXAQEA4A8N/RKYQPHZ6v17771XL7zwgq9OBwAALoPP3u6Wm5uriIgIX50OAACfOv8+dm9eAuPDYPzI48Q+dOhQl8+maaq4uFh5eXmaMWOGzwIDAMCXmGOvR0xMjMvnsLAwJSUlae7cubr99tt9FhgAAPCcR4m9pqZGo0ePVpcuXdS8eXN/xQQAgM+xeO4CGjVqpNtvv523uAEAgo7hgz/BwONV8Z07d9ann37qj1gAAPCb2ordmy0YeJzYH3/8cU2bNk2bN29WcXGxysvLXTYAABA4bs+xz507Vw8++KDuvPNOSdJdd93l8mhZ0zRlGIZqamp8HyUAAF6yyhy724l9zpw5Gj9+vP7+97/7Mx4AAPzCMIyLvuvEneODgduJ3TRNSVLv3r39FgwAAPCOR7e7BctvKwAAfB+t+Au47rrrLpncT5w44VVAAAD4A0+eu4A5c+bUefIcAAD44fAosf/yl79Uy5Yt/RULAAB+E2YYXr0ExptjG5LbiZ35dQBAMLPKHLvbD6ipXRUPAAB+uNyu2B0Ohz/jAADAv7xcPBckj4r3/LWtAAAEozAZCvMiO3tzbEMisQMALMEqt7t5/BIYAADww0XFDgCwBKusiiexAwAswSr3sdOKBwAghFCxAwAswSqL50jsAABLCJOXrfggud2NVjwAACGEih0AYAm04gEACCFh8q5NHSwt7mCJEwAAuIGKHQBgCYZhePUK8mB5fTmJHQBgCYa8e0FbcKR1EjsAwCJ48hwAAAg6VOwAAMsIjprbOyR2AIAlWOU+dlrxAACEECp2AIAlcLsbAAAhhCfPAQCAy5aRkaEbb7xR0dHRatmypdLS0lRQUHDJ41577TV16NBBERER6tKli95++22PrktiBwBYQm0r3pvNE9u2bdOECRP0wQcfKDs7W2fPntXtt9+uysrKeo/ZtWuXhg8frjFjxmjfvn1KS0tTWlqaDhw44P7PaZqm6VGkPyDl5eWKiYlR6ddlstvtgQ4H8IvmN04MdAiA35g11ara/5zKyvz373htrsjccUhNoqIv+zynK05p1K0dLjvWr776Si1bttS2bdt02223XXCfe+65R5WVldq8ebNz7Oabb1a3bt20YsUKt65DxQ4AQAMoKyuTJLVo0aLefXJzc9WvXz+XsQEDBig3N9ft67B4DgBgCb5aFV9eXu4ybrPZZLPZLnqsw+HQ5MmTdcstt6hz58717ldSUqLY2FiXsdjYWJWUlLgdJxU7AMASwnywSVJCQoJiYmKcW0ZGxiWvPWHCBB04cEDr1q3z7Q91AVTsAABL8FXFXlRU5DLHfqlqfeLEidq8ebO2b9+uNm3aXHTfuLg4lZaWuoyVlpYqLi7O7Tip2AEA8IDdbnfZ6kvspmlq4sSJ2rRpk9577z21b9/+kudOSUnR1q1bXcays7OVkpLidnxU7AAAS2jo97FPmDBBa9eu1RtvvKHo6GjnPHlMTIwiIyMlSSNHjlTr1q2d7fxJkyapd+/eWrhwoQYNGqR169YpLy9Pq1atcvu6VOwAAEuofQmMN5snli9frrKyMvXp00etWrVybuvXr3fuU1hYqOLiYufn1NRUrV27VqtWrVJycrI2bNigrKysiy64+z4qdgAA/MCdx8Tk5OTUGRs2bJiGDRt22dclsQMALCFMhsK8aMZ7c2xDIrEDACyB97EDAICgQ8UOALAE47s/3hwfDEjsAABLoBUPAACCDhU7AMASDC9XxdOKBwDgB8QqrXgSOwDAEqyS2JljBwAghFCxAwAsgdvdAAAIIWHG+c2b44MBrXgAAEIIFTsAwBJoxQMAEEJYFQ8AAIIOFTsAwBIMeddOD5KCncQOALAGVsUDAICgQ2KHW97fe1i/nLJCHQc+puY3TtRbOR8HOiTAbyan99c3u5fqiam/CHQo8CHDB3+CQUAT+/bt2zV48GDFx8fLMAxlZWUFMhxcxOlvq9T5utZa8PA9gQ4F8Kvundpq1M9v0YF/Hgt0KPCx2lXx3mzBIKCJvbKyUsnJyVq2bFkgw4Ab+t9yvX7/28H6P32TAx0K4DdNI8O1au4oTXrizzp56ttAhwMfM3ywBYOALp4bOHCgBg4cGMgQAMBpwcP36K/vH9C2jwo07Td3BDoc4LIE1ar4qqoqVVVVOT+Xl5cHMBoAoWRo/55