Robot.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f822d172",
   "metadata": {},
   "source": [
    "# Wall-Following navigation task with mobile robot SCITOS-G5\n",
    "For Each Attribute: \n",
    "\n",
    "   -- File sensor_readings_24.data:\n",
    "   \n",
    "\t 1. US1: ultrasound sensor at the front of the robot (reference angle: 180°) - (numeric: real)\n",
    "\t 2. US2: ultrasound reading (reference angle: -165°) - (numeric: real)\n",
    "\t 3. US3: ultrasound reading (reference angle: -150°) - (numeric: real)\n",
    "\t 4. US4: ultrasound reading (reference angle: -135°) - (numeric: real)\n",
    "\t 5. US5: ultrasound reading (reference angle: -120°) - (numeric: real)\n",
    "\t 6. US6: ultrasound reading (reference angle: -105°) - (numeric: real)\n",
    "\t 7. US7: ultrasound reading (reference angle: -90°) - (numeric: real)\n",
    "\t 8. US8: ultrasound reading (reference angle: -75°) - (numeric: real)\n",
    "\t 9. US9: ultrasound reading (reference angle: -60°) - (numeric: real)\n",
    "\t10. US10: ultrasound reading (reference angle: -45°) - (numeric: real)\n",
    "\t11. US11: ultrasound reading (reference angle: -30°) - (numeric: real)\n",
    "\t12. US12: ultrasound reading (reference angle: -15°) - (numeric: real)\n",
    "\t13. US13: reading of ultrasound sensor situated at the back of the robot (reference angle: 0°) - (numeric: real)\n",
    "\t14. US14: ultrasound reading (reference angle: 15°) - (numeric: real)\n",
    "\t15. US15: ultrasound reading (reference angle: 30°) - (numeric: real)\n",
    "\t16. US16: ultrasound reading (reference angle: 45°) - (numeric: real)\n",
    "\t17. US17: ultrasound reading (reference angle: 60°) - (numeric: real)\n",
    "\t18. US18: ultrasound reading (reference angle: 75°) - (numeric: real)\n",
    "\t19. US19: ultrasound reading (reference angle: 90°) - (numeric: real)\n",
    "\t20. US20: ultrasound reading (reference angle: 105°) - (numeric: real)\n",
    "\t21. US21: ultrasound reading (reference angle: 120°) - (numeric: real)\n",
    "\t22. US22: ultrasound reading (reference angle: 135°) - (numeric: real)\n",
    "\t23. US23: ultrasound reading (reference angle: 150°) - (numeric: real)\n",
    "\t24. US24: ultrasound reading (reference angle: 165°) - (numeric: real)\n",
    "   \t25. Class: \n",
    "      \t\t-- Move-Forward\n",
    "      \t\t-- Slight-Right-Turn\n",
    "      \t\t-- Sharp-Right-Turn\n",
    "      \t\t-- Slight-Left-Turn\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dad9e870",
   "metadata": {},
   "source": [
    "# importing library for reading, writing and perform basic operations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4fb90461",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "#Importing library for visualization\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "#Filter the unwanted warning\n",
    "import warnings\n",
    "warnings.simplefilter(\"ignore\")\n",
    "#Importing library for splitting model into train and test and for data transformation\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import r2_score,mean_squared_error,accuracy_score,recall_score,precision_score,confusion_matrix,roc_auc_score,classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "82fab2da",
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv(r\"C:\\Users\\TRUPTI\\Desktop\\MLProject\\wall+following+robot+navigation+data\\sensor_readings_24.csv\",header=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "04446cc4",
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.438</td>\n",
       "      <td>0.498</td>\n",
       "      <td>3.625</td>\n",
       "      <td>3.645</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.351</td>\n",
       "      <td>2.332</td>\n",
       "      <td>2.643</td>\n",
       "      <td>...</td>\n",
       "      <td>0.593</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.445</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.440</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.438</td>\n",
       "      <td>0.498</td>\n",
       "      <td>3.625</td>\n",
       "      <td>3.648</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.637</td>\n",
       "      <td>2.332</td>\n",
       "      <td>2.649</td>\n",
       "      <td>...</td>\n",
       "      <td>0.592</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.449</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.443</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.438</td>\n",
       "      <td>0.498</td>\n",
       "      <td>3.625</td>\n",
       "      <td>3.629</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.637</td>\n",
       "      <td>2.334</td>\n",
       "      <td>2.643</td>\n",
       "      <td>...</td>\n",
       "      <td>0.593</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.449</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.446</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.437</td>\n",
       "      <td>0.501</td>\n",
       "      <td>3.625</td>\n",
       "      <td>3.626</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.353</td>\n",
       "      <td>2.334</td>\n",
       "      <td>2.642</td>\n",
       "      <td>...</td>\n",
       "      <td>0.593</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.449</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.438</td>\n",
       "      <td>0.498</td>\n",
       "      <td>3.626</td>\n",
       "      <td>3.629</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.640</td>\n",
       "      <td>2.334</td>\n",
       "      <td>2.639</td>\n",
       "      <td>...</td>\n",
       "      <td>0.592</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.449</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.441</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      0      1      2      3    4      5    6      7      8      9   ...  \\\n",
       "0  0.438  0.498  3.625  3.645  5.0  2.918  5.0  2.351  2.332  2.643  ...   \n",
       "1  0.438  0.498  3.625  3.648  5.0  2.918  5.0  2.637  2.332  2.649  ...   \n",
       "2  0.438  0.498  3.625  3.629  5.0  2.918  5.0  2.637  2.334  2.643  ...   \n",
       "3  0.437  0.501  3.625  3.626  5.0  2.918  5.0  2.353  2.334  2.642  ...   \n",
       "4  0.438  0.498  3.626  3.629  5.0  2.918  5.0  2.640  2.334  2.639  ...   \n",
       "\n",
       "      15     16     17     18     19     20     21     22     23  \\\n",
       "0  0.593  0.502  0.493  0.504  0.445  0.431  0.444  0.440  0.429   \n",
       "1  0.592  0.502  0.493  0.504  0.449  0.431  0.444  0.443  0.429   \n",
       "2  0.593  0.502  0.493  0.504  0.449  0.431  0.444  0.446  0.429   \n",
       "3  0.593  0.502  0.493  0.504  0.449  0.431  0.444  0.444  0.429   \n",
       "4  0.592  0.502  0.493  0.504  0.449  0.431  0.444  0.441  0.429   \n",
       "\n",
       "                  24  \n",
       "0  Slight-Right-Turn  \n",
       "1  Slight-Right-Turn  \n",
       "2  Slight-Right-Turn  \n",
       "3  Slight-Right-Turn  \n",
       "4  Slight-Right-Turn  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9d7dd67",
   "metadata": {},
   "source": [
    "# Basic Description of dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5748322f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5456, 25)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Dimension of dataset.\n",
    "\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3c9d72cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 5456 entries, 0 to 5455\n",
      "Data columns (total 25 columns):\n",
      " #   Column  Non-Null Count  Dtype  \n",
      "---  ------  --------------  -----  \n",
      " 0   0       5456 non-null   float64\n",
      " 1   1       5456 non-null   float64\n",
      " 2   2       5456 non-null   float64\n",
      " 3   3       5456 non-null   float64\n",
      " 4   4       5456 non-null   float64\n",
      " 5   5       5456 non-null   float64\n",
      " 6   6       5456 non-null   float64\n",
      " 7   7       5456 non-null   float64\n",
      " 8   8       5456 non-null   float64\n",
      " 9   9       5456 non-null   float64\n",
      " 10  10      5456 non-null   float64\n",
      " 11  11      5456 non-null   float64\n",
      " 12  12      5456 non-null   float64\n",
      " 13  13      5456 non-null   float64\n",
      " 14  14      5456 non-null   float64\n",
      " 15  15      5456 non-null   float64\n",
      " 16  16      5456 non-null   float64\n",
      " 17  17      5456 non-null   float64\n",
      " 18  18      5456 non-null   float64\n",
      " 19  19      5456 non-null   float64\n",
      " 20  20      5456 non-null   float64\n",
      " 21  21      5456 non-null   float64\n",
      " 22  22      5456 non-null   float64\n",
      " 23  23      5456 non-null   float64\n",
      " 24  24      5456 non-null   object \n",
      "dtypes: float64(24), object(1)\n",
      "memory usage: 1.0+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9d8914fe",
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.00000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "      <td>5456.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.471617</td>\n",
       "      <td>2.327043</td>\n",
       "      <td>2.489347</td>\n",
       "      <td>2.796501</td>\n",
       "      <td>2.958552</td>\n",
       "      <td>2.893073</td>\n",
       "      <td>3.351113</td>\n",
       "      <td>2.540397</td>\n",
       "      <td>3.125621</td>\n",
       "      <td>2.832386</td>\n",
       "      <td>...</td>\n",
       "      <td>2.205772</td>\n",
       "      <td>1.202111</td>\n",
       "      <td>0.989831</td>\n",
       "      <td>0.910273</td>\n",
       "      <td>1.05811</td>\n",
       "      <td>1.076320</td>\n",
       "      <td>1.015923</td>\n",
       "      <td>1.778034</td>\n",
       "      <td>1.555045</td>\n",
       "      <td>1.578508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.802801</td>\n",
       "      <td>1.410146</td>\n",
       "      <td>1.247435</td>\n",
       "      <td>1.309368</td>\n",
       "      <td>1.339225</td>\n",
       "      <td>1.282575</td>\n",
       "      <td>1.413692</td>\n",
       "      <td>1.111554</td>\n",
       "      <td>1.356965</td>\n",
       "      <td>1.307843</td>\n",
       "      <td>...</td>\n",
       "      <td>1.715435</td>\n",
       "      <td>1.098568</td>\n",
       "      <td>0.942075</td>\n",
       "      <td>0.889527</td>\n",
       "      <td>1.14463</td>\n",
       "      <td>1.141498</td>\n",
       "      <td>0.887439</td>\n",
       "      <td>1.571686</td>\n",
       "      <td>1.291447</td>\n",
       "      <td>1.150480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.437000</td>\n",
       "      <td>0.470000</td>\n",
       "      <td>0.833000</td>\n",
       "      <td>1.120000</td>\n",
       "      <td>1.114000</td>\n",
       "      <td>1.122000</td>\n",
       "      <td>0.859000</td>\n",
       "      <td>0.836000</td>\n",
       "      <td>0.810000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.495000</td>\n",
       "      <td>0.424000</td>\n",
       "      <td>0.373000</td>\n",
       "      <td>0.354000</td>\n",
       "      <td>0.34000</td>\n",
       "      <td>0.355000</td>\n",
       "      <td>0.380000</td>\n",
       "      <td>0.370000</td>\n",
       "      <td>0.367000</td>\n",
       "      <td>0.377000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.921000</td>\n",
       "      <td>1.362000</td>\n",
       "      <td>1.538750</td>\n",
       "      <td>1.731000</td>\n",
       "      <td>1.774000</td>\n",
       "      <td>1.785750</td>\n",
       "      <td>1.930750</td>\n",
       "      <td>1.618000</td>\n",
       "      <td>1.799750</td>\n",
       "      <td>1.636000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.860000</td>\n",
       "      <td>0.690000</td>\n",
       "      <td>0.581000</td>\n",
       "      <td>0.529750</td>\n",
       "      <td>0.52300</td>\n",
       "      <td>0.541750</td>\n",
       "      <td>0.567000</td>\n",
       "      <td>0.743000</td>\n",
       "      <td>0.792000</td>\n",
       "      <td>0.884000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.335000</td>\n",
       "      <td>1.904500</td>\n",
       "      <td>2.064000</td>\n",
       "      <td>2.458000</td>\n",
       "      <td>2.667000</td>\n",
       "      <td>2.682500</td>\n",
       "      <td>3.225500</td>\n",
       "      <td>2.172000</td>\n",
       "      <td>2.802000</td>\n",
       "      <td>2.679000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.328500</td>\n",
       "      <td>0.803000</td>\n",
       "      <td>0.738000</td>\n",
       "      <td>0.685000</td>\n",
       "      <td>0.69100</td>\n",
       "      <td>0.693000</td>\n",
       "      <td>0.764000</td>\n",
       "      <td>1.030500</td>\n",
       "      <td>1.071000</td>\n",
       "      <td>1.289000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.814000</td>\n",
       "      <td>2.681500</td>\n",
       "      <td>2.739250</td>\n",
       "      <td>4.093500</td>\n",
       "      <td>4.314500</td>\n",
       "      <td>3.835250</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>3.193000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>3.526250</td>\n",
       "      <td>...</td>\n",
       "      <td>4.436250</td>\n",
       "      <td>1.159000</td>\n",
       "      <td>0.913000</td>\n",
       "      <td>0.837000</td>\n",
       "      <td>0.85700</td>\n",
       "      <td>0.863000</td>\n",
       "      <td>1.002250</td>\n",
       "      <td>2.068250</td>\n",
       "      <td>1.559500</td>\n",
       "      <td>1.657250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.025000</td>\n",
       "      <td>5.029000</td>\n",
       "      <td>5.017000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.005000</td>\n",
       "      <td>5.008000</td>\n",
       "      <td>5.087000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.022000</td>\n",
       "      <td>...</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.00000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                0            1            2            3            4   \\\n",
       "count  5456.000000  5456.000000  5456.000000  5456.000000  5456.000000   \n",
       "mean      1.471617     2.327043     2.489347     2.796501     2.958552   \n",
       "std       0.802801     1.410146     1.247435     1.309368     1.339225   \n",
       "min       0.400000     0.437000     0.470000     0.833000     1.120000   \n",
       "25%       0.921000     1.362000     1.538750     1.731000     1.774000   \n",
       "50%       1.335000     1.904500     2.064000     2.458000     2.667000   \n",
       "75%       1.814000     2.681500     2.739250     4.093500     4.314500   \n",
       "max       5.000000     5.025000     5.029000     5.017000     5.000000   \n",
       "\n",
       "                5            6            7            8            9   ...  \\\n",
       "count  5456.000000  5456.000000  5456.000000  5456.000000  5456.000000  ...   \n",
       "mean      2.893073     3.351113     2.540397     3.125621     2.832386  ...   \n",
       "std       1.282575     1.413692     1.111554     1.356965     1.307843  ...   \n",
       "min       1.114000     1.122000     0.859000     0.836000     0.810000  ...   \n",
       "25%       1.785750     1.930750     1.618000     1.799750     1.636000  ...   \n",
       "50%       2.682500     3.225500     2.172000     2.802000     2.679000  ...   \n",
       "75%       3.835250     5.000000     3.193000     5.000000     3.526250  ...   \n",
       "max       5.005000     5.008000     5.087000     5.000000     5.022000  ...   \n",
       "\n",
       "                14           15           16           17          18  \\\n",
       "count  5456.000000  5456.000000  5456.000000  5456.000000  5456.00000   \n",
       "mean      2.205772     1.202111     0.989831     0.910273     1.05811   \n",
       "std       1.715435     1.098568     0.942075     0.889527     1.14463   \n",
       "min       0.495000     0.424000     0.373000     0.354000     0.34000   \n",
       "25%       0.860000     0.690000     0.581000     0.529750     0.52300   \n",
       "50%       1.328500     0.803000     0.738000     0.685000     0.69100   \n",
       "75%       4.436250     1.159000     0.913000     0.837000     0.85700   \n",
       "max       5.000000     5.000000     5.000000     5.000000     5.00000   \n",
       "\n",
       "                19           20           21           22           23  \n",
       "count  5456.000000  5456.000000  5456.000000  5456.000000  5456.000000  \n",
       "mean      1.076320     1.015923     1.778034     1.555045     1.578508  \n",
       "std       1.141498     0.887439     1.571686     1.291447     1.150480  \n",
       "min       0.355000     0.380000     0.370000     0.367000     0.377000  \n",
       "25%       0.541750     0.567000     0.743000     0.792000     0.884000  \n",
       "50%       0.693000     0.764000     1.030500     1.071000     1.289000  \n",
       "75%       0.863000     1.002250     2.068250     1.559500     1.657250  \n",
       "max       5.000000     5.000000     5.000000     5.000000     5.000000  \n",
       "\n",
       "[8 rows x 24 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0f46ca77",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     0\n",
       "1     0\n",
       "2     0\n",
       "3     0\n",
       "4     0\n",
       "5     0\n",
       "6     0\n",
       "7     0\n",
       "8     0\n",
       "9     0\n",
       "10    0\n",
       "11    0\n",
       "12    0\n",
       "13    0\n",
       "14    0\n",
       "15    0\n",
       "16    0\n",
       "17    0\n",
       "18    0\n",
       "19    0\n",
       "20    0\n",
       "21    0\n",
       "22    0\n",
       "23    0\n",
       "24    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "56fbbdb4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generating final dataset\n",
    "new_df=df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "24750e8f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Creating column names\n",
    "\n",
    "final_df_columns=[\"US1(angle:-180°)\",\"US2(angle:-165°)\",\"US3(angle:-150°)\",\"US4(angle:-135°)\",\"US5(angle:-120°)\",\"US6(angle:-105°)\",\n",
    "        \"US7(angle:-90°)\",\"US8(angle:-75°)\",\"US9(angle:-60°)\",\"US10(angle:-45°)\",\"US11(angle:-30°)\",\n",
    "                  \"US12(angle:-15°)\",\"US13(angle:0°)\",\"US14(angle:+15°)\",\"US15(angle:+30°)\",\"US16(angle:+45°)\",\n",
    "        \"US17(angle:+60°)\",\"US18(angle:+75°)\",\"US19(angle:+90°)\",\"US20(angle:+105°)\",\"US21(angle:+120°)\",\n",
    "                  \"US22(angle:+135°)\",\"US23(angle:+150°)\",\"US24(angle:+165°)\",\"CLASS\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "36a8cc89",
   "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>US1(angle:-180°)</th>\n",
       "      <th>US2(angle:-165°)</th>\n",
       "      <th>US3(angle:-150°)</th>\n",
       "      <th>US4(angle:-135°)</th>\n",
       "      <th>US5(angle:-120°)</th>\n",
       "      <th>US6(angle:-105°)</th>\n",
       "      <th>US7(angle:-90°)</th>\n",
       "      <th>US8(angle:-75°)</th>\n",
       "      <th>US9(angle:-60°)</th>\n",
       "      <th>US10(angle:-45°)</th>\n",
       "      <th>...</th>\n",
       "      <th>US16(angle:+45°)</th>\n",
       "      <th>US17(angle:+60°)</th>\n",
       "      <th>US18(angle:+75°)</th>\n",
       "      <th>US19(angle:+90°)</th>\n",
       "      <th>US20(angle:+105°)</th>\n",
       "      <th>US21(angle:+120°)</th>\n",
       "      <th>US22(angle:+135°)</th>\n",
       "      <th>US23(angle:+150°)</th>\n",
       "      <th>US24(angle:+165°)</th>\n",
       "      <th>CLASS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.438</td>\n",
       "      <td>0.498</td>\n",
       "      <td>3.625</td>\n",
       "      <td>3.645</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.351</td>\n",
       "      <td>2.332</td>\n",
       "      <td>2.643</td>\n",
       "      <td>...</td>\n",
       "      <td>0.593</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.445</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.440</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.438</td>\n",
       "      <td>0.498</td>\n",
       "      <td>3.625</td>\n",
       "      <td>3.648</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.637</td>\n",
       "      <td>2.332</td>\n",
       "      <td>2.649</td>\n",
       "      <td>...</td>\n",
       "      <td>0.592</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.449</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.443</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.438</td>\n",
       "      <td>0.498</td>\n",
       "      <td>3.625</td>\n",
       "      <td>3.629</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.637</td>\n",
       "      <td>2.334</td>\n",
       "      <td>2.643</td>\n",
       "      <td>...</td>\n",
       "      <td>0.593</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.449</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.446</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.437</td>\n",
       "      <td>0.501</td>\n",
       "      <td>3.625</td>\n",
       "      <td>3.626</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.353</td>\n",
       "      <td>2.334</td>\n",
       "      <td>2.642</td>\n",
       "      <td>...</td>\n",
       "      <td>0.593</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.449</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.438</td>\n",
       "      <td>0.498</td>\n",
       "      <td>3.626</td>\n",
       "      <td>3.629</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.918</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.640</td>\n",
       "      <td>2.334</td>\n",
       "      <td>2.639</td>\n",
       "      <td>...</td>\n",
       "      <td>0.592</td>\n",
       "      <td>0.502</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.504</td>\n",
       "      <td>0.449</td>\n",
       "      <td>0.431</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.441</td>\n",
       "      <td>0.429</td>\n",
       "      <td>Slight-Right-Turn</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   US1(angle:-180°)  US2(angle:-165°)  US3(angle:-150°)  US4(angle:-135°)  \\\n",
       "0             0.438             0.498             3.625             3.645   \n",
       "1             0.438             0.498             3.625             3.648   \n",
       "2             0.438             0.498             3.625             3.629   \n",
       "3             0.437             0.501             3.625             3.626   \n",
       "4             0.438             0.498             3.626             3.629   \n",
       "\n",
       "   US5(angle:-120°)  US6(angle:-105°)  US7(angle:-90°)  US8(angle:-75°)  \\\n",
       "0               5.0             2.918              5.0            2.351   \n",
       "1               5.0             2.918              5.0            2.637   \n",
       "2               5.0             2.918              5.0            2.637   \n",
       "3               5.0             2.918              5.0            2.353   \n",
       "4               5.0             2.918              5.0            2.640   \n",
       "\n",
       "   US9(angle:-60°)  US10(angle:-45°)  ...  US16(angle:+45°)  US17(angle:+60°)  \\\n",
       "0            2.332             2.643  ...             0.593             0.502   \n",
       "1            2.332             2.649  ...             0.592             0.502   \n",
       "2            2.334             2.643  ...             0.593             0.502   \n",
       "3            2.334             2.642  ...             0.593             0.502   \n",
       "4            2.334             2.639  ...             0.592             0.502   \n",
       "\n",
       "   US18(angle:+75°)  US19(angle:+90°)  US20(angle:+105°)  US21(angle:+120°)  \\\n",
       "0             0.493             0.504              0.445              0.431   \n",
       "1             0.493             0.504              0.449              0.431   \n",
       "2             0.493             0.504              0.449              0.431   \n",
       "3             0.493             0.504              0.449              0.431   \n",
       "4             0.493             0.504              0.449              0.431   \n",
       "\n",
       "   US22(angle:+135°)  US23(angle:+150°)  US24(angle:+165°)              CLASS  \n",
       "0              0.444              0.440              0.429  Slight-Right-Turn  \n",
       "1              0.444              0.443              0.429  Slight-Right-Turn  \n",
       "2              0.444              0.446              0.429  Slight-Right-Turn  \n",
       "3              0.444              0.444              0.429  Slight-Right-Turn  \n",
       "4              0.444              0.441              0.429  Slight-Right-Turn  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Adding Column names to dataset\n",
    "new_df.columns=final_df_columns\n",
    "new_df.to_csv(\"new data with target class.csv\")\n",
    "new_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c176590",
   "metadata": {},
   "source": [
    "* As our dataframe is completely cleaned and preprocessed. we will remove the target attribute and store our final dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a84b8700",
   "metadata": {},
   "outputs": [],
   "source": [
    "target=new_df[\"CLASS\"]\n",
    "final_df = new_df.drop(columns =\"CLASS\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "29f1d354",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5456, 24)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a053496a",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis (EDA)\n",
    "Analyzing data sets to summarize their main characteristics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "068f4467",
   "metadata": {},
   "outputs": [],
   "source": [
    "#List with class names\n",
    "\n",
    "class_names = [\"Move-Forward\",\n",
    "               \"Slight-Right-Turn\",\n",
    "               \"Sharp-Right-Turn\",\n",
    "               \"Slight-Left-Turn\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a234ca47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CLASS\n",
       "Slight-Right-Turn     826\n",
       "Sharp-Right-Turn     2097\n",
       "Move-Forward         2205\n",
       "Slight-Left-Turn      328\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Values count based on class \n",
    "new_df[\"CLASS\"].value_counts(sort=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "3cb619e5",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x ='CLASS',data =new_df) \n",
    "plt.show() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "dbadebc9",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6,6))\n",
    "#a=np.random.random(4)\n",
    "#cs=cm.Set1(np.arange(4)/7.)\n",
    "plt.pie(new_df[\"CLASS\"].value_counts().sort_index(),labels=class_names,autopct='%1.1f%%',startangle=90) \n",
    "plt.title(\"ROBOT  navigation  class\") \n",
    "plt.show() #displaying the plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c175130d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "cf020824",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(data=new_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e3498b17",
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in new_df:                  # i = column name \n",
    "    if new_df[i].dtypes in (\"float64\",\"int64\"): # df[i] = every column will select float & int column only\n",
    "        q1 = new_df[i].quantile(0.25)  # for 1 column compute q1\n",
    "        q3 = new_df[i].quantile(0.75)  # for 1 column compute q3\n",
    "        iqr = q3-q1  # for 1 column compute IQR\n",
    "        ul = q3 + 1.5*iqr # for 1 column compute UPPER LIMIT\n",
    "        ll = q1 - 1.5*iqr # for 1 column compute LOWER LIMIT \n",
    "        new_df[i].clip(upper=ul,lower=ll,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "54210fbd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(data=new_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f83d3e6e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_360ba_row0_col0, #T_360ba_row1_col1, #T_360ba_row2_col2, #T_360ba_row3_col3, #T_360ba_row4_col4, #T_360ba_row5_col5, #T_360ba_row6_col6, #T_360ba_row7_col7, #T_360ba_row8_col8, #T_360ba_row9_col9, #T_360ba_row10_col10, #T_360ba_row11_col11, #T_360ba_row12_col12, #T_360ba_row13_col13, #T_360ba_row14_col14, #T_360ba_row15_col15, #T_360ba_row16_col16, #T_360ba_row17_col17, #T_360ba_row18_col18, #T_360ba_row19_col19, #T_360ba_row20_col20, #T_360ba_row21_col21, #T_360ba_row22_col22, #T_360ba_row23_col23 {\n",
       "  background-color: yellow;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_360ba\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_360ba_level0_col0\" class=\"col_heading level0 col0\" >US1(angle:-180°)</th>\n",
       "      <th id=\"T_360ba_level0_col1\" class=\"col_heading level0 col1\" >US2(angle:-165°)</th>\n",
       "      <th id=\"T_360ba_level0_col2\" class=\"col_heading level0 col2\" >US3(angle:-150°)</th>\n",
       "      <th id=\"T_360ba_level0_col3\" class=\"col_heading level0 col3\" >US4(angle:-135°)</th>\n",
       "      <th id=\"T_360ba_level0_col4\" class=\"col_heading level0 col4\" >US5(angle:-120°)</th>\n",
       "      <th id=\"T_360ba_level0_col5\" class=\"col_heading level0 col5\" >US6(angle:-105°)</th>\n",
       "      <th id=\"T_360ba_level0_col6\" class=\"col_heading level0 col6\" >US7(angle:-90°)</th>\n",
       "      <th id=\"T_360ba_level0_col7\" class=\"col_heading level0 col7\" >US8(angle:-75°)</th>\n",
       "      <th id=\"T_360ba_level0_col8\" class=\"col_heading level0 col8\" >US9(angle:-60°)</th>\n",
       "      <th id=\"T_360ba_level0_col9\" class=\"col_heading level0 col9\" >US10(angle:-45°)</th>\n",
       "      <th id=\"T_360ba_level0_col10\" class=\"col_heading level0 col10\" >US11(angle:-30°)</th>\n",
       "      <th id=\"T_360ba_level0_col11\" class=\"col_heading level0 col11\" >US12(angle:-15°)</th>\n",
       "      <th id=\"T_360ba_level0_col12\" class=\"col_heading level0 col12\" >US13(angle:0°)</th>\n",
       "      <th id=\"T_360ba_level0_col13\" class=\"col_heading level0 col13\" >US14(angle:+15°)</th>\n",
       "      <th id=\"T_360ba_level0_col14\" class=\"col_heading level0 col14\" >US15(angle:+30°)</th>\n",
       "      <th id=\"T_360ba_level0_col15\" class=\"col_heading level0 col15\" >US16(angle:+45°)</th>\n",
       "      <th id=\"T_360ba_level0_col16\" class=\"col_heading level0 col16\" >US17(angle:+60°)</th>\n",
       "      <th id=\"T_360ba_level0_col17\" class=\"col_heading level0 col17\" >US18(angle:+75°)</th>\n",
       "      <th id=\"T_360ba_level0_col18\" class=\"col_heading level0 col18\" >US19(angle:+90°)</th>\n",
       "      <th id=\"T_360ba_level0_col19\" class=\"col_heading level0 col19\" >US20(angle:+105°)</th>\n",
       "      <th id=\"T_360ba_level0_col20\" class=\"col_heading level0 col20\" >US21(angle:+120°)</th>\n",
       "      <th id=\"T_360ba_level0_col21\" class=\"col_heading level0 col21\" >US22(angle:+135°)</th>\n",
       "      <th id=\"T_360ba_level0_col22\" class=\"col_heading level0 col22\" >US23(angle:+150°)</th>\n",
       "      <th id=\"T_360ba_level0_col23\" class=\"col_heading level0 col23\" >US24(angle:+165°)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row0\" class=\"row_heading level0 row0\" >US1(angle:-180°)</th>\n",
       "      <td id=\"T_360ba_row0_col0\" class=\"data row0 col0\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row0_col1\" class=\"data row0 col1\" >0.227981</td>\n",
       "      <td id=\"T_360ba_row0_col2\" class=\"data row0 col2\" >0.140681</td>\n",
       "      <td id=\"T_360ba_row0_col3\" class=\"data row0 col3\" >0.057997</td>\n",
       "      <td id=\"T_360ba_row0_col4\" class=\"data row0 col4\" >0.019948</td>\n",
       "      <td id=\"T_360ba_row0_col5\" class=\"data row0 col5\" >-0.113879</td>\n",
       "      <td id=\"T_360ba_row0_col6\" class=\"data row0 col6\" >-0.284184</td>\n",