K7pCgn6Y/FehQ4CdhMhTmRT89LEhq9qBaPJeRkaGYmBjnlpCQEOiQAISA1rHNlPHgL/R/Z2SqqvpcoMOBn9CK/wGaPn26pk6d6vxcXl5OcgfgteQObdXySrtyXn7EOda4cSOldv+xxg27TbG3TJbDYQYwQsB9QZXYbTabbDZboMMAEGK27y5Q6i/nu4wtnXmv/vVZqRavySaphwpvy+4gKdmDKrEjcCpOV+lo0VfOz59/+bX2FxxTs5gmSohrEcDIAO9VnK7SwSPFLmOnv63WibLKOuMIXrzdrQFUVFTo8OHDzs9Hjx5Vfn6+WrRoobZt2wYwMnxf/sHPNXj8Eufn3y16XZI0fFAvPTv714EKCwDwPQFN7Hl5eerbt6/zc+38eXp6ujIzMwMUFS7kJz2v0ze7lwY6DKDBDB6/ONAhwNe8fchMcBTsgU3sffr0kWkydwUA8D+LTLEH1+1uAADg4lg8BwCwBouU7CR2AIAlsCoeAIAQ4u0b2ni7GwAAaHBU7AAAS7DIFDuJHQBgERbJ7LTiAQAIIVTsAABLYFU8AAAhhFXxAAAg6FCxAwAswSJr50jsAACLsEhmpxUPAEAIoWIHAFgCq+IBAAghVlkVT2IHAFiCRabYmWMHACCUULEDAKzBIiU7iR0AYAlWWTxHKx4AAD/Yvn27Bg8erPj4eBmGoaysrIvun5OTI8Mw6mwlJSUeXZfEDgCwhNpV8d5snqisrFRycrKWLVvm0XEFBQUqLi52bi1btvToeFrxAABLaOgp9oEDB2rgwIEeX6dly5Zq1qyZx8fVomIHAMAD5eXlLltVVZVPz9+tWze1atVK/fv31/vvv+/x8SR2AIA1GD7YJCUkJCgmJsa5ZWRk+CS8Vq1aacWKFdq4caM2btyohIQE9enTR3v37vXoPLTiAQCW4KtV8UVFRbLb7c5xm83mdWySlJSUpKSkJOfn1NRUHTlyRIsWLdLLL7/s9nlI7AAAeMBut7skdn+66aabtHPnTo+OIbEDACwhGJ8Vn5+fr1atWnl0DIkdAGAJDb0qvqKiQocPH3Z+Pnr0qPLz89WiRQu1bdtW06dP1xdffKE1a9ZIkp555hm1b99e119/vc6cOaPVq1frvffe01//+lePrktiBwBYQwNn9ry8PPXt29f5eerUqZKk9PR0ZWZmqri4WIWFhc7vq6ur9eCDD+qLL75QkyZN1LVrV/3tb39zOYdbYZqmaXoW6g9HeXm5YmJiVPp1WYPNdwANrfmNEwMdAuA3Zk21qvY/p7Iy//07Xpsr9vyrWFHRl3+NilPl6nltK7/G6gtU7AAAS7DKs+JJ7AAAa/By8VyQ5HUeUAMAQCihYgcAWIJFXsdOYgcAWIRFMjuteAAAQggVOwDAElgVDwBACAnGR8peDlrxAACEECp2AIAlWGTtHIkdAGARFsnsJHYAgCVYZfEcc+wAAIQQKnYAgCUY8nJVvM8i8S8SOwDAEiwyxU4rHgCAUELFDgCwBKs8oIbEDgCwCGs042nFAwAQQqjYAQCWQCseAIAQYo1GPK14AABCChU7AMASaMUDABBCrPKseBI7AMAaLDLJzhw7AAAhhIodAGAJFinYSewAAGuwyuI5WvEAAIQQKnYAgCWwKh4AgFBikUl2WvEAAIQQKnYAgCVYpGAnsQMArIFV8QAAIOhQsQMALMK7VfHB0ownsQMALIFWPAAACDokdgAAQgiteACAJVilFU9iBwBYglUeKUsrHgCAEELFDgCwBFrxAACEEKs8UpZWPAAAIYSKHQBgDRYp2UnsAABLYFU8AAAIOlTsAABLYFU8AAAhxCJT7CR2AIBFWCSzM8cOAIAfbN++XYMHD1Z8fLwMw1BWVtYlj8nJyVGPHj1ks9mUmJiozMxMj69LYgcAWILhgz+eqKysVHJyspYtW+bW/kePHtWgQYPUt29f5efna/LkyRo7dqy2bNni0XVpxQMALKGhF88NHDhQAwcOdHv/FStWqH379lq4cKEkqWPHjtq5c6cWLVqkAQMGuH2eoE7spmlKkk6Vlwc4EsB/zJrqQIcA+E3t3+/af8/9qdzLXFF7/PfPY7PZZLPZvDq3JOXm5qpfv34uYwMGDNDkyZM9Ok9QJ/ZTp05JkhLbJwQ4EgCAN06dOqWYmBi/nDs8PFxxcXG61ge