       "      <td id=\"T_360ba_row0_col7\" class=\"data row0 col7\" >-0.193975</td>\n",
       "      <td id=\"T_360ba_row0_col8\" class=\"data row0 col8\" >-0.305249</td>\n",
       "      <td id=\"T_360ba_row0_col9\" class=\"data row0 col9\" >-0.279354</td>\n",
       "      <td id=\"T_360ba_row0_col10\" class=\"data row0 col10\" >-0.207491</td>\n",
       "      <td id=\"T_360ba_row0_col11\" class=\"data row0 col11\" >-0.235220</td>\n",
       "      <td id=\"T_360ba_row0_col12\" class=\"data row0 col12\" >-0.157099</td>\n",
       "      <td id=\"T_360ba_row0_col13\" class=\"data row0 col13\" >-0.220978</td>\n",
       "      <td id=\"T_360ba_row0_col14\" class=\"data row0 col14\" >-0.135658</td>\n",
       "      <td id=\"T_360ba_row0_col15\" class=\"data row0 col15\" >0.065822</td>\n",
       "      <td id=\"T_360ba_row0_col16\" class=\"data row0 col16\" >0.137558</td>\n",
       "      <td id=\"T_360ba_row0_col17\" class=\"data row0 col17\" >0.163184</td>\n",
       "      <td id=\"T_360ba_row0_col18\" class=\"data row0 col18\" >0.185377</td>\n",
       "      <td id=\"T_360ba_row0_col19\" class=\"data row0 col19\" >0.181525</td>\n",
       "      <td id=\"T_360ba_row0_col20\" class=\"data row0 col20\" >0.280650</td>\n",
       "      <td id=\"T_360ba_row0_col21\" class=\"data row0 col21\" >0.341608</td>\n",
       "      <td id=\"T_360ba_row0_col22\" class=\"data row0 col22\" >0.382660</td>\n",
       "      <td id=\"T_360ba_row0_col23\" class=\"data row0 col23\" >0.468430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row1\" class=\"row_heading level0 row1\" >US2(angle:-165°)</th>\n",
       "      <td id=\"T_360ba_row1_col0\" class=\"data row1 col0\" >0.227981</td>\n",
       "      <td id=\"T_360ba_row1_col1\" class=\"data row1 col1\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row1_col2\" class=\"data row1 col2\" >0.323046</td>\n",
       "      <td id=\"T_360ba_row1_col3\" class=\"data row1 col3\" >0.163488</td>\n",
       "      <td id=\"T_360ba_row1_col4\" class=\"data row1 col4\" >0.149413</td>\n",
       "      <td id=\"T_360ba_row1_col5\" class=\"data row1 col5\" >0.048565</td>\n",
       "      <td id=\"T_360ba_row1_col6\" class=\"data row1 col6\" >-0.055028</td>\n",
       "      <td id=\"T_360ba_row1_col7\" class=\"data row1 col7\" >-0.008596</td>\n",
       "      <td id=\"T_360ba_row1_col8\" class=\"data row1 col8\" >-0.052476</td>\n",
       "      <td id=\"T_360ba_row1_col9\" class=\"data row1 col9\" >-0.092430</td>\n",
       "      <td id=\"T_360ba_row1_col10\" class=\"data row1 col10\" >-0.219792</td>\n",
       "      <td id=\"T_360ba_row1_col11\" class=\"data row1 col11\" >-0.147006</td>\n",
       "      <td id=\"T_360ba_row1_col12\" class=\"data row1 col12\" >-0.206211</td>\n",
       "      <td id=\"T_360ba_row1_col13\" class=\"data row1 col13\" >-0.118625</td>\n",
       "      <td id=\"T_360ba_row1_col14\" class=\"data row1 col14\" >-0.061509</td>\n",
       "      <td id=\"T_360ba_row1_col15\" class=\"data row1 col15\" >-0.031577</td>\n",
       "      <td id=\"T_360ba_row1_col16\" class=\"data row1 col16\" >-0.064508</td>\n",
       "      <td id=\"T_360ba_row1_col17\" class=\"data row1 col17\" >-0.032970</td>\n",
       "      <td id=\"T_360ba_row1_col18\" class=\"data row1 col18\" >0.028959</td>\n",
       "      <td id=\"T_360ba_row1_col19\" class=\"data row1 col19\" >0.072769</td>\n",
       "      <td id=\"T_360ba_row1_col20\" class=\"data row1 col20\" >0.053314</td>\n",
       "      <td id=\"T_360ba_row1_col21\" class=\"data row1 col21\" >-0.001614</td>\n",
       "      <td id=\"T_360ba_row1_col22\" class=\"data row1 col22\" >0.057817</td>\n",
       "      <td id=\"T_360ba_row1_col23\" class=\"data row1 col23\" >0.105036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row2\" class=\"row_heading level0 row2\" >US3(angle:-150°)</th>\n",
       "      <td id=\"T_360ba_row2_col0\" class=\"data row2 col0\" >0.140681</td>\n",
       "      <td id=\"T_360ba_row2_col1\" class=\"data row2 col1\" >0.323046</td>\n",
       "      <td id=\"T_360ba_row2_col2\" class=\"data row2 col2\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row2_col3\" class=\"data row2 col3\" >0.396655</td>\n",
       "      <td id=\"T_360ba_row2_col4\" class=\"data row2 col4\" >0.329760</td>\n",
       "      <td id=\"T_360ba_row2_col5\" class=\"data row2 col5\" >0.161001</td>\n",
       "      <td id=\"T_360ba_row2_col6\" class=\"data row2 col6\" >-0.072938</td>\n",
       "      <td id=\"T_360ba_row2_col7\" class=\"data row2 col7\" >-0.186465</td>\n",
       "      <td id=\"T_360ba_row2_col8\" class=\"data row2 col8\" >-0.220046</td>\n",
       "      <td id=\"T_360ba_row2_col9\" class=\"data row2 col9\" >-0.179122</td>\n",
       "      <td id=\"T_360ba_row2_col10\" class=\"data row2 col10\" >-0.157230</td>\n",
       "      <td id=\"T_360ba_row2_col11\" class=\"data row2 col11\" >-0.235032</td>\n",
       "      <td id=\"T_360ba_row2_col12\" class=\"data row2 col12\" >-0.182399</td>\n",
       "      <td id=\"T_360ba_row2_col13\" class=\"data row2 col13\" >-0.162224</td>\n",
       "      <td id=\"T_360ba_row2_col14\" class=\"data row2 col14\" >-0.133359</td>\n",
       "      <td id=\"T_360ba_row2_col15\" class=\"data row2 col15\" >-0.019796</td>\n",
       "      <td id=\"T_360ba_row2_col16\" class=\"data row2 col16\" >-0.069217</td>\n",
       "      <td id=\"T_360ba_row2_col17\" class=\"data row2 col17\" >-0.103051</td>\n",
       "      <td id=\"T_360ba_row2_col18\" class=\"data row2 col18\" >-0.063635</td>\n",
       "      <td id=\"T_360ba_row2_col19\" class=\"data row2 col19\" >-0.006546</td>\n",
       "      <td id=\"T_360ba_row2_col20\" class=\"data row2 col20\" >-0.010129</td>\n",
       "      <td id=\"T_360ba_row2_col21\" class=\"data row2 col21\" >0.130244</td>\n",
       "      <td id=\"T_360ba_row2_col22\" class=\"data row2 col22\" >0.112309</td>\n",
       "      <td id=\"T_360ba_row2_col23\" class=\"data row2 col23\" >0.076078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row3\" class=\"row_heading level0 row3\" >US4(angle:-135°)</th>\n",
       "      <td id=\"T_360ba_row3_col0\" class=\"data row3 col0\" >0.057997</td>\n",
       "      <td id=\"T_360ba_row3_col1\" class=\"data row3 col1\" >0.163488</td>\n",
       "      <td id=\"T_360ba_row3_col2\" class=\"data row3 col2\" >0.396655</td>\n",
       "      <td id=\"T_360ba_row3_col3\" class=\"data row3 col3\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row3_col4\" class=\"data row3 col4\" >0.626803</td>\n",
       "      <td id=\"T_360ba_row3_col5\" class=\"data row3 col5\" >0.315309</td>\n",
       "      <td id=\"T_360ba_row3_col6\" class=\"data row3 col6\" >-0.067964</td>\n",
       "      <td id=\"T_360ba_row3_col7\" class=\"data row3 col7\" >-0.127967</td>\n",
       "      <td id=\"T_360ba_row3_col8\" class=\"data row3 col8\" >-0.169434</td>\n",
       "      <td id=\"T_360ba_row3_col9\" class=\"data row3 col9\" >-0.115552</td>\n",
       "      <td id=\"T_360ba_row3_col10\" class=\"data row3 col10\" >-0.116770</td>\n",
       "      <td id=\"T_360ba_row3_col11\" class=\"data row3 col11\" >-0.165835</td>\n",
       "      <td id=\"T_360ba_row3_col12\" class=\"data row3 col12\" >-0.145734</td>\n",
       "      <td id=\"T_360ba_row3_col13\" class=\"data row3 col13\" >-0.308163</td>\n",
       "      <td id=\"T_360ba_row3_col14\" class=\"data row3 col14\" >-0.271507</td>\n",
       "      <td id=\"T_360ba_row3_col15\" class=\"data row3 col15\" >0.022011</td>\n",
       "      <td id=\"T_360ba_row3_col16\" class=\"data row3 col16\" >-0.120829</td>\n",
       "      <td id=\"T_360ba_row3_col17\" class=\"data row3 col17\" >-0.181193</td>\n",
       "      <td id=\"T_360ba_row3_col18\" class=\"data row3 col18\" >-0.189575</td>\n",
       "      <td id=\"T_360ba_row3_col19\" class=\"data row3 col19\" >-0.094104</td>\n",
       "      <td id=\"T_360ba_row3_col20\" class=\"data row3 col20\" >-0.073197</td>\n",
       "      <td id=\"T_360ba_row3_col21\" class=\"data row3 col21\" >0.270021</td>\n",
       "      <td id=\"T_360ba_row3_col22\" class=\"data row3 col22\" >0.161952</td>\n",
       "      <td id=\"T_360ba_row3_col23\" class=\"data row3 col23\" >0.079978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row4\" class=\"row_heading level0 row4\" >US5(angle:-120°)</th>\n",
       "      <td id=\"T_360ba_row4_col0\" class=\"data row4 col0\" >0.019948</td>\n",
       "      <td id=\"T_360ba_row4_col1\" class=\"data row4 col1\" >0.149413</td>\n",
       "      <td id=\"T_360ba_row4_col2\" class=\"data row4 col2\" >0.329760</td>\n",
       "      <td id=\"T_360ba_row4_col3\" class=\"data row4 col3\" >0.626803</td>\n",
       "      <td id=\"T_360ba_row4_col4\" class=\"data row4 col4\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row4_col5\" class=\"data row4 col5\" >0.489932</td>\n",
       "      <td id=\"T_360ba_row4_col6\" class=\"data row4 col6\" >0.077817</td>\n",
       "      <td id=\"T_360ba_row4_col7\" class=\"data row4 col7\" >-0.002550</td>\n",
       "      <td id=\"T_360ba_row4_col8\" class=\"data row4 col8\" >-0.170375</td>\n",
       "      <td id=\"T_360ba_row4_col9\" class=\"data row4 col9\" >-0.100226</td>\n",
       "      <td id=\"T_360ba_row4_col10\" class=\"data row4 col10\" >-0.123872</td>\n",
       "      <td id=\"T_360ba_row4_col11\" class=\"data row4 col11\" >-0.104604</td>\n",
       "      <td id=\"T_360ba_row4_col12\" class=\"data row4 col12\" >-0.091065</td>\n",
       "      <td id=\"T_360ba_row4_col13\" class=\"data row4 col13\" >-0.280737</td>\n",
       "      <td id=\"T_360ba_row4_col14\" class=\"data row4 col14\" >-0.323145</td>\n",
       "      <td id=\"T_360ba_row4_col15\" class=\"data row4 col15\" >-0.051745</td>\n",
       "      <td id=\"T_360ba_row4_col16\" class=\"data row4 col16\" >-0.161187</td>\n",
       "      <td id=\"T_360ba_row4_col17\" class=\"data row4 col17\" >-0.222219</td>\n",
       "      <td id=\"T_360ba_row4_col18\" class=\"data row4 col18\" >-0.248406</td>\n",
       "      <td id=\"T_360ba_row4_col19\" class=\"data row4 col19\" >-0.195669</td>\n",
       "      <td id=\"T_360ba_row4_col20\" class=\"data row4 col20\" >-0.153191</td>\n",
       "      <td id=\"T_360ba_row4_col21\" class=\"data row4 col21\" >0.114636</td>\n",
       "      <td id=\"T_360ba_row4_col22\" class=\"data row4 col22\" >0.115356</td>\n",
       "      <td id=\"T_360ba_row4_col23\" class=\"data row4 col23\" >0.087500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row5\" class=\"row_heading level0 row5\" >US6(angle:-105°)</th>\n",
       "      <td id=\"T_360ba_row5_col0\" class=\"data row5 col0\" >-0.113879</td>\n",
       "      <td id=\"T_360ba_row5_col1\" class=\"data row5 col1\" >0.048565</td>\n",
       "      <td id=\"T_360ba_row5_col2\" class=\"data row5 col2\" >0.161001</td>\n",
       "      <td id=\"T_360ba_row5_col3\" class=\"data row5 col3\" >0.315309</td>\n",
       "      <td id=\"T_360ba_row5_col4\" class=\"data row5 col4\" >0.489932</td>\n",
       "      <td id=\"T_360ba_row5_col5\" class=\"data row5 col5\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row5_col6\" class=\"data row5 col6\" >0.382777</td>\n",
       "      <td id=\"T_360ba_row5_col7\" class=\"data row5 col7\" >0.200489</td>\n",
       "      <td id=\"T_360ba_row5_col8\" class=\"data row5 col8\" >0.114697</td>\n",
       "      <td id=\"T_360ba_row5_col9\" class=\"data row5 col9\" >-0.019028</td>\n",
       "      <td id=\"T_360ba_row5_col10\" class=\"data row5 col10\" >-0.147906</td>\n",
       "      <td id=\"T_360ba_row5_col11\" class=\"data row5 col11\" >-0.163720</td>\n",
       "      <td id=\"T_360ba_row5_col12\" class=\"data row5 col12\" >-0.137650</td>\n",
       "      <td id=\"T_360ba_row5_col13\" class=\"data row5 col13\" >-0.153418</td>\n",
       "      <td id=\"T_360ba_row5_col14\" class=\"data row5 col14\" >-0.186635</td>\n",
       "      <td id=\"T_360ba_row5_col15\" class=\"data row5 col15\" >-0.147336</td>\n",
       "      <td id=\"T_360ba_row5_col16\" class=\"data row5 col16\" >-0.213275</td>\n",
       "      <td id=\"T_360ba_row5_col17\" class=\"data row5 col17\" >-0.250157</td>\n",
       "      <td id=\"T_360ba_row5_col18\" class=\"data row5 col18\" >-0.251119</td>\n",
       "      <td id=\"T_360ba_row5_col19\" class=\"data row5 col19\" >-0.228092</td>\n",
       "      <td id=\"T_360ba_row5_col20\" class=\"data row5 col20\" >-0.215363</td>\n",
       "      <td id=\"T_360ba_row5_col21\" class=\"data row5 col21\" >-0.103676</td>\n",
       "      <td id=\"T_360ba_row5_col22\" class=\"data row5 col22\" >-0.097973</td>\n",
       "      <td id=\"T_360ba_row5_col23\" class=\"data row5 col23\" >-0.081810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row6\" class=\"row_heading level0 row6\" >US7(angle:-90°)</th>\n",
       "      <td id=\"T_360ba_row6_col0\" class=\"data row6 col0\" >-0.284184</td>\n",
       "      <td id=\"T_360ba_row6_col1\" class=\"data row6 col1\" >-0.055028</td>\n",
       "      <td id=\"T_360ba_row6_col2\" class=\"data row6 col2\" >-0.072938</td>\n",
       "      <td id=\"T_360ba_row6_col3\" class=\"data row6 col3\" >-0.067964</td>\n",
       "      <td id=\"T_360ba_row6_col4\" class=\"data row6 col4\" >0.077817</td>\n",
       "      <td id=\"T_360ba_row6_col5\" class=\"data row6 col5\" >0.382777</td>\n",
       "      <td id=\"T_360ba_row6_col6\" class=\"data row6 col6\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row6_col7\" class=\"data row6 col7\" >0.406891</td>\n",
       "      <td id=\"T_360ba_row6_col8\" class=\"data row6 col8\" >0.432018</td>\n",
       "      <td id=\"T_360ba_row6_col9\" class=\"data row6 col9\" >0.256569</td>\n",
       "      <td id=\"T_360ba_row6_col10\" class=\"data row6 col10\" >0.098363</td>\n",
       "      <td id=\"T_360ba_row6_col11\" class=\"data row6 col11\" >0.031076</td>\n",
       "      <td id=\"T_360ba_row6_col12\" class=\"data row6 col12\" >0.024852</td>\n",
       "      <td id=\"T_360ba_row6_col13\" class=\"data row6 col13\" >0.089082</td>\n",
       "      <td id=\"T_360ba_row6_col14\" class=\"data row6 col14\" >0.093002</td>\n",
       "      <td id=\"T_360ba_row6_col15\" class=\"data row6 col15\" >-0.142813</td>\n",
       "      <td id=\"T_360ba_row6_col16\" class=\"data row6 col16\" >-0.201018</td>\n",
       "      <td id=\"T_360ba_row6_col17\" class=\"data row6 col17\" >-0.220990</td>\n",
       "      <td id=\"T_360ba_row6_col18\" class=\"data row6 col18\" >-0.272319</td>\n",
       "      <td id=\"T_360ba_row6_col19\" class=\"data row6 col19\" >-0.240093</td>\n",
       "      <td id=\"T_360ba_row6_col20\" class=\"data row6 col20\" >-0.273235</td>\n",
       "      <td id=\"T_360ba_row6_col21\" class=\"data row6 col21\" >-0.372338</td>\n",
       "      <td id=\"T_360ba_row6_col22\" class=\"data row6 col22\" >-0.235191</td>\n",
       "      <td id=\"T_360ba_row6_col23\" class=\"data row6 col23\" >-0.259634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row7\" class=\"row_heading level0 row7\" >US8(angle:-75°)</th>\n",
       "      <td id=\"T_360ba_row7_col0\" class=\"data row7 col0\" >-0.193975</td>\n",
       "      <td id=\"T_360ba_row7_col1\" class=\"data row7 col1\" >-0.008596</td>\n",
       "      <td id=\"T_360ba_row7_col2\" class=\"data row7 col2\" >-0.186465</td>\n",
       "      <td id=\"T_360ba_row7_col3\" class=\"data row7 col3\" >-0.127967</td>\n",
       "      <td id=\"T_360ba_row7_col4\" class=\"data row7 col4\" >-0.002550</td>\n",
       "      <td id=\"T_360ba_row7_col5\" class=\"data row7 col5\" >0.200489</td>\n",
       "      <td id=\"T_360ba_row7_col6\" class=\"data row7 col6\" >0.406891</td>\n",
       "      <td id=\"T_360ba_row7_col7\" class=\"data row7 col7\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row7_col8\" class=\"data row7 col8\" >0.536807</td>\n",
       "      <td id=\"T_360ba_row7_col9\" class=\"data row7 col9\" >0.551395</td>\n",
       "      <td id=\"T_360ba_row7_col10\" class=\"data row7 col10\" >0.292638</td>\n",
       "      <td id=\"T_360ba_row7_col11\" class=\"data row7 col11\" >0.003404</td>\n",
       "      <td id=\"T_360ba_row7_col12\" class=\"data row7 col12\" >-0.011000</td>\n",
       "      <td id=\"T_360ba_row7_col13\" class=\"data row7 col13\" >-0.021795</td>\n",
       "      <td id=\"T_360ba_row7_col14\" class=\"data row7 col14\" >-0.064063</td>\n",
       "      <td id=\"T_360ba_row7_col15\" class=\"data row7 col15\" >-0.124273</td>\n",
       "      <td id=\"T_360ba_row7_col16\" class=\"data row7 col16\" >-0.193578</td>\n",
       "      <td id=\"T_360ba_row7_col17\" class=\"data row7 col17\" >-0.209798</td>\n",
       "      <td id=\"T_360ba_row7_col18\" class=\"data row7 col18\" >-0.265394</td>\n",
       "      <td id=\"T_360ba_row7_col19\" class=\"data row7 col19\" >-0.246415</td>\n",
       "      <td id=\"T_360ba_row7_col20\" class=\"data row7 col20\" >-0.278243</td>\n",
       "      <td id=\"T_360ba_row7_col21\" class=\"data row7 col21\" >-0.258908</td>\n",
       "      <td id=\"T_360ba_row7_col22\" class=\"data row7 col22\" >-0.171624</td>\n",
       "      <td id=\"T_360ba_row7_col23\" class=\"data row7 col23\" >-0.143237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row8\" class=\"row_heading level0 row8\" >US9(angle:-60°)</th>\n",
       "      <td id=\"T_360ba_row8_col0\" class=\"data row8 col0\" >-0.305249</td>\n",
       "      <td id=\"T_360ba_row8_col1\" class=\"data row8 col1\" >-0.052476</td>\n",
       "      <td id=\"T_360ba_row8_col2\" class=\"data row8 col2\" >-0.220046</td>\n",
       "      <td id=\"T_360ba_row8_col3\" class=\"data row8 col3\" >-0.169434</td>\n",
       "      <td id=\"T_360ba_row8_col4\" class=\"data row8 col4\" >-0.170375</td>\n",
       "      <td id=\"T_360ba_row8_col5\" class=\"data row8 col5\" >0.114697</td>\n",
       "      <td id=\"T_360ba_row8_col6\" class=\"data row8 col6\" >0.432018</td>\n",
       "      <td id=\"T_360ba_row8_col7\" class=\"data row8 col7\" >0.536807</td>\n",
       "      <td id=\"T_360ba_row8_col8\" class=\"data row8 col8\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row8_col9\" class=\"data row8 col9\" >0.564116</td>\n",
       "      <td id=\"T_360ba_row8_col10\" class=\"data row8 col10\" >0.262173</td>\n",
       "      <td id=\"T_360ba_row8_col11\" class=\"data row8 col11\" >-0.012491</td>\n",
       "      <td id=\"T_360ba_row8_col12\" class=\"data row8 col12\" >0.018771</td>\n",
       "      <td id=\"T_360ba_row8_col13\" class=\"data row8 col13\" >0.032239</td>\n",
       "      <td id=\"T_360ba_row8_col14\" class=\"data row8 col14\" >0.127596</td>\n",
       "      <td id=\"T_360ba_row8_col15\" class=\"data row8 col15\" >-0.117984</td>\n",
       "      <td id=\"T_360ba_row8_col16\" class=\"data row8 col16\" >-0.232477</td>\n",
       "      <td id=\"T_360ba_row8_col17\" class=\"data row8 col17\" >-0.264316</td>\n",
       "      <td id=\"T_360ba_row8_col18\" class=\"data row8 col18\" >-0.269450</td>\n",
       "      <td id=\"T_360ba_row8_col19\" class=\"data row8 col19\" >-0.156058</td>\n",
       "      <td id=\"T_360ba_row8_col20\" class=\"data row8 col20\" >-0.189301</td>\n",
       "      <td id=\"T_360ba_row8_col21\" class=\"data row8 col21\" >-0.271245</td>\n",
       "      <td id=\"T_360ba_row8_col22\" class=\"data row8 col22\" >-0.273312</td>\n",
       "      <td id=\"T_360ba_row8_col23\" class=\"data row8 col23\" >-0.334902</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row9\" class=\"row_heading level0 row9\" >US10(angle:-45°)</th>\n",
       "      <td id=\"T_360ba_row9_col0\" class=\"data row9 col0\" >-0.279354</td>\n",
       "      <td id=\"T_360ba_row9_col1\" class=\"data row9 col1\" >-0.092430</td>\n",
       "      <td id=\"T_360ba_row9_col2\" class=\"data row9 col2\" >-0.179122</td>\n",
       "      <td id=\"T_360ba_row9_col3\" class=\"data row9 col3\" >-0.115552</td>\n",
       "      <td id=\"T_360ba_row9_col4\" class=\"data row9 col4\" >-0.100226</td>\n",
       "      <td id=\"T_360ba_row9_col5\" class=\"data row9 col5\" >-0.019028</td>\n",
       "      <td id=\"T_360ba_row9_col6\" class=\"data row9 col6\" >0.256569</td>\n",
       "      <td id=\"T_360ba_row9_col7\" class=\"data row9 col7\" >0.551395</td>\n",
       "      <td id=\"T_360ba_row9_col8\" class=\"data row9 col8\" >0.564116</td>\n",
       "      <td id=\"T_360ba_row9_col9\" class=\"data row9 col9\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row9_col10\" class=\"data row9 col10\" >0.469736</td>\n",
       "      <td id=\"T_360ba_row9_col11\" class=\"data row9 col11\" >0.128584</td>\n",
       "      <td id=\"T_360ba_row9_col12\" class=\"data row9 col12\" >0.065392</td>\n",
       "      <td id=\"T_360ba_row9_col13\" class=\"data row9 col13\" >-0.001732</td>\n",
       "      <td id=\"T_360ba_row9_col14\" class=\"data row9 col14\" >0.039170</td>\n",
       "      <td id=\"T_360ba_row9_col15\" class=\"data row9 col15\" >-0.030759</td>\n",
       "      <td id=\"T_360ba_row9_col16\" class=\"data row9 col16\" >-0.199329</td>\n",
       "      <td id=\"T_360ba_row9_col17\" class=\"data row9 col17\" >-0.239952</td>\n",
       "      <td id=\"T_360ba_row9_col18\" class=\"data row9 col18\" >-0.297213</td>\n",
       "      <td id=\"T_360ba_row9_col19\" class=\"data row9 col19\" >-0.224384</td>\n",
       "      <td id=\"T_360ba_row9_col20\" class=\"data row9 col20\" >-0.210885</td>\n",
       "      <td id=\"T_360ba_row9_col21\" class=\"data row9 col21\" >-0.179306</td>\n",
       "      <td id=\"T_360ba_row9_col22\" class=\"data row9 col22\" >-0.208651</td>\n",
       "      <td id=\"T_360ba_row9_col23\" class=\"data row9 col23\" >-0.263427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row10\" class=\"row_heading level0 row10\" >US11(angle:-30°)</th>\n",
       "      <td id=\"T_360ba_row10_col0\" class=\"data row10 col0\" >-0.207491</td>\n",
       "      <td id=\"T_360ba_row10_col1\" class=\"data row10 col1\" >-0.219792</td>\n",
       "      <td id=\"T_360ba_row10_col2\" class=\"data row10 col2\" >-0.157230</td>\n",
       "      <td id=\"T_360ba_row10_col3\" class=\"data row10 col3\" >-0.116770</td>\n",
       "      <td id=\"T_360ba_row10_col4\" class=\"data row10 col4\" >-0.123872</td>\n",
       "      <td id=\"T_360ba_row10_col5\" class=\"data row10 col5\" >-0.147906</td>\n",
       "      <td id=\"T_360ba_row10_col6\" class=\"data row10 col6\" >0.098363</td>\n",
       "      <td id=\"T_360ba_row10_col7\" class=\"data row10 col7\" >0.292638</td>\n",
       "      <td id=\"T_360ba_row10_col8\" class=\"data row10 col8\" >0.262173</td>\n",
       "      <td id=\"T_360ba_row10_col9\" class=\"data row10 col9\" >0.469736</td>\n",
       "      <td id=\"T_360ba_row10_col10\" class=\"data row10 col10\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row10_col11\" class=\"data row10 col11\" >0.321842</td>\n",
       "      <td id=\"T_360ba_row10_col12\" class=\"data row10 col12\" >0.256480</td>\n",
       "      <td id=\"T_360ba_row10_col13\" class=\"data row10 col13\" >0.056979</td>\n",
       "      <td id=\"T_360ba_row10_col14\" class=\"data row10 col14\" >-0.026911</td>\n",
       "      <td id=\"T_360ba_row10_col15\" class=\"data row10 col15\" >-0.004358</td>\n",
       "      <td id=\"T_360ba_row10_col16\" class=\"data row10 col16\" >-0.062371</td>\n",
       "      <td id=\"T_360ba_row10_col17\" class=\"data row10 col17\" >-0.109127</td>\n",
       "      <td id=\"T_360ba_row10_col18\" class=\"data row10 col18\" >-0.174173</td>\n",
       "      <td id=\"T_360ba_row10_col19\" class=\"data row10 col19\" >-0.262037</td>\n",
       "      <td id=\"T_360ba_row10_col20\" class=\"data row10 col20\" >-0.302757</td>\n",
       "      <td id=\"T_360ba_row10_col21\" class=\"data row10 col21\" >-0.084473</td>\n",
       "      <td id=\"T_360ba_row10_col22\" class=\"data row10 col22\" >-0.115065</td>\n",
       "      <td id=\"T_360ba_row10_col23\" class=\"data row10 col23\" >-0.168136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row11\" class=\"row_heading level0 row11\" >US12(angle:-15°)</th>\n",
       "      <td id=\"T_360ba_row11_col0\" class=\"data row11 col0\" >-0.235220</td>\n",
       "      <td id=\"T_360ba_row11_col1\" class=\"data row11 col1\" >-0.147006</td>\n",
       "      <td id=\"T_360ba_row11_col2\" class=\"data row11 col2\" >-0.235032</td>\n",
       "      <td id=\"T_360ba_row11_col3\" class=\"data row11 col3\" >-0.165835</td>\n",
       "      <td id=\"T_360ba_row11_col4\" class=\"data row11 col4\" >-0.104604</td>\n",
       "      <td id=\"T_360ba_row11_col5\" class=\"data row11 col5\" >-0.163720</td>\n",
       "      <td id=\"T_360ba_row11_col6\" class=\"data row11 col6\" >0.031076</td>\n",
       "      <td id=\"T_360ba_row11_col7\" class=\"data row11 col7\" >0.003404</td>\n",
       "      <td id=\"T_360ba_row11_col8\" class=\"data row11 col8\" >-0.012491</td>\n",
       "      <td id=\"T_360ba_row11_col9\" class=\"data row11 col9\" >0.128584</td>\n",
       "      <td id=\"T_360ba_row11_col10\" class=\"data row11 col10\" >0.321842</td>\n",
       "      <td id=\"T_360ba_row11_col11\" class=\"data row11 col11\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row11_col12\" class=\"data row11 col12\" >0.469767</td>\n",
       "      <td id=\"T_360ba_row11_col13\" class=\"data row11 col13\" >0.354018</td>\n",
       "      <td id=\"T_360ba_row11_col14\" class=\"data row11 col14\" >0.150226</td>\n",
       "      <td id=\"T_360ba_row11_col15\" class=\"data row11 col15\" >-0.048639</td>\n",
       "      <td id=\"T_360ba_row11_col16\" class=\"data row11 col16\" >0.019440</td>\n",
       "      <td id=\"T_360ba_row11_col17\" class=\"data row11 col17\" >0.032984</td>\n",
       "      <td id=\"T_360ba_row11_col18\" class=\"data row11 col18\" >0.003972</td>\n",
       "      <td id=\"T_360ba_row11_col19\" class=\"data row11 col19\" >-0.128698</td>\n",
       "      <td id=\"T_360ba_row11_col20\" class=\"data row11 col20\" >-0.264360</td>\n",
       "      <td id=\"T_360ba_row11_col21\" class=\"data row11 col21\" >-0.302604</td>\n",
       "      <td id=\"T_360ba_row11_col22\" class=\"data row11 col22\" >-0.186996</td>\n",
       "      <td id=\"T_360ba_row11_col23\" class=\"data row11 col23\" >-0.160546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row12\" class=\"row_heading level0 row12\" >US13(angle:0°)</th>\n",
       "      <td id=\"T_360ba_row12_col0\" class=\"data row12 col0\" >-0.157099</td>\n",
       "      <td id=\"T_360ba_row12_col1\" class=\"data row12 col1\" >-0.206211</td>\n",
       "      <td id=\"T_360ba_row12_col2\" class=\"data row12 col2\" >-0.182399</td>\n",
       "      <td id=\"T_360ba_row12_col3\" class=\"data row12 col3\" >-0.145734</td>\n",
       "      <td id=\"T_360ba_row12_col4\" class=\"data row12 col4\" >-0.091065</td>\n",
       "      <td id=\"T_360ba_row12_col5\" class=\"data row12 col5\" >-0.137650</td>\n",
       "      <td id=\"T_360ba_row12_col6\" class=\"data row12 col6\" >0.024852</td>\n",
       "      <td id=\"T_360ba_row12_col7\" class=\"data row12 col7\" >-0.011000</td>\n",
       "      <td id=\"T_360ba_row12_col8\" class=\"data row12 col8\" >0.018771</td>\n",
       "      <td id=\"T_360ba_row12_col9\" class=\"data row12 col9\" >0.065392</td>\n",
       "      <td id=\"T_360ba_row12_col10\" class=\"data row12 col10\" >0.256480</td>\n",
       "      <td id=\"T_360ba_row12_col11\" class=\"data row12 col11\" >0.469767</td>\n",
       "      <td id=\"T_360ba_row12_col12\" class=\"data row12 col12\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row12_col13\" class=\"data row12 col13\" >0.513060</td>\n",
       "      <td id=\"T_360ba_row12_col14\" class=\"data row12 col14\" >0.197737</td>\n",
       "      <td id=\"T_360ba_row12_col15\" class=\"data row12 col15\" >-0.049350</td>\n",
       "      <td id=\"T_360ba_row12_col16\" class=\"data row12 col16\" >0.011522</td>\n",
       "      <td id=\"T_360ba_row12_col17\" class=\"data row12 col17\" >0.053853</td>\n",
       "      <td id=\"T_360ba_row12_col18\" class=\"data row12 col18\" >0.042635</td>\n",
       "      <td id=\"T_360ba_row12_col19\" class=\"data row12 col19\" >-0.064894</td>\n",
       "      <td id=\"T_360ba_row12_col20\" class=\"data row12 col20\" >-0.196783</td>\n",
       "      <td id=\"T_360ba_row12_col21\" class=\"data row12 col21\" >-0.185089</td>\n",
       "      <td id=\"T_360ba_row12_col22\" class=\"data row12 col22\" >-0.139779</td>\n",
       "      <td id=\"T_360ba_row12_col23\" class=\"data row12 col23\" >-0.114091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row13\" class=\"row_heading level0 row13\" >US14(angle:+15°)</th>\n",
       "      <td id=\"T_360ba_row13_col0\" class=\"data row13 col0\" >-0.220978</td>\n",
       "      <td id=\"T_360ba_row13_col1\" class=\"data row13 col1\" >-0.118625</td>\n",
       "      <td id=\"T_360ba_row13_col2\" class=\"data row13 col2\" >-0.162224</td>\n",
       "      <td id=\"T_360ba_row13_col3\" class=\"data row13 col3\" >-0.308163</td>\n",
       "      <td id=\"T_360ba_row13_col4\" class=\"data row13 col4\" >-0.280737</td>\n",
       "      <td id=\"T_360ba_row13_col5\" class=\"data row13 col5\" >-0.153418</td>\n",
       "      <td id=\"T_360ba_row13_col6\" class=\"data row13 col6\" >0.089082</td>\n",
       "      <td id=\"T_360ba_row13_col7\" class=\"data row13 col7\" >-0.021795</td>\n",
       "      <td id=\"T_360ba_row13_col8\" class=\"data row13 col8\" >0.032239</td>\n",
       "      <td id=\"T_360ba_row13_col9\" class=\"data row13 col9\" >-0.001732</td>\n",
       "      <td id=\"T_360ba_row13_col10\" class=\"data row13 col10\" >0.056979</td>\n",
       "      <td id=\"T_360ba_row13_col11\" class=\"data row13 col11\" >0.354018</td>\n",
       "      <td id=\"T_360ba_row13_col12\" class=\"data row13 col12\" >0.513060</td>\n",
       "      <td id=\"T_360ba_row13_col13\" class=\"data row13 col13\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row13_col14\" class=\"data row13 col14\" >0.370364</td>\n",
       "      <td id=\"T_360ba_row13_col15\" class=\"data row13 col15\" >-0.005494</td>\n",
       "      <td id=\"T_360ba_row13_col16\" class=\"data row13 col16\" >0.066959</td>\n",
       "      <td id=\"T_360ba_row13_col17\" class=\"data row13 col17\" >0.087975</td>\n",
       "      <td id=\"T_360ba_row13_col18\" class=\"data row13 col18\" >0.107681</td>\n",
       "      <td id=\"T_360ba_row13_col19\" class=\"data row13 col19\" >0.028992</td>\n",
       "      <td id=\"T_360ba_row13_col20\" class=\"data row13 col20\" >-0.092524</td>\n",
       "      <td id=\"T_360ba_row13_col21\" class=\"data row13 col21\" >-0.284082</td>\n",
       "      <td id=\"T_360ba_row13_col22\" class=\"data row13 col22\" >-0.280006</td>\n",
       "      <td id=\"T_360ba_row13_col23\" class=\"data row13 col23\" >-0.196876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row14\" class=\"row_heading level0 row14\" >US15(angle:+30°)</th>\n",
       "      <td id=\"T_360ba_row14_col0\" class=\"data row14 col0\" >-0.135658</td>\n",
       "      <td id=\"T_360ba_row14_col1\" class=\"data row14 col1\" >-0.061509</td>\n",
       "      <td id=\"T_360ba_row14_col2\" class=\"data row14 col2\" >-0.133359</td>\n",
       "      <td id=\"T_360ba_row14_col3\" class=\"data row14 col3\" >-0.271507</td>\n",
       "      <td id=\"T_360ba_row14_col4\" class=\"data row14 col4\" >-0.323145</td>\n",
       "      <td id=\"T_360ba_row14_col5\" class=\"data row14 col5\" >-0.186635</td>\n",
       "      <td id=\"T_360ba_row14_col6\" class=\"data row14 col6\" >0.093002</td>\n",
       "      <td id=\"T_360ba_row14_col7\" class=\"data row14 col7\" >-0.064063</td>\n",
       "      <td id=\"T_360ba_row14_col8\" class=\"data row14 col8\" >0.127596</td>\n",
       "      <td id=\"T_360ba_row14_col9\" class=\"data row14 col9\" >0.039170</td>\n",
       "      <td id=\"T_360ba_row14_col10\" class=\"data row14 col10\" >-0.026911</td>\n",
       "      <td id=\"T_360ba_row14_col11\" class=\"data row14 col11\" >0.150226</td>\n",
       "      <td id=\"T_360ba_row14_col12\" class=\"data row14 col12\" >0.197737</td>\n",
       "      <td id=\"T_360ba_row14_col13\" class=\"data row14 col13\" >0.370364</td>\n",
       "      <td id=\"T_360ba_row14_col14\" class=\"data row14 col14\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row14_col15\" class=\"data row14 col15\" >0.100354</td>\n",