5IioqSgkJrueZNWuWZs+e7fW5S0pKFBsb6zIWGxur8vJyffvtt4qMjHTrPEGd2OPj41VUVKTo6GgZwXKDYZArLy9XQkKCioqKZLfbAx0O4FP8/W54pmnq1KlTio+P99s1IiIidPToUVVXe9/9Mk2zTr7xRbXuS0Gd2MPCwtSmTZtAh2FJdrudf/gQsvj73bD8Van/t4iICEVERPj9Ot6Ii4tTaWmpy1hpaansdrvb1brEqngAAH4QUlJStHXrVpex7OxspaSkeHQeEjsAAH5QUVGh/Px85efnSzp/O1t+fr4KCwslSdOnT9fIkSOd+48fP16ffvqpHn74YR06dEjPPvusXn31VU2ZMsWj65LY4RGbzaZZs2b94OaUAF/g7zd8KS8vT927d1f37t0lSVOnTlX37t01c+ZMSVJxcbEzyUtS+/bt9dZbbyk7O1vJyclauHChVq9e7dGtbpJkmA1xjwEAAGgQVOwAAIQQEjsAACGExA4AQAghsQMAEEJI7HDbsmXLdPXVVysiIkK9evXSRx99FOiQAJ+4nNdrAj9UJHa4Zf369Zo6dapmzZqlvXv3Kjk5WQMGDNDx48cDHRrgNU9frwn8kHG7G9zSq1cv3XjjjVq6dKkkyeFwKCEhQffff78effTRAEcH+I5hGNq0aZPS0tICHQpwWajYcUnV1dXas2ePy+sEw8LC1K9fP+Xm5gYwMgDA95HYcUn//ve/VVNTc8HXCZaUlAQoKgDAhZDYAQAIISR2XNJVV12lRo0aXfB1gnFxcQGKCgBwISR2XFJ4eLh69uzp8jpBh8OhrVu3evw6QQCAfzUOdAAIDlOnTlV6erpuuOEG3XTTTXrmmWdUWVmp0aNHBzo0wGsVFRU6fPiw83Pt6zVbtGihtm3bBjAywHPc7ga3LV26VAsWLFBJSYm6deumJUuWqFevXoEOC/BaTk6O+vbtW2c8PT1dmZmZDR8Q4AUSOwAAIYQ5dgAAQgiJHQCAEEJiBwAghJDYAQAIISR2AABCCIkdAIAQQmIHACCEkNgBL40aNcrl3d19+vTR5MmTGzyOnJwcGYahkydP1ruPYRjKyspy+5yzZ89Wt27dvIrrs88+k2EYys/P9+o8ANxDYkdIGjVqlAzDkGEYCg8PV2JioubOnatz5875/dqvv/665s2b59a+7iRjAPAEz4pHyLrjjjv04osvqqqqSm+//bYmTJigK664QtOnT6+zb3V1tcLDw31y3RYtWvjkPABwOajYEbJsNpvi4uLUrl07/fa3v1W/fv30l7/8RdJ/2ufz589XfHy8kpKSJElFRUW6++671axZM7Vo0UJDhgzRZ5995jxnTU2Npk6dqmbNmunKK6/Uww8/rO8/lfn7rfiqqio98sgjSkhIkM1mU2Jiop5//nl99tlnzueTN2/eXIZhaNSoUZLOvz0vIyND7du3V2RkpJKTk7VhwwaX67z99tu67rrrFBkZqb59+7rE6a5HHnlE1113nZo0aaJrrrlGM2bM0NmzZ+vst3LlSiUkJKhJkya6++67VVZW5vL96tWr1bFjR0VERKhDhw569tlnPY4FgG+Q2GEZkZGRqq6udn7eunWrCgoKlJ2drc2bN+vs2bMaMGCAoqOjtWPHDr3//vuKiorSHXfc4Txu4cKFyszM1AsvvKCdO3fqxIkT2rRp00WvO3LkSP35z3/WkiVLdPDgQa1cuVJRUVFKSEjQxo0bJUkFBQUqLi7W4sWLJUkZGRlas2aNVqxYoU8++URTpkzRvffeq23btkk6/wvI0KFDNXjwYOXn52vs2LF69NFHPf5/Eh0drczMTP3jH//Q4sWL9dxzz2nRokUu+xw+fFivvvqq3nzzTb377rvat2+f7rvvPuf3r7zyimbOnKn58+fr4MGDeuKJJzRjxgy99NJLHscDwAdMIASlp6ebQ4YMMU3TNB0Oh5mdnW3abDZz2rRpzu9jY2PNqqoq5zEvv/yymZSUZDocDudYVVWVGRkZaW7ZssU0TdNs1aqV+dRTTzm/P3v2rNmmTRvntUzTNHv37m1OmjTJNE3TLCgoMCWZ2dnZF4zz73//uynJ/Oabb5xjZ86cMZs0aWLu2rXLZd8xY8aYw4cPN03TNKdPn2526tTJ5ftHHnmkzrm+T5K5adOmer9fsGCB2bNnT+fnWbNmmY0aNTKPHTvmHHvnnXfMsLAws7i42DRN0/zxj39srl271uU88+bNM1NSUkzTNM2jR4+aksx9+/bVe10AvsMcO0LW5s2bFRUVpbNnz8rhcOhXv/qVZs+e7fy+S5cuLvPqH3/8sQ4fPqzo6GiX85w5c0ZHjhxRWVmZiouLXV5V27hxY91www112vG18vPz1ahRI/Xu3dvtuA8fPqzTp0+rf//+LuPV1dXq3r27JOngwYN1XpmbkpLi9jVqrV+/XkuWLNGRI0dUUVGhc+fOyW63u+zTtm1btW7d2uU6DodDBQUFio6O1pEjRzRmzBiNGzfOuc+5c+cUExPjcTwAvEdiR8jq27evli9frvDwcMXHx6txY9e/7k2bNnX5XFFRoZ49e+qVV16pc64f/ehHlxVDZGSkx8dUVFRIkt566y2XhCqdXzfgK7m5uRoxYoTmzJmjAQMGKCYmRuvWrdPChQs9jvW5556r84tGo0aNfBYrAPeR2BGymjZtqsTERLf379Gjh9avX6+WLVvWqVprtWrVSh9++KFuu+02Secr0z179qhHjx4X3L9Lly5yOBzatm2b+vXrV+f72o5BTU2Nc6xTp06y2WwqLCyst9Lv2LGjcyFgrQ8++ODSP+R/2bVrl9q1a6ff/e53zrHPP/+8zn6FhYX68ssvFR8f77xOWFiYkpKSFBsbq/j4eH366acaMWKER9cH4B8sngO+M2LECF111VUaMmSIduzYoaNHjyonJ0cPPPCAjh07JkmaNGmSnnzySWVlZenQoUO67777LnoP+tVXX6309HT95je/UVZWlvOcr776qiSpXbt2MgxDmzdv1ldffaWKigpFR0dr2rRpmjJlil566SUdOXJEe/fu1R//+EfngrTx48frX//6lx566CEVFBRo7dq1yszM9Ojnvfbaa1VYWKh169bpyJEjWrJkyQUXAkZERCg9PV0ff/yxduzYoQceeEB333234uLiJElz5sxRRkaGlixZon/+85/av3+/XnzxRT399NMexQPAN0jswHeaNGmi7du3q23btho6dKg6duyoMWPG6MyZM84K/sEHH9Svf/1rpaenKyUlRdHR0fr5z39+0fMuX75c//M//6P77rtPHTp00Lhx41RZWSlJat26tebMmaNHH31UsbGxmjhxoiRp3rx5mjFjhjIyMtSxY0fdcccdeuutt9S+fXtJ5+e9N27cqKysLCUnJ2vFihV64oknPPp577rrLk2ZMkUTJ05Ut27dtGvXLs2YMaPOfomJiRo6dKjuvPNO3X777eratavL7Wxjx47V6tWr9eKLL6pLly7q3bu3MjMznbECaFiGWd+qHwAAEHSo2AEACCEkdgAAQgiJHQCAEEJiBwAghJDYAQAIISR2AABCCIkdAIAQQmIHACCEkNgBAAghJHYAAEIIiR0AgBBCYgcAIIT8fzD7g83Q8Ik8AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "SVM_model = SVC(C = 1, kernel = 'sigmoid')\n",
    "SVM_model.fit(X_train, Y_train)\n",
    "SVM_prediction = SVM_model.predict(X_test)\n",
    "\n",
    "SVM_train_accuracy = SVM_model.score(X_train, Y_train)\n",
    "SVM_test_accuracy = SVM_model.score(X_test, Y_test)\n",
    "print ('Точность модели на обучающей выборке: ', round (SVM_train_accuracy, 2))\n",
    "print ('Точность модели на тестовой выборке: ', round (SVM_test_accuracy, 2))\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix(SVM_prediction, Y_test).T)\n",