       "      <td id=\"T_360ba_row14_col16\" class=\"data row14 col16\" >0.054402</td>\n",
       "      <td id=\"T_360ba_row14_col17\" class=\"data row14 col17\" >0.051709</td>\n",
       "      <td id=\"T_360ba_row14_col18\" class=\"data row14 col18\" >0.045517</td>\n",
       "      <td id=\"T_360ba_row14_col19\" class=\"data row14 col19\" >0.056800</td>\n",
       "      <td id=\"T_360ba_row14_col20\" class=\"data row14 col20\" >0.054904</td>\n",
       "      <td id=\"T_360ba_row14_col21\" class=\"data row14 col21\" >-0.182287</td>\n",
       "      <td id=\"T_360ba_row14_col22\" class=\"data row14 col22\" >-0.164560</td>\n",
       "      <td id=\"T_360ba_row14_col23\" class=\"data row14 col23\" >-0.210914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row15\" class=\"row_heading level0 row15\" >US16(angle:+45°)</th>\n",
       "      <td id=\"T_360ba_row15_col0\" class=\"data row15 col0\" >0.065822</td>\n",
       "      <td id=\"T_360ba_row15_col1\" class=\"data row15 col1\" >-0.031577</td>\n",
       "      <td id=\"T_360ba_row15_col2\" class=\"data row15 col2\" >-0.019796</td>\n",
       "      <td id=\"T_360ba_row15_col3\" class=\"data row15 col3\" >0.022011</td>\n",
       "      <td id=\"T_360ba_row15_col4\" class=\"data row15 col4\" >-0.051745</td>\n",
       "      <td id=\"T_360ba_row15_col5\" class=\"data row15 col5\" >-0.147336</td>\n",
       "      <td id=\"T_360ba_row15_col6\" class=\"data row15 col6\" >-0.142813</td>\n",
       "      <td id=\"T_360ba_row15_col7\" class=\"data row15 col7\" >-0.124273</td>\n",
       "      <td id=\"T_360ba_row15_col8\" class=\"data row15 col8\" >-0.117984</td>\n",
       "      <td id=\"T_360ba_row15_col9\" class=\"data row15 col9\" >-0.030759</td>\n",
       "      <td id=\"T_360ba_row15_col10\" class=\"data row15 col10\" >-0.004358</td>\n",
       "      <td id=\"T_360ba_row15_col11\" class=\"data row15 col11\" >-0.048639</td>\n",
       "      <td id=\"T_360ba_row15_col12\" class=\"data row15 col12\" >-0.049350</td>\n",
       "      <td id=\"T_360ba_row15_col13\" class=\"data row15 col13\" >-0.005494</td>\n",
       "      <td id=\"T_360ba_row15_col14\" class=\"data row15 col14\" >0.100354</td>\n",
       "      <td id=\"T_360ba_row15_col15\" class=\"data row15 col15\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row15_col16\" class=\"data row15 col16\" >0.374540</td>\n",
       "      <td id=\"T_360ba_row15_col17\" class=\"data row15 col17\" >0.230554</td>\n",
       "      <td id=\"T_360ba_row15_col18\" class=\"data row15 col18\" >0.026933</td>\n",
       "      <td id=\"T_360ba_row15_col19\" class=\"data row15 col19\" >0.062628</td>\n",
       "      <td id=\"T_360ba_row15_col20\" class=\"data row15 col20\" >0.158016</td>\n",
       "      <td id=\"T_360ba_row15_col21\" class=\"data row15 col21\" >0.096513</td>\n",
       "      <td id=\"T_360ba_row15_col22\" class=\"data row15 col22\" >0.049848</td>\n",
       "      <td id=\"T_360ba_row15_col23\" class=\"data row15 col23\" >0.025650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row16\" class=\"row_heading level0 row16\" >US17(angle:+60°)</th>\n",
       "      <td id=\"T_360ba_row16_col0\" class=\"data row16 col0\" >0.137558</td>\n",
       "      <td id=\"T_360ba_row16_col1\" class=\"data row16 col1\" >-0.064508</td>\n",
       "      <td id=\"T_360ba_row16_col2\" class=\"data row16 col2\" >-0.069217</td>\n",
       "      <td id=\"T_360ba_row16_col3\" class=\"data row16 col3\" >-0.120829</td>\n",
       "      <td id=\"T_360ba_row16_col4\" class=\"data row16 col4\" >-0.161187</td>\n",
       "      <td id=\"T_360ba_row16_col5\" class=\"data row16 col5\" >-0.213275</td>\n",
       "      <td id=\"T_360ba_row16_col6\" class=\"data row16 col6\" >-0.201018</td>\n",
       "      <td id=\"T_360ba_row16_col7\" class=\"data row16 col7\" >-0.193578</td>\n",
       "      <td id=\"T_360ba_row16_col8\" class=\"data row16 col8\" >-0.232477</td>\n",
       "      <td id=\"T_360ba_row16_col9\" class=\"data row16 col9\" >-0.199329</td>\n",
       "      <td id=\"T_360ba_row16_col10\" class=\"data row16 col10\" >-0.062371</td>\n",
       "      <td id=\"T_360ba_row16_col11\" class=\"data row16 col11\" >0.019440</td>\n",
       "      <td id=\"T_360ba_row16_col12\" class=\"data row16 col12\" >0.011522</td>\n",
       "      <td id=\"T_360ba_row16_col13\" class=\"data row16 col13\" >0.066959</td>\n",
       "      <td id=\"T_360ba_row16_col14\" class=\"data row16 col14\" >0.054402</td>\n",
       "      <td id=\"T_360ba_row16_col15\" class=\"data row16 col15\" >0.374540</td>\n",
       "      <td id=\"T_360ba_row16_col16\" class=\"data row16 col16\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row16_col17\" class=\"data row16 col17\" >0.514918</td>\n",
       "      <td id=\"T_360ba_row16_col18\" class=\"data row16 col18\" >0.271167</td>\n",
       "      <td id=\"T_360ba_row16_col19\" class=\"data row16 col19\" >0.069659</td>\n",
       "      <td id=\"T_360ba_row16_col20\" class=\"data row16 col20\" >0.161366</td>\n",
       "      <td id=\"T_360ba_row16_col21\" class=\"data row16 col21\" >0.096269</td>\n",
       "      <td id=\"T_360ba_row16_col22\" class=\"data row16 col22\" >0.130416</td>\n",
       "      <td id=\"T_360ba_row16_col23\" class=\"data row16 col23\" >0.150685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row17\" class=\"row_heading level0 row17\" >US18(angle:+75°)</th>\n",
       "      <td id=\"T_360ba_row17_col0\" class=\"data row17 col0\" >0.163184</td>\n",
       "      <td id=\"T_360ba_row17_col1\" class=\"data row17 col1\" >-0.032970</td>\n",
       "      <td id=\"T_360ba_row17_col2\" class=\"data row17 col2\" >-0.103051</td>\n",
       "      <td id=\"T_360ba_row17_col3\" class=\"data row17 col3\" >-0.181193</td>\n",
       "      <td id=\"T_360ba_row17_col4\" class=\"data row17 col4\" >-0.222219</td>\n",
       "      <td id=\"T_360ba_row17_col5\" class=\"data row17 col5\" >-0.250157</td>\n",
       "      <td id=\"T_360ba_row17_col6\" class=\"data row17 col6\" >-0.220990</td>\n",
       "      <td id=\"T_360ba_row17_col7\" class=\"data row17 col7\" >-0.209798</td>\n",
       "      <td id=\"T_360ba_row17_col8\" class=\"data row17 col8\" >-0.264316</td>\n",
       "      <td id=\"T_360ba_row17_col9\" class=\"data row17 col9\" >-0.239952</td>\n",
       "      <td id=\"T_360ba_row17_col10\" class=\"data row17 col10\" >-0.109127</td>\n",
       "      <td id=\"T_360ba_row17_col11\" class=\"data row17 col11\" >0.032984</td>\n",
       "      <td id=\"T_360ba_row17_col12\" class=\"data row17 col12\" >0.053853</td>\n",
       "      <td id=\"T_360ba_row17_col13\" class=\"data row17 col13\" >0.087975</td>\n",
       "      <td id=\"T_360ba_row17_col14\" class=\"data row17 col14\" >0.051709</td>\n",
       "      <td id=\"T_360ba_row17_col15\" class=\"data row17 col15\" >0.230554</td>\n",
       "      <td id=\"T_360ba_row17_col16\" class=\"data row17 col16\" >0.514918</td>\n",
       "      <td id=\"T_360ba_row17_col17\" class=\"data row17 col17\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row17_col18\" class=\"data row17 col18\" >0.388338</td>\n",
       "      <td id=\"T_360ba_row17_col19\" class=\"data row17 col19\" >0.157571</td>\n",
       "      <td id=\"T_360ba_row17_col20\" class=\"data row17 col20\" >0.176008</td>\n",
       "      <td id=\"T_360ba_row17_col21\" class=\"data row17 col21\" >0.064938</td>\n",
       "      <td id=\"T_360ba_row17_col22\" class=\"data row17 col22\" >0.122593</td>\n",
       "      <td id=\"T_360ba_row17_col23\" class=\"data row17 col23\" >0.164096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row18\" class=\"row_heading level0 row18\" >US19(angle:+90°)</th>\n",
       "      <td id=\"T_360ba_row18_col0\" class=\"data row18 col0\" >0.185377</td>\n",
       "      <td id=\"T_360ba_row18_col1\" class=\"data row18 col1\" >0.028959</td>\n",
       "      <td id=\"T_360ba_row18_col2\" class=\"data row18 col2\" >-0.063635</td>\n",
       "      <td id=\"T_360ba_row18_col3\" class=\"data row18 col3\" >-0.189575</td>\n",
       "      <td id=\"T_360ba_row18_col4\" class=\"data row18 col4\" >-0.248406</td>\n",
       "      <td id=\"T_360ba_row18_col5\" class=\"data row18 col5\" >-0.251119</td>\n",
       "      <td id=\"T_360ba_row18_col6\" class=\"data row18 col6\" >-0.272319</td>\n",
       "      <td id=\"T_360ba_row18_col7\" class=\"data row18 col7\" >-0.265394</td>\n",
       "      <td id=\"T_360ba_row18_col8\" class=\"data row18 col8\" >-0.269450</td>\n",
       "      <td id=\"T_360ba_row18_col9\" class=\"data row18 col9\" >-0.297213</td>\n",
       "      <td id=\"T_360ba_row18_col10\" class=\"data row18 col10\" >-0.174173</td>\n",
       "      <td id=\"T_360ba_row18_col11\" class=\"data row18 col11\" >0.003972</td>\n",
       "      <td id=\"T_360ba_row18_col12\" class=\"data row18 col12\" >0.042635</td>\n",
       "      <td id=\"T_360ba_row18_col13\" class=\"data row18 col13\" >0.107681</td>\n",
       "      <td id=\"T_360ba_row18_col14\" class=\"data row18 col14\" >0.045517</td>\n",
       "      <td id=\"T_360ba_row18_col15\" class=\"data row18 col15\" >0.026933</td>\n",
       "      <td id=\"T_360ba_row18_col16\" class=\"data row18 col16\" >0.271167</td>\n",
       "      <td id=\"T_360ba_row18_col17\" class=\"data row18 col17\" >0.388338</td>\n",
       "      <td id=\"T_360ba_row18_col18\" class=\"data row18 col18\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row18_col19\" class=\"data row18 col19\" >0.284291</td>\n",
       "      <td id=\"T_360ba_row18_col20\" class=\"data row18 col20\" >0.175877</td>\n",
       "      <td id=\"T_360ba_row18_col21\" class=\"data row18 col21\" >0.053998</td>\n",
       "      <td id=\"T_360ba_row18_col22\" class=\"data row18 col22\" >0.072240</td>\n",
       "      <td id=\"T_360ba_row18_col23\" class=\"data row18 col23\" >0.118344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row19\" class=\"row_heading level0 row19\" >US20(angle:+105°)</th>\n",
       "      <td id=\"T_360ba_row19_col0\" class=\"data row19 col0\" >0.181525</td>\n",
       "      <td id=\"T_360ba_row19_col1\" class=\"data row19 col1\" >0.072769</td>\n",
       "      <td id=\"T_360ba_row19_col2\" class=\"data row19 col2\" >-0.006546</td>\n",
       "      <td id=\"T_360ba_row19_col3\" class=\"data row19 col3\" >-0.094104</td>\n",
       "      <td id=\"T_360ba_row19_col4\" class=\"data row19 col4\" >-0.195669</td>\n",
       "      <td id=\"T_360ba_row19_col5\" class=\"data row19 col5\" >-0.228092</td>\n",
       "      <td id=\"T_360ba_row19_col6\" class=\"data row19 col6\" >-0.240093</td>\n",
       "      <td id=\"T_360ba_row19_col7\" class=\"data row19 col7\" >-0.246415</td>\n",
       "      <td id=\"T_360ba_row19_col8\" class=\"data row19 col8\" >-0.156058</td>\n",
       "      <td id=\"T_360ba_row19_col9\" class=\"data row19 col9\" >-0.224384</td>\n",
       "      <td id=\"T_360ba_row19_col10\" class=\"data row19 col10\" >-0.262037</td>\n",
       "      <td id=\"T_360ba_row19_col11\" class=\"data row19 col11\" >-0.128698</td>\n",
       "      <td id=\"T_360ba_row19_col12\" class=\"data row19 col12\" >-0.064894</td>\n",
       "      <td id=\"T_360ba_row19_col13\" class=\"data row19 col13\" >0.028992</td>\n",
       "      <td id=\"T_360ba_row19_col14\" class=\"data row19 col14\" >0.056800</td>\n",
       "      <td id=\"T_360ba_row19_col15\" class=\"data row19 col15\" >0.062628</td>\n",
       "      <td id=\"T_360ba_row19_col16\" class=\"data row19 col16\" >0.069659</td>\n",
       "      <td id=\"T_360ba_row19_col17\" class=\"data row19 col17\" >0.157571</td>\n",
       "      <td id=\"T_360ba_row19_col18\" class=\"data row19 col18\" >0.284291</td>\n",
       "      <td id=\"T_360ba_row19_col19\" class=\"data row19 col19\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row19_col20\" class=\"data row19 col20\" >0.276597</td>\n",
       "      <td id=\"T_360ba_row19_col21\" class=\"data row19 col21\" >0.175442</td>\n",
       "      <td id=\"T_360ba_row19_col22\" class=\"data row19 col22\" >0.117640</td>\n",
       "      <td id=\"T_360ba_row19_col23\" class=\"data row19 col23\" >0.097039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row20\" class=\"row_heading level0 row20\" >US21(angle:+120°)</th>\n",
       "      <td id=\"T_360ba_row20_col0\" class=\"data row20 col0\" >0.280650</td>\n",
       "      <td id=\"T_360ba_row20_col1\" class=\"data row20 col1\" >0.053314</td>\n",
       "      <td id=\"T_360ba_row20_col2\" class=\"data row20 col2\" >-0.010129</td>\n",
       "      <td id=\"T_360ba_row20_col3\" class=\"data row20 col3\" >-0.073197</td>\n",
       "      <td id=\"T_360ba_row20_col4\" class=\"data row20 col4\" >-0.153191</td>\n",
       "      <td id=\"T_360ba_row20_col5\" class=\"data row20 col5\" >-0.215363</td>\n",
       "      <td id=\"T_360ba_row20_col6\" class=\"data row20 col6\" >-0.273235</td>\n",
       "      <td id=\"T_360ba_row20_col7\" class=\"data row20 col7\" >-0.278243</td>\n",
       "      <td id=\"T_360ba_row20_col8\" class=\"data row20 col8\" >-0.189301</td>\n",
       "      <td id=\"T_360ba_row20_col9\" class=\"data row20 col9\" >-0.210885</td>\n",
       "      <td id=\"T_360ba_row20_col10\" class=\"data row20 col10\" >-0.302757</td>\n",
       "      <td id=\"T_360ba_row20_col11\" class=\"data row20 col11\" >-0.264360</td>\n",
       "      <td id=\"T_360ba_row20_col12\" class=\"data row20 col12\" >-0.196783</td>\n",
       "      <td id=\"T_360ba_row20_col13\" class=\"data row20 col13\" >-0.092524</td>\n",
       "      <td id=\"T_360ba_row20_col14\" class=\"data row20 col14\" >0.054904</td>\n",
       "      <td id=\"T_360ba_row20_col15\" class=\"data row20 col15\" >0.158016</td>\n",
       "      <td id=\"T_360ba_row20_col16\" class=\"data row20 col16\" >0.161366</td>\n",
       "      <td id=\"T_360ba_row20_col17\" class=\"data row20 col17\" >0.176008</td>\n",
       "      <td id=\"T_360ba_row20_col18\" class=\"data row20 col18\" >0.175877</td>\n",
       "      <td id=\"T_360ba_row20_col19\" class=\"data row20 col19\" >0.276597</td>\n",
       "      <td id=\"T_360ba_row20_col20\" class=\"data row20 col20\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row20_col21\" class=\"data row20 col21\" >0.358239</td>\n",
       "      <td id=\"T_360ba_row20_col22\" class=\"data row20 col22\" >0.318696</td>\n",
       "      <td id=\"T_360ba_row20_col23\" class=\"data row20 col23\" >0.273300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row21\" class=\"row_heading level0 row21\" >US22(angle:+135°)</th>\n",
       "      <td id=\"T_360ba_row21_col0\" class=\"data row21 col0\" >0.341608</td>\n",
       "      <td id=\"T_360ba_row21_col1\" class=\"data row21 col1\" >-0.001614</td>\n",
       "      <td id=\"T_360ba_row21_col2\" class=\"data row21 col2\" >0.130244</td>\n",
       "      <td id=\"T_360ba_row21_col3\" class=\"data row21 col3\" >0.270021</td>\n",
       "      <td id=\"T_360ba_row21_col4\" class=\"data row21 col4\" >0.114636</td>\n",
       "      <td id=\"T_360ba_row21_col5\" class=\"data row21 col5\" >-0.103676</td>\n",
       "      <td id=\"T_360ba_row21_col6\" class=\"data row21 col6\" >-0.372338</td>\n",
       "      <td id=\"T_360ba_row21_col7\" class=\"data row21 col7\" >-0.258908</td>\n",
       "      <td id=\"T_360ba_row21_col8\" class=\"data row21 col8\" >-0.271245</td>\n",
       "      <td id=\"T_360ba_row21_col9\" class=\"data row21 col9\" >-0.179306</td>\n",
       "      <td id=\"T_360ba_row21_col10\" class=\"data row21 col10\" >-0.084473</td>\n",
       "      <td id=\"T_360ba_row21_col11\" class=\"data row21 col11\" >-0.302604</td>\n",
       "      <td id=\"T_360ba_row21_col12\" class=\"data row21 col12\" >-0.185089</td>\n",
       "      <td id=\"T_360ba_row21_col13\" class=\"data row21 col13\" >-0.284082</td>\n",
       "      <td id=\"T_360ba_row21_col14\" class=\"data row21 col14\" >-0.182287</td>\n",
       "      <td id=\"T_360ba_row21_col15\" class=\"data row21 col15\" >0.096513</td>\n",
       "      <td id=\"T_360ba_row21_col16\" class=\"data row21 col16\" >0.096269</td>\n",
       "      <td id=\"T_360ba_row21_col17\" class=\"data row21 col17\" >0.064938</td>\n",
       "      <td id=\"T_360ba_row21_col18\" class=\"data row21 col18\" >0.053998</td>\n",
       "      <td id=\"T_360ba_row21_col19\" class=\"data row21 col19\" >0.175442</td>\n",
       "      <td id=\"T_360ba_row21_col20\" class=\"data row21 col20\" >0.358239</td>\n",
       "      <td id=\"T_360ba_row21_col21\" class=\"data row21 col21\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row21_col22\" class=\"data row21 col22\" >0.454413</td>\n",
       "      <td id=\"T_360ba_row21_col23\" class=\"data row21 col23\" >0.295771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row22\" class=\"row_heading level0 row22\" >US23(angle:+150°)</th>\n",
       "      <td id=\"T_360ba_row22_col0\" class=\"data row22 col0\" >0.382660</td>\n",
       "      <td id=\"T_360ba_row22_col1\" class=\"data row22 col1\" >0.057817</td>\n",
       "      <td id=\"T_360ba_row22_col2\" class=\"data row22 col2\" >0.112309</td>\n",
       "      <td id=\"T_360ba_row22_col3\" class=\"data row22 col3\" >0.161952</td>\n",
       "      <td id=\"T_360ba_row22_col4\" class=\"data row22 col4\" >0.115356</td>\n",
       "      <td id=\"T_360ba_row22_col5\" class=\"data row22 col5\" >-0.097973</td>\n",
       "      <td id=\"T_360ba_row22_col6\" class=\"data row22 col6\" >-0.235191</td>\n",
       "      <td id=\"T_360ba_row22_col7\" class=\"data row22 col7\" >-0.171624</td>\n",
       "      <td id=\"T_360ba_row22_col8\" class=\"data row22 col8\" >-0.273312</td>\n",
       "      <td id=\"T_360ba_row22_col9\" class=\"data row22 col9\" >-0.208651</td>\n",
       "      <td id=\"T_360ba_row22_col10\" class=\"data row22 col10\" >-0.115065</td>\n",
       "      <td id=\"T_360ba_row22_col11\" class=\"data row22 col11\" >-0.186996</td>\n",
       "      <td id=\"T_360ba_row22_col12\" class=\"data row22 col12\" >-0.139779</td>\n",
       "      <td id=\"T_360ba_row22_col13\" class=\"data row22 col13\" >-0.280006</td>\n",
       "      <td id=\"T_360ba_row22_col14\" class=\"data row22 col14\" >-0.164560</td>\n",
       "      <td id=\"T_360ba_row22_col15\" class=\"data row22 col15\" >0.049848</td>\n",
       "      <td id=\"T_360ba_row22_col16\" class=\"data row22 col16\" >0.130416</td>\n",
       "      <td id=\"T_360ba_row22_col17\" class=\"data row22 col17\" >0.122593</td>\n",
       "      <td id=\"T_360ba_row22_col18\" class=\"data row22 col18\" >0.072240</td>\n",
       "      <td id=\"T_360ba_row22_col19\" class=\"data row22 col19\" >0.117640</td>\n",
       "      <td id=\"T_360ba_row22_col20\" class=\"data row22 col20\" >0.318696</td>\n",
       "      <td id=\"T_360ba_row22_col21\" class=\"data row22 col21\" >0.454413</td>\n",
       "      <td id=\"T_360ba_row22_col22\" class=\"data row22 col22\" >1.000000</td>\n",
       "      <td id=\"T_360ba_row22_col23\" class=\"data row22 col23\" >0.437620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_360ba_level0_row23\" class=\"row_heading level0 row23\" >US24(angle:+165°)</th>\n",
       "      <td id=\"T_360ba_row23_col0\" class=\"data row23 col0\" >0.468430</td>\n",
       "      <td id=\"T_360ba_row23_col1\" class=\"data row23 col1\" >0.105036</td>\n",
       "      <td id=\"T_360ba_row23_col2\" class=\"data row23 col2\" >0.076078</td>\n",
       "      <td id=\"T_360ba_row23_col3\" class=\"data row23 col3\" >0.079978</td>\n",
       "      <td id=\"T_360ba_row23_col4\" class=\"data row23 col4\" >0.087500</td>\n",
       "      <td id=\"T_360ba_row23_col5\" class=\"data row23 col5\" >-0.081810</td>\n",
       "      <td id=\"T_360ba_row23_col6\" class=\"data row23 col6\" >-0.259634</td>\n",
       "      <td id=\"T_360ba_row23_col7\" class=\"data row23 col7\" >-0.143237</td>\n",
       "      <td id=\"T_360ba_row23_col8\" class=\"data row23 col8\" >-0.334902</td>\n",
       "      <td id=\"T_360ba_row23_col9\" class=\"data row23 col9\" >-0.263427</td>\n",
       "      <td id=\"T_360ba_row23_col10\" class=\"data row23 col10\" >-0.168136</td>\n",
       "      <td id=\"T_360ba_row23_col11\" class=\"data row23 col11\" >-0.160546</td>\n",
       "      <td id=\"T_360ba_row23_col12\" class=\"data row23 col12\" >-0.114091</td>\n",
       "      <td id=\"T_360ba_row23_col13\" class=\"data row23 col13\" >-0.196876</td>\n",
       "      <td id=\"T_360ba_row23_col14\" class=\"data row23 col14\" >-0.210914</td>\n",
       "      <td id=\"T_360ba_row23_col15\" class=\"data row23 col15\" >0.025650</td>\n",
       "      <td id=\"T_360ba_row23_col16\" class=\"data row23 col16\" >0.150685</td>\n",
       "      <td id=\"T_360ba_row23_col17\" class=\"data row23 col17\" >0.164096</td>\n",
       "      <td id=\"T_360ba_row23_col18\" class=\"data row23 col18\" >0.118344</td>\n",
       "      <td id=\"T_360ba_row23_col19\" class=\"data row23 col19\" >0.097039</td>\n",
       "      <td id=\"T_360ba_row23_col20\" class=\"data row23 col20\" >0.273300</td>\n",
       "      <td id=\"T_360ba_row23_col21\" class=\"data row23 col21\" >0.295771</td>\n",
       "      <td id=\"T_360ba_row23_col22\" class=\"data row23 col22\" >0.437620</td>\n",
       "      <td id=\"T_360ba_row23_col23\" class=\"data row23 col23\" >1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x1f3d3ef0050>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Correlation = final_df.corr()\n",
    "Correlation.style.applymap(lambda x: 'background-color : yellow' if x > 0.7 else '')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3cd3d2f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "5a5e8c2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(final_df, target ,test_size=0.3, random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "db38b659",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "\n",
    "scaler.fit(X_train)\n",
    "X_train = scaler.transform(X_train)\n",
    "X_test = scaler.transform(X_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "7ea29691",
   "metadata": {},
   "outputs": [],
   "source": [
    "result = pd.DataFrame(columns=['Model','Train Accuracy','Test Accuracy'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b1afd00",
   "metadata": {},
   "source": [
    "# KNN Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "01896463",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "knnclassifier = KNeighborsClassifier()\n",
    "knnclassifier.fit(X_train, y_train)\n",
    "y_pred = knnclassifier.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ab963da2",
   "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>Model</th>\n",
       "      <th>Train Accuracy</th>\n",
       "      <th>Test Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNN Classifier</td>\n",
       "      <td>0.905473</td>\n",
       "      <td>0.842395</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Model  Train Accuracy  Test Accuracy\n",
       "0  KNN Classifier        0.905473       0.842395"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_train_accuracy = accuracy_score(y_train, knnclassifier.predict(X_train))\n",
    "knn_test_accuracy = accuracy_score(y_test, knnclassifier.predict(X_test))\n",
    "\n",
    "result = result._append(pd.Series({'Model':'KNN Classifier','Train Accuracy':knn_train_accuracy,'Test Accuracy':knn_test_accuracy}),ignore_index=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "3e3b5839",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   precision    recall  f1-score   support\n",
      "\n",
      "     Move-Forward       0.83      0.82      0.83       648\n",
      " Sharp-Right-Turn       0.88      0.88      0.88       661\n",
      " Slight-Left-Turn       0.89      0.86      0.87        99\n",
      "Slight-Right-Turn       0.75      0.79      0.77       229\n",
      "\n",
      "         accuracy                           0.84      1637\n",
      "        macro avg       0.84      0.84      0.84      1637\n",
      "     weighted avg       0.84      0.84      0.84      1637\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_pred,y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "297a3811",
   "metadata": {},
   "source": [
    "# Logistic regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "a2894913",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "lgclassifier = LogisticRegression() \n",
    "lgclassifier.fit(X_train, y_train) \n",
    "y_pred = lgclassifier.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "fe1786ed",
   "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>Model</th>\n",
       "      <th>Train Accuracy</th>\n",
       "      <th>Test Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNN Classifier</td>\n",
       "      <td>0.905473</td>\n",
       "      <td>0.842395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logestic Regression</td>\n",
       "      <td>0.713799</td>\n",
       "      <td>0.700672</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Model  Train Accuracy  Test Accuracy\n",
       "0       KNN Classifier        0.905473       0.842395\n",
       "1  Logestic Regression        0.713799       0.700672"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg_train_recall = recall_score(y_train, lgclassifier.predict(X_train),average='weighted')\n",
    "lg_test_recall = recall_score(y_test, lgclassifier.predict(X_test),average='weighted')\n",
    "lg_train_accuracy = accuracy_score(y_train, lgclassifier.predict(X_train))\n",
    "lg_test_accuracy = accuracy_score(y_test, lgclassifier.predict(X_test))\n",
    "result = result._append(pd.Series({'Model':'Logestic Regression','Train Accuracy':lg_train_accuracy,'Test Accuracy':lg_test_accuracy}),ignore_index=True )\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "1e502b70",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   precision    recall  f1-score   support\n",
      "\n",
      "     Move-Forward       0.70      0.73      0.72       611\n",
      " Sharp-Right-Turn       0.77      0.70      0.73       726\n",
      " Slight-Left-Turn       0.54      0.67      0.60        78\n",
      "Slight-Right-Turn       0.58      0.63      0.60       222\n",
      "\n",
      "         accuracy                           0.70      1637\n",
      "        macro avg       0.65      0.68      0.66      1637\n",
      "     weighted avg       0.71      0.70      0.70      1637\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_pred,y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92ba4269",
   "metadata": {},
   "source": [
    "# Decision Tree Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "eed67ed2",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier \n",
    "dtclassifier = DecisionTreeClassifier(criterion = 'entropy', random_state = 0,max_depth=4)\n",
    "dtclassifier.fit(X_train, y_train) \n",
    "y_pred_test = dtclassifier.predict(X_test)\n",
    "y_pred_train = dtclassifier.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "5ba832cd",
   "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>Model</th>\n",
       "      <th>Train Accuracy</th>\n",
       "      <th>Test Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNN Classifier</td>\n",
       "      <td>0.905473</td>\n",
       "      <td>0.842395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logestic Regression</td>\n",
       "      <td>0.713799</td>\n",
       "      <td>0.700672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision Tree Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.909591</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Model  Train Accuracy  Test Accuracy\n",
       "0            KNN Classifier        0.905473       0.842395\n",
       "1       Logestic Regression        0.713799       0.700672\n",
       "2  Decision Tree Classifier        0.913852       0.909591"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dt_train_accuracy = accuracy_score(y_train,y_pred_train )\n",
    "dt_test_accuracy = accuracy_score(y_test, y_pred_test)\n",
    "result = result._append(pd.Series({'Model':'Decision Tree Classifier','Train Accuracy':dt_train_accuracy,'Test Accuracy':dt_test_accuracy}),ignore_index=True )\n",
    "result"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e393ea01",
   "metadata": {},
   "source": [
    "# Random forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "b40d598f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#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>RandomForestClassifier(criterion=&#x27;entropy&#x27;, max_depth=4, max_features=100,\n",
       "                       max_leaf_nodes=30, n_estimators=50)</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\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(criterion=&#x27;entropy&#x27;, max_depth=4, max_features=100,\n",
       "                       max_leaf_nodes=30, n_estimators=50)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomForestClassifier(criterion='entropy', max_depth=4, max_features=100,\n",
       "                       max_leaf_nodes=30, n_estimators=50)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rf_clf = RandomForestClassifier(n_estimators=50, criterion='entropy',max_features=100,max_depth=4,max_leaf_nodes=30)\n",
    "rf_clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "6c1935bc",
   "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>Model</th>\n",
       "      <th>Train Accuracy</th>\n",
       "      <th>Test Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNN Classifier</td>\n",
       "      <td>0.905473</td>\n",
       "      <td>0.842395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logestic Regression</td>\n",
       "      <td>0.713799</td>\n",
       "      <td>0.700672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision Tree Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.909591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random Forest Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.911423</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Model  Train Accuracy  Test Accuracy\n",
       "0            KNN Classifier        0.905473       0.842395\n",
       "1       Logestic Regression        0.713799       0.700672\n",
       "2  Decision Tree Classifier        0.913852       0.909591\n",
       "3  Random Forest Classifier        0.913852       0.911423"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_train = rf_clf.predict(X_train)\n",
    "y_pred_test = rf_clf.predict(X_test)\n",
    "\n",
    "rf_train_accuracy_score = accuracy_score(y_train, y_pred_train)\n",
    "rf_test_accuracy_score = accuracy_score(y_test, y_pred_test)\n",
    "\n",
    "result = result._append(pd.Series({'Model':'Random Forest Classifier','Train Accuracy':rf_train_accuracy_score,'Test Accuracy':rf_test_accuracy_score}),ignore_index=True )\n",
    "result"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "414da94e",
   "metadata": {},
   "source": [
    "# PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "2cdee247",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train1, X_test1, y_train1, y_test1 = train_test_split(X_train, y_train , test_size=0.3, random_state=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "4637d734",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "scaler.fit(X_train1)\n",
    "X_train1 = scaler.transform(X_train1)\n",
    "X_test1 = scaler.transform(X_test1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "0f3c45b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "23"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(.98)\n",
    "pca.fit(X_train1)\n",
    "\n",
    "pca.n_components_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "44011ee3",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train1 = pca.transform(X_train1)\n",
    "X_test1 = pca.transform(X_test1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71ef3e2d",
   "metadata": {},
   "source": [
    "# KNN Classifier with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "2c0e5964",
   "metadata": {},
   "outputs": [],
   "source": [
    "classifier = KNeighborsClassifier()\n",
    "classifier.fit(X_train1, y_train1)\n",
    "Y_pred = classifier.predict(X_test1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "63a5ca26",
   "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>Model</th>\n",
       "      <th>Train Accuracy</th>\n",