    "disp.plot(cmap = 'Blues', ax=ax);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "cad13448",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 1.0, 'kernel': 'sigmoid'}"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "par = {'kernel': ['linear', 'poly', 'rbf', 'sigmoid'],\n",
    "'C': [0.5, 1.0, 2.0, 5.0, 10.0, 20.0, 50.0, 100.0]}\n",
    "search = GridSearchCV(SVM_model,par,cv=10)\n",
    "search.fit(X_train, Y_train)\n",
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "465e22a1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Точность модели на обучающей выборке:  0.72\n",
      "Точность модели на тестовой выборке:  0.5\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "NB_model = GaussianNB() \n",
    "NB_model.fit(X_train, Y_train)\n",
    "NB_prediction = NB_model.predict(X_test)\n",
    "\n",
    "NB_train_accuracy = NB_model.score(X_train, Y_train)\n",
    "NB_test_accuracy = NB_model.score(X_test, Y_test)\n",
    "print ('Точность модели на обучающей выборке: ', round (NB_train_accuracy, 2))\n",
    "print ('Точность модели на тестовой выборке: ', round (NB_test_accuracy, 2))\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix(NB_prediction, Y_test).T)\n",
    "disp.plot(cmap = 'Blues', ax=ax);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "86be0c1a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Точность модели на обучающей выборке:  0.68\n",
      "Точность модели на тестовой выборке:  0.75\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "KNN_model = KNeighborsClassifier(n_neighbors = 4)\n",
    "KNN_model.fit(X_train, Y_train)\n",
    "KNN_prediction = KNN_model.predict(X_test)\n",
    "\n",
    "KNN_train_accuracy = KNN_model.score(X_train, Y_train)\n",
    "KNN_test_accuracy = KNN_model.score(X_test, Y_test)\n",
    "print ('Точность модели на обучающей выборке: ', round (KNN_train_accuracy, 2))\n",
    "print ('Точность модели на тестовой выборке: ', round (KNN_test_accuracy, 2))\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix(KNN_prediction, Y_test).T)\n",
    "disp.plot(cmap = 'Blues', ax=ax);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6ce6bf2d",
   "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": "3.10.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}