       "      <th>Test Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNN Classifier</td>\n",
       "      <td>0.905473</td>\n",
       "      <td>0.842395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logestic Regression</td>\n",
       "      <td>0.713799</td>\n",
       "      <td>0.700672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision Tree Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.909591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random Forest Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.911423</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>KNN with PCA</td>\n",
       "      <td>0.893004</td>\n",
       "      <td>0.831588</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Model  Train Accuracy  Test Accuracy\n",
       "0            KNN Classifier        0.905473       0.842395\n",
       "1       Logestic Regression        0.713799       0.700672\n",
       "2  Decision Tree Classifier        0.913852       0.909591\n",
       "3  Random Forest Classifier        0.913852       0.911423\n",
       "4              KNN with PCA        0.893004       0.831588"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knnp_train_accuracy = accuracy_score(y_train1,classifier.predict(X_train1))\n",
    "knnp_test_accuracy = accuracy_score(y_test1,Y_pred)\n",
    "#print(knnp_train_accuracy,knnp_test_accuracy)\n",
    "result = result._append(pd.Series({'Model':'KNN with PCA','Train Accuracy':knnp_train_accuracy,'Test Accuracy':knnp_test_accuracy}),ignore_index=True )\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "a510cbcf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   precision    recall  f1-score   support\n",
      "\n",
      "     Move-Forward       0.87      0.81      0.84       513\n",
      " Sharp-Right-Turn       0.79      0.89      0.83       393\n",
      " Slight-Left-Turn       0.87      0.88      0.87        59\n",
      "Slight-Right-Turn       0.81      0.75      0.78       181\n",
      "\n",
      "         accuracy                           0.83      1146\n",
      "        macro avg       0.84      0.83      0.83      1146\n",
      "     weighted avg       0.83      0.83      0.83      1146\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(Y_pred,y_test1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f78da0db",
   "metadata": {},
   "source": [
    "# Logistic Regression with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "30ddd2b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "lgpclassifier = LogisticRegression(C=10,random_state = 0) \n",
    "lgpclassifier.fit(X_train1, y_train1) \n",
    "y_pred_train1 = lgpclassifier.predict(X_train1)\n",
    "y_pred_test1 = lgpclassifier.predict(X_test1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "03229c26",
   "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>Model</th>\n",
       "      <th>Train Accuracy</th>\n",
       "      <th>Test Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNN Classifier</td>\n",
       "      <td>0.905473</td>\n",
       "      <td>0.842395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logestic Regression</td>\n",
       "      <td>0.713799</td>\n",
       "      <td>0.700672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision Tree Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.909591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random Forest Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.911423</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>KNN with PCA</td>\n",
       "      <td>0.893004</td>\n",
       "      <td>0.831588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Logestic Regression PCA</td>\n",
       "      <td>0.733633</td>\n",
       "      <td>0.717277</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Model  Train Accuracy  Test Accuracy\n",
       "0            KNN Classifier        0.905473       0.842395\n",
       "1       Logestic Regression        0.713799       0.700672\n",
       "2  Decision Tree Classifier        0.913852       0.909591\n",
       "3  Random Forest Classifier        0.913852       0.911423\n",
       "4              KNN with PCA        0.893004       0.831588\n",
       "5   Logestic Regression PCA        0.733633       0.717277"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lgp_train_accuracy = accuracy_score(y_train1,y_pred_train1)\n",
    "lgp_test_accuracy = accuracy_score(y_test1,y_pred_test1)\n",
    "#print(lgp_train_accuracy,knnp_test_accuracy)\n",
    "result = result._append(pd.Series({'Model':'Logestic Regression PCA','Train Accuracy':lgp_train_accuracy,'Test Accuracy':lgp_test_accuracy}),ignore_index=True )\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "d918e101",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   precision    recall  f1-score   support\n",
      "\n",
      "     Move-Forward       0.77      0.75      0.76       490\n",
      " Sharp-Right-Turn       0.74      0.71      0.72       460\n",
      " Slight-Left-Turn       0.45      0.57      0.50        47\n",
      "Slight-Right-Turn       0.61      0.68      0.65       149\n",
      "\n",
      "         accuracy                           0.72      1146\n",
      "        macro avg       0.64      0.68      0.66      1146\n",
      "     weighted avg       0.72      0.72      0.72      1146\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_pred_test1,y_test1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dea5e89b",
   "metadata": {},
   "source": [
    "# Decision Tree Classifier with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "2c8269cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier \n",
    "dtpclassifier = DecisionTreeClassifier(criterion = 'entropy', random_state = 0,max_depth=5)\n",
    "dtpclassifier.fit(X_train1, y_train1) \n",
    "y_pred_test = dtpclassifier.predict(X_test1)\n",
    "y_pred_train = dtpclassifier.predict(X_train1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "8963ae81",
   "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>Model</th>\n",
       "      <th>Train Accuracy</th>\n",
       "      <th>Test Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNN Classifier</td>\n",
       "      <td>0.905473</td>\n",
       "      <td>0.842395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logestic Regression</td>\n",
       "      <td>0.713799</td>\n",
       "      <td>0.700672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision Tree Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.909591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random Forest Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.911423</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>KNN with PCA</td>\n",
       "      <td>0.893004</td>\n",
       "      <td>0.831588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Logestic Regression PCA</td>\n",
       "      <td>0.733633</td>\n",
       "      <td>0.717277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Decision Tree with PCA</td>\n",
       "      <td>0.717920</td>\n",
       "      <td>0.657941</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Model  Train Accuracy  Test Accuracy\n",
       "0            KNN Classifier        0.905473       0.842395\n",
       "1       Logestic Regression        0.713799       0.700672\n",
       "2  Decision Tree Classifier        0.913852       0.909591\n",
       "3  Random Forest Classifier        0.913852       0.911423\n",
       "4              KNN with PCA        0.893004       0.831588\n",
       "5   Logestic Regression PCA        0.733633       0.717277\n",
       "6    Decision Tree with PCA        0.717920       0.657941"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dtp_train_recall_score = recall_score(y_train1, y_pred_train, average='weighted')\n",
    "dtp_test_recall_score = recall_score(y_test1, y_pred_test, average='weighted')\n",
    "\n",
    "dtp_train_accuracy_score = accuracy_score(y_train1, y_pred_train)\n",
    "dtp_test_accuracy_score = accuracy_score(y_test1, y_pred_test)\n",
    "result = result._append(pd.Series({'Model':'Decision Tree with PCA','Train Accuracy':dtp_train_accuracy_score,'Test Accuracy':dtp_test_accuracy_score}),ignore_index=True )\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ed5f19e8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "f85c41ed",
   "metadata": {},
   "source": [
    "# Random Forest with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "2c652215",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 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-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 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-2 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-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 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-2 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-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 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-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 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-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 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-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(criterion=&#x27;entropy&#x27;, max_depth=4, max_features=100,\n",
       "                       max_leaf_nodes=30, n_estimators=50)</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\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(criterion=&#x27;entropy&#x27;, max_depth=4, max_features=100,\n",
       "                       max_leaf_nodes=30, n_estimators=50)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomForestClassifier(criterion='entropy', max_depth=4, max_features=100,\n",
       "                       max_leaf_nodes=30, n_estimators=50)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rfp_clf = RandomForestClassifier(n_estimators=50, criterion='entropy',max_features=100,max_depth=4,max_leaf_nodes=30)\n",
    "rfp_clf.fit(X_train1, y_train1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "ec2b9c21",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train1 = rfp_clf.predict(X_train1)\n",
    "y_pred_test1 = rfp_clf.predict(X_test1)\n",
    "rfp_train_accuracy_score = accuracy_score(y_train1, y_pred_train1)\n",
    "rfp_test_accuracy_score = accuracy_score(y_test1, y_pred_test1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "5fa74539",
   "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>Model</th>\n",
       "      <th>Train Accuracy</th>\n",
       "      <th>Test Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNN Classifier</td>\n",
       "      <td>0.905473</td>\n",
       "      <td>0.842395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logestic Regression</td>\n",
       "      <td>0.713799</td>\n",
       "      <td>0.700672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Decision Tree Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.909591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random Forest Classifier</td>\n",
       "      <td>0.913852</td>\n",
       "      <td>0.911423</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>KNN with PCA</td>\n",
       "      <td>0.893004</td>\n",
       "      <td>0.831588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Logestic Regression PCA</td>\n",
       "      <td>0.733633</td>\n",
       "      <td>0.717277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Decision Tree with PCA</td>\n",
       "      <td>0.717920</td>\n",
       "      <td>0.657941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Random Forest with PCA</td>\n",
       "      <td>0.700711</td>\n",
       "      <td>0.674520</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Model  Train Accuracy  Test Accuracy\n",
       "0            KNN Classifier        0.905473       0.842395\n",
       "1       Logestic Regression        0.713799       0.700672\n",
       "2  Decision Tree Classifier        0.913852       0.909591\n",
       "3  Random Forest Classifier        0.913852       0.911423\n",
       "4              KNN with PCA        0.893004       0.831588\n",
       "5   Logestic Regression PCA        0.733633       0.717277\n",
       "6    Decision Tree with PCA        0.717920       0.657941\n",
       "7    Random Forest with PCA        0.700711       0.674520"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = result._append(pd.Series({'Model':'Random Forest with PCA','Train Accuracy':rfp_train_accuracy_score,'Test Accuracy':rfp_test_accuracy_score}),ignore_index=True )\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "0ec38f6a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABbMAAAGCCAYAAADaEl4NAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABNcElEQVR4nO3deXhN5/7//9fOPCAIgkpCzAQloYSUmocayjmmY6qhNCUftFXaqtAhqsdUNbQkUq0hWkMVRWqei4pSOa05QdRQQg1Bsn9/+Nnf7mYQEdkrPB/Xta/Lvte91nrfe99xel653ctkNpvNAgAAAAAAAADAwOxsXQAAAAAAAAAAAA9CmA0AAAAAAAAAMDzCbAAAAAAAAACA4RFmAwAAAAAAAAAMjzAbAAAAAAAAAGB4hNkAAAAAAAAAAMMjzAYAAAAAAAAAGB5hNgAAAAAAAADA8AizAQAAAAAAAACGR5gNAMBTymQyZem1adOmR7pPWFiYTCbTI11j586datiwoQoUKKAiRYqocePG2rhxY5p+uTUmSbpx44bCwsIe6loJCQkKCQlRhQoV5OrqqsKFC6tatWoaMGCAEhISHrmmvCw6OlpVq1aVq6urTCaTYmNjM+1//PhxDR482PJZurm5qWrVqnr33Xd15swZS78+ffqodOnSj7f4TJw8eVImk0lRUVFW7emNNyd+VgAAAIAnmYOtCwAAALaxc+dOq/fvv/++Nm7cqA0bNli1V6lS5ZHu079/f7Vs2TLb5586dUotWrRQ1apVtWDBAqWkpCgmJkZ79+7VCy+8YNU3t8Yk3Quzx44dK0lq1KjRA/ufPn1atWrVUsGCBfX666+rYsWKSkpK0uHDh7V48WIdP35c3t7ej1xXXnThwgX17NlTLVu21IwZM+Ts7KwKFSpk2H/lypXq2rWrihQposGDB6tmzZoymUw6ePCgIiMjtWrVKu3fvz8XR5CxEiVKaOfOnSpbtqylLaPxPurPCgAAAPCkI8wGAOApVbduXav3RYsWlZ2dXZr2f7px44bc3NyyfJ9SpUqpVKlS2apRklavXq1r165p7ty5qlSpkiSpffv26fbN7phyw+zZs3Xx4kX99NNPKlOmjKW9Q4cOevvtt5Wampprtdy8eVMuLi6GWQX8+++/686dO+rRo4caNmyYad8TJ06oa9euqlChgjZu3CgPDw/LscaNGys0NFTLli173CVnmbOzc5r5l9F43dzcHuln5Z8e9mcVAAAAMDq2GQEAABlq1KiR/P39tWXLFgUFBcnNzU19+/aVdG+bhObNm6tEiRJydXVV5cqVNXLkSF2/ft3qGultnVC6dGm9+OKLWrNmjWrVqiVXV1dVqlRJkZGRaWqwt7eXJP322285Mqbbt2/rgw8+UKVKleTs7KyiRYvq5Zdf1oULF6z6bdiwQY0aNZKnp6dcXV3l4+OjTp066caNGzp58qSKFi0qSRo7dqxl+5I+ffpkeN9Lly7Jzs5OxYoVS/e4nZ31f5bt3r1bbdu2laenp1xcXFS2bFkNHTrUqs+2bdvUpEkT5c+fX25ubgoKCtKqVaus+kRFRclkMmndunXq27evihYtKjc3NyUnJ0u69z3Wq1dP7u7uypcvn1q0aJFmVfPx48fVtWtXlSxZUs7OzvLy8lKTJk0euBWIJK1YsUL16tWTm5ub8ufPr2bNmlmtoO/Tp48aNGggSerSpYtMJlOmK90nTZqk69eva8aMGVZB9n0mk0kdO3bMtKbp06fr+eefV7FixeTu7q5q1appwoQJunPnjlW//fv368UXX1SxYsXk7OyskiVLqk2bNjp9+rSlzzfffKPnnntOHh4ecnNzk5+fn+VnREq7zUhm481om5GsfEd9+vRRvnz5dPDgQTVv3lz58+dXkyZNsjwOAAAAIC8gzAYAAJlKTExUjx491L17d61evVohISGSpCNHjqh169aKiIjQmjVrNHToUC1evFht27bN0nUPHDig119/XcOGDdN3332n6tWrq1+/ftqyZYtVv06dOqlw4cIaNGiQjh49+khjSU1NVfv27TV+/Hh1795dq1at0vjx4xUTE6NGjRrp5s2bku4FkG3atJGTk5MiIyO1Zs0ajR8/Xu7u7rp9+7ZKlCihNWvWSJL69eunnTt3aufOnRo9enSG965Xr55SU1PVsWNHrV27VlevXs2w79q1axUcHKz4+HhNmjRJP/zwg95991398ccflj6bN29W48aNlZSUpIiICC1cuFD58+dX27ZtFR0dneaaffv2laOjo7766it9++23cnR01EcffaRu3bqpSpUqWrx4sb766itdu3ZNwcHBOnz4sOXc1q1ba9++fZowYYJiYmI0c+ZM1axZU1euXMn0816wYIHat2+vAgUKaOHChYqIiNDly5fVqFEjbdu2TZI0evRoTZ8+XZL00UcfaefOnZoxY0aG11y3bp28vLweabX9sWPH1L17d3311VdauXKl+vXrp08++UQDBw609Ll+/bqaNWumP/74Q9OnT1dMTIymTJkiHx8fXbt2TdK9bW26dOkiPz8/LVq0SKtWrdJ7772nu3fvZnjvhx1vVr8j6d4vatq1a6fGjRvru+++09ixY7M0DgAAACDPMAMAAJjN5t69e5vd3d2t2ho2bGiWZF6/fn2m56ampprv3Llj3rx5s1mS+cCBA5ZjY8aMMf/zPzl8fX3NLi4u5lOnTlnabt68aS5cuLB54MCBVn1XrFhh9vLyMnt7e5u9vb3Nx44dy/aYFi5caJZkXrJkiVW/PXv2mCWZZ8yYYTabzeZvv/3WLMkcGxub4bUvXLhglmQeM2ZMlmpJTU01Dxw40GxnZ2eWZDaZTObKlSubhw0bZj5x4oRV37Jly5rLli1rvnnzZobXq1u3rrlYsWLma9euWdru3r1r9vf3N5cqVcqcmppqNpvN5rlz55olmXv16mV1fnx8vNnBwcE8ZMgQq/Zr166Zixcvbu7cubPZbDabL168aJZknjJlSpbGeV9KSoq5ZMmS5mrVqplTUlKsrl+sWDFzUFCQpW3jxo1mSeZvvvnmgdd1cXEx161bN8t19O7d2+zr65tpnXfu3DHPmzfPbG9vb/7zzz/NZrPZvHfvXrMk8/LlyzM897///a9ZkvnKlSsZ9jlx4oRZknnu3LmWtozG+8+flax+R/fHKckcGRlp1Tcr4wAAAADyClZmAwCATBUqVEiNGzdO0378+HF1795dxYsXl729vRwdHS37/8bFxT3wus8++6x8fHws711cXFShQgWdOnXK0rZjxw516tRJM2bM0Pbt2+Xo6KgXXnhBJ06csPTp37+/fH19szSWlStXqmDBgmrbtq3u3r1reT377LMqXry4Nm3aZKnNyclJr7zyir788ksdP348S9fPjMlk0qxZs3T8+HHNmDFDL7/8su7cuaPJkyeratWq2rx5s6R7+ykfO3ZM/fr1k4uLS7rXun79unbv3q1//etfypcvn6Xd3t5ePXv21OnTp9Nsy9KpUyer92vXrtXdu3fVq1cvq8/CxcVFDRs2tHwWhQsXVtmyZfXJJ59o0qRJ2r9/f5b29/7tt9909uxZ9ezZ02oLlXz58qlTp07atWuXbty4kaXPLqft379f7dq1k6enp2Xu9urVSykpKfr9998lSeXKlVOhQoX01ltvadasWWlWQUtS7dq1JUmdO3fW4sWLdebMmRytM6vf0d/983vOyjgAAACAvIIwGwAAZKpEiRJp2v766y8FBwdr9+7d+uCDD7Rp0ybt2bNHS5culSTLdh2Z8fT0TNPm7Oxsde6HH36oihUrqmPHjvL29tbmzZstgfapU6eUmpqqrVu3qk2bNlkayx9//KErV67IyclJjo6OVq9z587p4sWLkqSyZcvqxx9/VLFixfTaa6+pbNmyKlu2rKZOnZql+2TG19dXr776qiIiInTkyBFFR0fr1q1bevPNNyXJsnd3Zg8CvHz5ssxmc7rfTcmSJSXd26P77/7Z9/6WJbVr107zWURHR1s+C5PJpPXr16tFixaaMGGCatWqpaJFiyo0NDTTbSru3z+jGlNTU3X58uUMz8+Ij4+P1S8zHlZ8fLyCg4N15swZTZ06VVu3btWePXssW3/cn38eHh7avHmznn32Wb399tuqWrWqSpYsqTFjxlj21n7++ee1fPlyS+BcqlQp+fv7a+HChdmu7++y+h3d5+bmpgIFCli1ZWUcAAAAQF7hYOsCAACAsaX3QLoNGzbo7Nmz2rRpk2U1tqQH7qH8sI4dO2YVepcqVUqbN29Wo0aN9MILL6hPnz46deqU3njjjSxdr0iRIvL09LTsd/1P+fPnt/w5ODhYwcHBSklJ0d69ezVt2jQNHTpUXl5e6tq166MN7G86d+6s8PBwHTp0SJIsD5bM7OF8hQoVkp2dnRITE9McO3v2rKR7Y/27f36P949/++23D1zZ7uvrq4iICEn3Vo4vXrxYYWFhun37tmbNmpXuOfe/t4xqtLOzU6FChTK9b3patGihadOmadeuXdnaN3v58uW6fv26li5dajXu9B5mWa1aNS1atEhms1m//PKLoqKiNG7cOLm6umrkyJGSpPbt26t9+/ZKTk7Wrl27FB4eru7du6t06dKqV6/eQ9f3dw/zHUnp/6xmdRwAAABAXsDKbAAA8NDuh2bOzs5W7Z9//nmO3sff31/79u2z2hrhmWee0ebNm2U2mzVmzBiNHDlSfn5+Wbreiy++qEuXLiklJUWBgYFpXhUrVkxzjr29vZ577jnLyt2ff/5Z0v8be1ZWoUvph7rSvVXuCQkJlhXVFSpUUNmyZRUZGank5OR0z3F3d9dzzz2npUuXWt0/NTVVX3/9tUqVKqUKFSpkWk+LFi3k4OCgY8eOpftZBAYGpntehQoV9O6776patWqWzyI9FStW1DPPPKMFCxbIbDZb2q9fv64lS5aoXr16cnNzy7TG9AwbNkzu7u4KCQlRUlJSmuNms1nLli3L8Pz05q7ZbNbs2bMzPadGjRqaPHmyChYsmO64nZ2d1bBhQ3388ceS7m1l8qiy+x09yjgAAAAAI2NlNgAAeGhBQUEqVKiQBg0apDFjxsjR0VHz58/XgQMHcvQ+H3zwgTZs2KBGjRrpzTffVK1atfTnn39q1apVOn36tEqVKqWZM2eqS5cuqly58gOv17VrV82fP1+tW7fW//3f/6lOnTpydHTU6dOntXHjRrVv314vvfSSZs2apQ0bNqhNmzby8fHRrVu3FBkZKUlq2rSppHuruH19ffXdd9+pSZMmKly4sIoUKaLSpUune+8PP/xQ27dvV5cuXfTss8/K1dVVJ06c0GeffaZLly7pk08+sfSdPn262rZtq7p162rYsGHy8fFRfHy81q5dq/nz50uSwsPD1axZM73wwgt644035OTkpBkzZujQoUNauHBhhqt07ytdurTGjRund955R8ePH1fLli1VqFAh/fHHH/rpp5/k7u6usWPH6pdfftHgwYP173//W+XLl5eTk5M2bNigX375JdNVvXZ2dpowYYL+85//6MUXX9TAgQOVnJysTz75RFeuXNH48eMf+H2lp0yZMlq0aJHlcxw8eLBq1qwpSTp8+LAiIyNlNpv10ksvpXt+s2bN5OTkpG7dumnEiBG6deuWZs6cmWbLk5UrV2rGjBnq0KGD/Pz8ZDabtXTpUl25ckXNmjWTJL333ns6ffq0mjRpolKlSunKlSuaOnWq1f7xjyKr31FmsjIOAAAAIK8gzAYAAA/N09NTq1at0uuvv64ePXrI3d1d7du3V3R0tGrVqpVj96lUqZJ+/vlnjR07VlOmTNH58+dVuHBhNWzYUFu3bpWfn5+CgoLUuHFjbd68+YGrke3t7bVixQpNnTpVX331lcLDw+Xg4KBSpUqpYcOGqlatmqR7D4Bct26dxowZo3Pnzilfvnzy9/fXihUr1Lx5c8v1IiIi9Oabb6pdu3ZKTk5W7969FRUVle69e/bsKUlatGiRPvnkEyUlJalw4cIKCAjQ6tWr1apVK0vfFi1aaMuWLRo3bpxCQ0N169YtlSpVSu3atbP0adiwoTZs2KAxY8aoT58+Sk1NVY0aNbRixQq9+OKLWfp8R40apSpVqmjq1KlauHChkpOTVbx4cdWuXVuDBg2SJBUvXlxly5bVjBkzlJCQIJPJJD8/P02cOFFDhgzJ9Prdu3eXu7u7wsPD1aVLF9nb26tu3brauHGjgoKCslRjel588UUdPHhQEydO1KxZs5SQkCA7OzuVKVNGLVu2zLSuSpUqacmSJXr33XfVsWNHeXp6qnv37ho+fLjVd1C+fHkVLFhQEyZM0NmzZ+Xk5KSKFSsqKipKvXv3liQ999xz2rt3r9566y1duHBBBQsWVGBgoDZs2KCqVatme3x/l5XvKDNZGQcAAACQV5jMf/93nwAAAAAAAAAAGBB7ZgMAAAAAAAAADI8wGwAAAAAAAABgeITZAAAAAAAAAADDI8wGAAAAAAAAABgeYTYAAAAAAAAAwPAIswEAAAAAAAAAhudg6wJyW2pqqs6ePav8+fPLZDLZuhwAAAAAAADYkNls1rVr11SyZEnZ2bHuEzCypy7MPnv2rLy9vW1dBgAAAAAAAAwkISFBpUqVsnUZADLx1IXZ+fPnl3TvL6gCBQrYuBoAAAAAAADY0tWrV+Xt7W3JjAAY11MXZt/fWqRAgQKE2QAAAAAAAJAktqMF8gA2AgIAAAAAAAAAGB5hNgAAAAAAAADA8AizAQAAAAAAAACGR5gNAAAAAAAAADA8wmwAAAAAAAAAgOERZgMAAAAAAAAADI8wGwAAAAAAAABgeITZAAAAAAAAAADDI8wGAAAAAAAAABgeYTYAAAAAAAAAwPAIswEAAAAAAAAAhkeYDQAAAAAAAAAwPAdbFwAAMIbSI1fZuoQ0To5vY+sS8JCMNo9OunS3dQnWwpJsXQEAAAAA5FmE2QAA4wrzsHUF1ggiAQAAAACwGcJsgzDcSjZWQwIAAAAAAAAwEPbMBgAAAAAAAAAYHmE2AAAAAAAAAMDwCLMBAAAAAAAAAIbHntkAAAAwDJ4jAgAAACAjrMwGAAAAAAAAABgeYTYAAAAAAAAAwPAIswEAAAAAAAAAhkeYDQAAAAAAAAAwPMJsAAAAAAAAAIDhEWYDAAAAAAAAAAyPMBsAAAAAAAAAYHiE2QAAAAAAAAAAwyPMBgAAAAAAAAAYHmE2AAAAAAAAAMDwCLMBAAAAAAAAAIZHmA0AAAAAAAAAMDzCbAAAAAAAAACA4RFmAwAAAAAAAAAMjzAbAAAAAAAAAGB4DrYuAAYV5mHrCqyFJdm6AgAAAAAAAAA2xMpsAAAAAAAAAIDhEWYDAAAAAAAAAAyPMBsAAAAAAAAAYHjsmQ0AAABkxGjPEZF4lggAAACeWqzMBgAAAAAAAAAYHmE2AAAAAAAAAMDwCLMBAAAAAAAAAIZHmA0AAAAAAAAAMDzCbAAAAAAAAACA4dk8zJ4xY4bKlCkjFxcXBQQEaOvWrZn2nz9/vmrUqCE3NzeVKFFCL7/8si5dupRL1QIAAAAAAAAAbMGmYXZ0dLSGDh2qd955R/v371dwcLBatWql+Pj4dPtv27ZNvXr1Ur9+/fTrr7/qm2++0Z49e9S/f/9crhwAAAAAAAAAkJtsGmZPmjRJ/fr1U//+/VW5cmVNmTJF3t7emjlzZrr9d+3apdKlSys0NFRlypRRgwYNNHDgQO3duzeXKwcAAAAAAAAA5Cabhdm3b9/Wvn371Lx5c6v25s2ba8eOHemeExQUpNOnT2v16tUym836448/9O2336pNmzYZ3ic5OVlXr161egEAAAAAAAAA8habhdkXL15USkqKvLy8rNq9vLx07ty5dM8JCgrS/Pnz1aVLFzk5Oal48eIqWLCgpk2bluF9wsPD5eHhYXl5e3vn6DgAAAAAAAAAAI+fzR8AaTKZrN6bzeY0bfcdPnxYoaGheu+997Rv3z6tWbNGJ06c0KBBgzK8/qhRo5SUlGR5JSQk5Gj9AAAAAAAAAIDHz8FWNy5SpIjs7e3TrMI+f/58mtXa94WHh6t+/fp68803JUnVq1eXu7u7goOD9cEHH6hEiRJpznF2dpazs3PODwAAAAAAAAAAkGtstjLbyclJAQEBiomJsWqPiYlRUFBQuufcuHFDdnbWJdvb20u6t6IbAAAAAAAAAPBksuk2I8OHD9ecOXMUGRmpuLg4DRs2TPHx8ZZtQ0aNGqVevXpZ+rdt21ZLly7VzJkzdfz4cW3fvl2hoaGqU6eOSpYsaathAAAAAAAAAAAeM5ttMyJJXbp00aVLlzRu3DglJibK399fq1evlq+vryQpMTFR8fHxlv59+vTRtWvX9Nlnn+n1119XwYIF1bhxY3388ce2GgIAAAAAAAAAIBfYNMyWpJCQEIWEhKR7LCoqKk3bkCFDNGTIkMdcFQAAAAAAAADASGy6zQgAAAAAAAAAAFlBmA0AAAAAAAAAMDzCbAAAAAAAAACA4RFmAwAAAAAAAAAMjzAbAAAAAAAAAGB4hNkAAAAAAAAAAMMjzAYAAAAAAAAAGB5hNgAAAAAAAADA8AizAQAAAAAAAACGR5gNAAAAAAAAADA8wmwAAAAAAAAAgOERZgMAAAAAAAAADI8wGwAAAAAAAABgeITZAAAAAAAAAADDc7B1AQAAAACQU0qPXGXrEtI4Ob6NrUsAAAB4IrAyGwAAAAAAAABgeITZAAAAAAAAAADDI8wGAAAAAAAAABgeYTYAAAAAAAAAwPB4ACQAAAAAAH9jtAeJ8hBRAADuIcwGAAAAgMcpzMPWFVgLS7J1BQAAANlCmA08IVg9AgAAAAAAgCcZe2YDAAAAAAAAAAyPMBsAAAAAAAAAYHiE2QAAAAAAAAAAw2PPbACPh9EedCTxsCMAAAAAuYJnGgHA48HKbAAAAAAAAACA4RFmAwAAAAAAAAAMjzAbAAAAAAAAAGB4hNkAAAAAAAAAAMMjzAYAAAAAAAAAGB5hNgAAAAAAAADA8AizAQAAAAAAAACG52DrAgAAAAAAAPAYhXnYuoK0wpJsXQGAPIiV2QAAAAAAAAAAwyPMBgAAAAAAAAAYHmE2AAAAAAAAAMDwCLMBAAAAAAAAAIZHmA0AAAAAAAAAMDwHWxcAAAAAAAAyEeZh6wqshSXZugIAwFOKldkAAAAAAAAAAMMjzAYAAAAAAAAAGB5hNgAAAAAAAADA8AizAQAAAAAAAACGR5gNAAAAAAAAADA8wmwAAAAAAAAAgOERZgMAAAAAAAAADI8wGwAAAAAAAABgeITZAAAAAAAAAADDI8wGAAAAAAAAABgeYTYAAAAAAAAAwPAIswEAAAAAAAAAhkeYDQAAAAAAAAAwPJuH2TNmzFCZMmXk4uKigIAAbd26NdP+ycnJeuedd+Tr6ytnZ2eVLVtWkZGRuVQtAAAAAAAAAMAWHGx58+joaA0dOlQzZsxQ/fr19fnnn6tVq1Y6fPiwfHx80j2nc+fO+uOPPxQREaFy5crp/Pnzunv3bi5XDgAAAAAAAADITTYNsydNmqR+/fqpf//+kqQpU6Zo7dq1mjlzpsLDw9P0X7NmjTZv3qzjx4+rcOHCkqTSpUvnZskAAAAAAAAAABuw2TYjt2/f1r59+9S8eXOr9ubNm2vHjh3pnrNixQoFBgZqwoQJeuaZZ1ShQgW98cYbunnzZob3SU5O1tWrV61eAAAAAAAAAIC8xWYrsy9evKiUlBR5eXlZtXt5eencuXPpnnP8+HFt27ZNLi4uWrZsmS5evKiQkBD9+eefGe6bHR4errFjx+Z4/QAAAAAAAACA3GPzB0CaTCar92azOU3bfampqTKZTJo/f77q1Kmj1q1ba9KkSYqKispwdfaoUaOUlJRkeSUkJOT4GAAAAAAAAAAAj5fNVmYXKVJE9vb2aVZhnz9/Ps1q7ftKlCihZ555Rh4eHpa2ypUry2w26/Tp0ypfvnyac5ydneXs7JyzxQMAAAAAAAAAcpXNVmY7OTkpICBAMTExVu0xMTEKCgpK95z69evr7Nmz+uuvvyxtv//+u+zs7FSqVKnHWi8AAAAAAAAAwHZsus3I8OHDNWfOHEVGRiouLk7Dhg1TfHy8Bg0aJOneFiG9evWy9O/evbs8PT318ssv6/Dhw9qyZYvefPNN9e3bV66urrYaBgAAAAAAAADgMbPZNiOS1KVLF126dEnjxo1TYmKi/P39tXr1avn6+kqSEhMTFR8fb+mfL18+xcTEaMiQIQoMDJSnp6c6d+6sDz74wFZDAAAAAAAAAADkApuG2ZIUEhKikJCQdI9FRUWlaatUqVKarUkAAAAAAAAAAE82m24zAgAAAAAAAABAVhBmAwAAAAAAAAAMjzAbAAAAAAAAAGB4hNkAAAAAAAAAAMMjzAYAAAAAAAAAGB5hNgAAAAAAAADA8AizAQAAAAAAAACGR5gNAAAAAAAAADA8wmwAAAAAAAAAgOERZgMAAAAAAAAADI8wGwAAAAAAAABgeDkSZl+9elXLly9XXFxcTlwOAAAAAAAAAAAr2QqzO3furM8++0ySdPPmTQUGBqpz586qXr26lixZkqMFAgAAAAAAAACQrTB7y5YtCg4OliQtW7ZMZrNZV65c0aeffqoPPvggRwsEAAAAAAAAACBbYXZSUpIKFy4sSVqzZo06deokNzc3tWnTRkeOHMnRAgEAAAAAAAAAyFaY7e3trZ07d+r69etas2aNmjdvLkm6fPmyXFxccrRAAAAAAAAAAAAcsnPS0KFD9Z///Ef58uWTj4+PGjVqJOne9iPVqlXLyfoAAAAAAAAAAMhemB0SEqI6deooISFBzZo1k53dvQXefn5+7JkNAAAAAAAAAMhx2QqzJSkwMFDVq1fXiRMnVLZsWTk4OKhNmzY5WRsAAAAAAAAAAJKyuWf2jRs31K9fP7m5ualq1aqKj4+XJIWGhmr8+PE5WiAAAAAAAAAAANkKs0eNGqUDBw5o06ZNVg98bNq0qaKjo3OsOAAAAAAAAAAApGxuM7J8+XJFR0erbt26MplMlvYqVaro2LFjOVYcAAAAAAAAAABSNldmX7hwQcWKFUvTfv36datwGwAAAAAAAACAnJCtMLt27dpatWqV5f39AHv27NmqV69ezlQGAAAAAAAAAMD/L1vbjISHh6tly5Y6fPiw7t69q6lTp+rXX3/Vzp07tXnz5pyuEQAAAAAAAADwlMvWyuygoCDt2LFDN27cUNmyZbVu3Tp5eXlp586dCggIyOkaAQAAAAAAAABPuYdemX3nzh298sorGj16tL788svHURMAAAAAAAAAAFYeemW2o6Ojli1b9jhqAQAAAAAAAAAgXdnaZuSll17S8uXLc7gUAAAAAAAAAADSl60HQJYrV07vv/++duzYoYCAALm7u1sdDw0NzZHiAAAAAAAAAACQshlmz5kzRwULFtS+ffu0b98+q2Mmk4kwGwAAAAAAAACQo7IVZp84cSKn6wAAAAAAAAAAIEPZ2jP778xms8xmc07UAgAAAAAAAABAurIdZs+bN0/VqlWTq6urXF1dVb16dX311Vc5WRsAAAAAAAAAAJKyuc3IpEmTNHr0aA0ePFj169eX2WzW9u3bNWjQIF28eFHDhg3L6ToBAAAAAAAAAE+xbIXZ06ZN08yZM9WrVy9LW/v27VW1alWFhYURZgMAAAAAAAAAclS2thlJTExUUFBQmvagoCAlJiY+clEAAAAAAAAAAPxdtsLscuXKafHixWnao6OjVb58+UcuCgAAAAAAAACAv8vWNiNjx45Vly5dtGXLFtWvX18mk0nbtm3T+vXr0w25AQAAAAAAAAB4FNlamd2pUyft3r1bRYoU0fLly7V06VIVKVJEP/30k1566aWcrhEAAAAAAAAA8JTL1spsSQoICNDXX3+dk7UAAAAAAAAAAJCubK3MXr16tdauXZumfe3atfrhhx8euSgAAAAAAAAAAP4uW2H2yJEjlZKSkqbdbDZr5MiRj1wUAAAAAAAAAAB/l60w+8iRI6pSpUqa9kqVKuno0aOPXBQAAAAAAAAAAH+XrTDbw8NDx48fT9N+9OhRubu7P3JRAAAAAAAAAAD8XbbC7Hbt2mno0KE6duyYpe3o0aN6/fXX1a5duxwrDgAAAAAAAAAAKZth9ieffCJ3d3dVqlRJZcqUUZkyZVSpUiV5enrqv//9b07XCAAAAAAAAAB4yjlk5yQPDw/t2LFDMTExOnDggFxdXVWjRg0FBwfndH0AAAAAAAAAADzcyuzdu3frhx9+kCSZTCY1b95cxYoV03//+1916tRJr7zyipKTkx9LoQAAAAAAAACAp9dDhdlhYWH65ZdfLO8PHjyoAQMGqFmzZho5cqS+//57hYeH53iRAAAAAAAAAICn20OF2bGxsWrSpInl/aJFi1SnTh3Nnj1bw4cP16effqrFixfneJEAAAAAAAAAgKfbQ4XZly9flpeXl+X95s2b1bJlS8v72rVrKyEhIeeqAwAAAAAAAABADxlme3l56cSJE5Kk27dv6+eff1a9evUsx69duyZHR8ecrRAAAAAAAAAA8NR7qDC7ZcuWGjlypLZu3apRo0bJzc1NwcHBluO//PKLypYt+1AFzJgxQ2XKlJGLi4sCAgK0devWLJ23fft2OTg46Nlnn32o+wEAAAAAAAAA8p6HCrM/+OAD2dvbq2HDhpo9e7Zmz54tJycny/HIyEg1b948y9eLjo7W0KFD9c4772j//v0KDg5Wq1atFB8fn+l5SUlJ6tWrl9X+3QAAAAAAAACAJ5fDw3QuWrSotm7dqqSkJOXLl0/29vZWx7/55hvly5cvy9ebNGmS+vXrp/79+0uSpkyZorVr12rmzJkKDw/P8LyBAweqe/fusre31/Llyx9mCAAAAAAAAACAPOihVmbf5+HhkSbIlqTChQtbrdTOzO3bt7Vv3740K7mbN2+uHTt2ZHje3LlzdezYMY0ZMyZL90lOTtbVq1etXgAAAAAAAACAvCVbYXZOuHjxolJSUuTl5WXV7uXlpXPnzqV7zpEjRzRy5EjNnz9fDg5ZW1QeHh4uDw8Py8vb2/uRawcAAAAAAAAA5C6bhdn3mUwmq/dmszlNmySlpKSoe/fuGjt2rCpUqJDl648aNUpJSUmWV0JCwiPXDAAAAAAAAADIXQ+1Z3ZOKlKkiOzt7dOswj5//nya1dqSdO3aNe3du1f79+/X4MGDJUmpqakym81ycHDQunXr1Lhx4zTnOTs7y9nZ+fEMAgAAAAAAAACQK2y2MtvJyUkBAQGKiYmxao+JiVFQUFCa/gUKFNDBgwcVGxtreQ0aNEgVK1ZUbGysnnvuudwqHQAAAAAAAACQy2y2MluShg8frp49eyowMFD16tXTF198ofj4eA0aNEjSvS1Czpw5o3nz5snOzk7+/v5W5xcrVkwuLi5p2gEAAAAAAAAATxabhtldunTRpUuXNG7cOCUmJsrf31+rV6+Wr6+vJCkxMVHx8fG2LBEAAAAAAAAAYAA2DbMlKSQkRCEhIekei4qKyvTcsLAwhYWF5XxRAAAAAAAAAABDsdme2QAAAAAAAAAAZBVhNgAAAAAAAADA8AizAQAAAAAAAACGR5gNAAAAAAAAADA8wmwAAAAAAAAAgOERZgMAAAAAAAAADI8wGwAAAAAAAABgeITZAAAAAAAAAADDI8wGAAAAAAAAABgeYTYAAAAAAAAAwPAIswEAAAAAAAAAhkeYDQAAAAAAAAAwPMJsAAAAAAAAAIDhEWYDAAAAAAAAAAyPMBsAAAAAAAAAYHiE2QAAAAAAAAAAwyPMBgAAAAAAAAAYHmE2AAAAAAAAAMDwCLMBAAAAAAAAAIZHmA0AAAAAAAAAMDzCbAAAAAAAAACA4RFmAwAAAAAAAAAMjzAbAAAAAAAAAGB4hNkAAAAAAAAAAMMjzAYAAAAAAAAAGB5hNgAAAAAAAADA8AizAQAAAAAAAACGR5gNAAAAAAAAADA8wmwAAAAAAAAAgOERZgMAAAAAAAAADI8wGwAAAAAAAABgeITZAAAAAAAAAADDI8wGAAAAAAAAABgeYTYAAAAAAAAAwPAIswEAAAAAAAAAhkeYDQAAAAAAAAAwPMJsAAAAAAAAAIDhEWYDAAAAAAAAAAyPMBsAAAAAAAAAYHiE2QAAAAAAAAAAwyPMBgAAAAAAAAAYHmE2AAAAAAAAAMDwCLMBAAAAAAAAAIZHmA0AAAAAAAAAMDzCbAAAAAAAAACA4RFmAwAAAAAAAAAMjzAbAAAAAAAAAGB4hNkAAAAAAAAAAMMjzAYAAAAAAAAAGB5hNgAAAAAAAADA8AizAQAAAAAAAACGR5gNAAAAAAAAADA8wmwAAAAAAAAAgOERZgMAAAAAAAAADM/mYfaMGTNUpkwZubi4KCAgQFu3bs2w79KlS9WsWTMVLVpUBQoUUL169bR27dpcrBYAAAAAAAAAYAs2DbOjo6M1dOhQvfPOO9q/f7+Cg4PVqlUrxcfHp9t/y5YtatasmVavXq19+/bphRdeUNu2bbV///5crhwAAAAAAAAAkJtsGmZPmjRJ/fr1U//+/VW5cmVNmTJF3t7emjlzZrr9p0yZohEjRqh27doqX768PvroI5UvX17ff/99LlcOAAAAAAAAAMhNNguzb9++rX379ql58+ZW7c2bN9eOHTuydI3U1FRdu3ZNhQsXzrBPcnKyrl69avUCAAAAAAAAAOQtNguzL168qJSUFHl5eVm1e3l56dy5c1m6xsSJE3X9+nV17tw5wz7h4eHy8PCwvLy9vR+pbgAAAAAAAABA7rP5AyBNJpPVe7PZnKYtPQsXLlRYWJiio6NVrFixDPuNGjVKSUlJlldCQsIj1wwAAAAAAAAAyF0OtrpxkSJFZG9vn2YV9vnz59Os1v6n6Oho9evXT998842aNm2aaV9nZ2c5Ozs/cr0AAAAAAAAAANux2cpsJycnBQQEKCYmxqo9JiZGQUFBGZ63cOFC9enTRwsWLFCbNm0ed5kAAAAAAAAAAAOw2cpsSRo+fLh69uypwMBA1atXT1988YXi4+M1aNAgSfe2CDlz5ozmzZsn6V6Q3atXL02dOlV169a1rOp2dXWVh4eHzcYBAAAAAAAAAHi8bBpmd+nSRZcuXdK4ceOUmJgof39/rV69Wr6+vpKkxMRExcfHW/p//vnnunv3rl577TW99tprlvbevXsrKioqt8sHAAAAAAAAAOQSm4bZkhQSEqKQkJB0j/0zoN60adPjLwgAAAAAAAAAYDg22zMbAAAAAAAAAICsIswGAAAAAAAAABgeYTYAAAAAAAAAwPBsvmc2AAAAAAAAgLwjJSVFd+7csXUZeEI4OTnJzi5ra64JswEAAAAAAAA8kNls1rlz53TlyhVbl4IniJ2dncqUKSMnJ6cH9iXMBgAAAAAAAPBA94PsYsWKyc3NTSaTydYlIY9LTU3V2bNnlZiYKB8fnwfOKcJsAAAAAAAAAJlKSUmxBNmenp62LgdPkKJFi+rs2bO6e/euHB0dM+3LAyABAAAAAAAAZOr+Htlubm42rgRPmvvbi6SkpDywL2E2AAAAAAAAgCxhaxHktIeZU4TZAAAAAAAAAADDI8wGAAAAAAAAgCxq1KiRhg4dausynko8ABIAAAAAAABAtpUeuSrX7nVyfJss933Q9hW9e/dWVFTUQ9ewdOnSBz6oMKt27Nih4OBgNWvWTGvWrMmRaz7JCLMBAAAAAAAAPHESExMtf46OjtZ7772n3377zdLm6upq1f/OnTtZCqkLFy6cYzVGRkZqyJAhmjNnjuLj4+Xj45Nj135YWR2/LbHNCAAAAAAAAIAnTvHixS0vDw8PmUwmy/tbt26pYMGCWrx4sRo1aiQXFxd9/fXXunTpkrp166ZSpUrJzc1N1apV08KFC62u+89tRkqXLq2PPvpIffv2Vf78+eXj46MvvvjigfVdv35dixcv1quvvqoXX3wx3VXiK1asUGBgoFxcXFSkSBF17NjRciw5OVkjRoyQt7e3nJ2dVb58eUVEREiSoqKiVLBgQatrLV++3Gq1elhYmJ599llFRkbKz89Pzs7OMpvNWrNmjRo0aKCCBQvK09NTL774oo4dO2Z1rdOnT6tr164qXLiw3N3dFRgYqN27d+vkyZOys7PT3r17rfpPmzZNvr6+MpvND/xcMkOYDQAAAAAAAOCp9NZbbyk0NFRxcXFq0aKFbt26pYCAAK1cuVKHDh3SK6+8op49e2r37t2ZXmfixIkKDAzU/v37FRISoldffVX/+9//Mj0nOjpaFStWVMWKFdWjRw/NnTvXKuxdtWqVOnbsqDZt2mj//v1av369AgMDLcd79eqlRYsW6dNPP1VcXJxmzZqlfPnyPdT4jx49qsWLF2vJkiWKjY2VdC9kHz58uPbs2aP169fLzs5OL730klJTUyVJf/31lxo2bKizZ89qxYoVOnDggEaMGKHU1FSVLl1aTZs21dy5c63uM3fuXPXp0+eBW788CNuMAAAAAAAAAHgqDR061Gq1syS98cYblj8PGTJEa9as0TfffKPnnnsuw+u0bt1aISEhku4F5JMnT9amTZtUqVKlDM+JiIhQjx49JEktW7bUX3/9pfXr16tp06aSpA8//FBdu3bV2LFjLefUqFFDkvT7779r8eLFiomJsfT38/N7mKFLkm7fvq2vvvpKRYsWtbR16tQpTZ3FihXT4cOH5e/vrwULFujChQvas2ePZcuVcuXKWfr3799fgwYN0qRJk+Ts7KwDBw4oNjZWS5cufej6/omV2QAAAAAAAACeSn9f6SxJKSkp+vDDD1W9enV5enoqX758WrduneLj4zO9TvXq1S1/vr+dyfnz5zPs/9tvv+mnn35S165dJUkODg7q0qWLIiMjLX1iY2PVpEmTdM+PjY2Vvb29GjZs+MAxZsbX19cqyJakY8eOqXv37vLz81OBAgVUpkwZSbJ8BrGxsapZs2aGe4d36NBBDg4OWrZsmaR7+4K/8MILKl269CPVKrEyGwAAAAAAAMBTyt3d3er9xIkTNXnyZE2ZMkXVqlWTu7u7hg4dqtu3b2d6nX8+ONFkMlm25UhPRESE7t69q2eeecbSZjab5ejoqMuXL6tQoUJpHlD5d5kdkyQ7O7s0+1PfuXMnTb9/jl+S2rZtK29vb82ePVslS5ZUamqq/P39LZ/Bg+7t5OSknj17au7cuerYsaMWLFigKVOmZHpOVrEyGwAAAAAAAAAkbd26Ve3bt1ePHj1Uo0YN+fn56ciRIzl6j7t372revHmaOHGiYmNjLa8DBw7I19dX8+fPl3Rvtff69evTvUa1atWUmpqqzZs3p3u8aNGiunbtmq5fv25pu78ndmYuXbqkuLg4vfvuu2rSpIkqV66sy5cvW/WpXr26YmNj9eeff2Z4nf79++vHH3/UjBkzdOfOnTRbuWQXYTYAAAAAAAAA6N7ezzExMdqxY4fi4uI0cOBAnTt3LkfvsXLlSl2+fFn9+vWTv7+/1etf//qXIiIiJEljxozRwoULNWbMGMXFxengwYOaMGGCJKl06dLq3bu3+vbtq+XLl+vEiRPatGmTFi9eLEl67rnn5ObmprfffltHjx7VggULFBUV9cDaChUqJE9PT33xxRc6evSoNmzYoOHDh1v16datm4oXL64OHTpo+/btOn78uJYsWaKdO3da+lSuXFl169bVW2+9pW7duj1wNXdWsc0IAAAAAAAAgGw7Ob6NrUvIMaNHj9aJEyfUokULubm56ZVXXlGHDh2UlJSUY/eIiIhQ06ZN5eHhkeZYp06d9NFHH+nnn39Wo0aN9M033+j999/X+PHjVaBAAT3//POWvjNnztTbb7+tkJAQXbp0ST4+Pnr77bclSYULF9bXX3+tN998U1988YWaNm2qsLAwvfLKK5nWZmdnp0WLFik0NFT+/v6qWLGiPv30UzVq1MjSx8nJSevWrdPrr7+u1q1b6+7du6pSpYqmT59uda1+/fppx44d6tu37yN8WtYIswEAAAAAAAA80fr06aM+ffpY3pcuXTrNntLSvRB4+fLlmV5r06ZNVu9PnjyZpk9mW3p8//33GR6rVauWVV0dO3bMcIsOFxcXTZo0SZMmTUr3eIcOHdShQwertgEDBlj+HBYWprCwsDTnNW3aVIcPH7Zq++dn5evrq2+//TbDcUhSYmKi/P39Vbt27Uz7PQy2GQEAAAAAAAAA5Ii//vpLe/bs0bRp0xQaGpqj1ybMBgAAAAAAAADkiMGDB6tBgwZq2LBhjm4xIrHNCAAAAAAAAAAgh0RFRWXpYZPZwcpsAAAAAAAAAIDhEWYDAAAAAAAAAAyPMBsAAAAAAAAAYHiE2QAAAAAAAAAAwyPMBgAAAAAAAAAYHmE2AAAAAAAAAMDwCLMBAAAAAAAAAIbnYOsCAAAAAAAAAORhYR65eK+kLHc1mUyZHu/du7eioqKyVUbp0qU1dOhQDR06NEv9P/roI40ePVoffvihRo4cma17gpXZAAAAAAAAAJ5AiYmJlteUKVNUoEABq7apU6fmWi1z587ViBEjFBkZmWv3zMjt27dtXUK2EWYDAAAAAAAAeOIUL17c8vLw8JDJZLJq27JliwICAuTi4iI/Pz+NHTtWd+/etZwfFhYmHx8fOTs7q2TJkgoNDZUkNWrUSKdOndKwYcNkMpkeuAJ88+bNunnzpsaNG6fr169ry5YtVsdTU1P18ccfq1y5cnJ2dpaPj48+/PBDy/HTp0+ra9euKly4sNzd3RUYGKjdu3dLkvr06aMOHTpYXW/o0KFq1KiR5X2jRo00ePBgDR8+XEWKFFGzZs0kSZMmTVK1atXk7u4ub29vhYSE6K+//rK61vbt29WwYUO5ubmpUKFCatGihS5fvqx58+bJ09NTycnJVv07deqkXr16Zfp5PArCbAAAAAAAAABPlbVr16pHjx4KDQ3V4cOH9fnnnysqKsoSIn/77beaPHmyPv/8cx05ckTLly9XtWrVJElLly5VqVKlNG7cOMsq78xERESoW7ducnR0VLdu3RQREWF1fNSoUfr44481evRoHT58WAsWLJCXl5ck6a+//lLDhg119uxZrVixQgcOHNCIESOUmpr6UOP98ssv5eDgoO3bt+vzzz+XJNnZ2enTTz/VoUOH9OWXX2rDhg0aMWKE5ZzY2Fg1adJEVatW1c6dO7Vt2za1bdtWKSkp+ve//62UlBStWLHC0v/ixYtauXKlXn755Yeq7WGwZzYAAAAAAACAp8r9vat79+4tSfLz89P777+vESNGaMyYMYqPj1fx4sXVtGlTOTo6ysfHR3Xq1JEkFS5cWPb29sqfP7+KFy+e6X2uXr2qJUuWaMeOHZKkHj16qH79+po2bZoKFCiga9euaerUqfrss88stZQtW1YNGjSQJC1YsEAXLlzQnj17VLhwYUlSuXLlHnq85cqV04QJE6za/r7fd5kyZfT+++/r1Vdf1YwZMyRJEyZMUGBgoOW9JFWtWtXy5+7du2vu3Ln697//LUmaP3++SpUqZbUqPKexMhsAAAAAAADAU2Xfvn0aN26c8uXLZ3kNGDBAiYmJunHjhv7973/r5s2b8vPz04ABA7Rs2TKrLUiyasGCBfLz81ONGjUkSc8++6z8/Py0aNEiSVJcXJySk5PVpEmTdM+PjY1VzZo1LUF2dgUGBqZp27hxo5o1a6ZnnnlG+fPnV69evXTp0iVdv37dcu+M6pKkAQMGaN26dTpz5oyke/uC9+nT54HbrjwKwmwAAAAAAAAAT5XU1FSNHTtWsbGxltfBgwd15MgRubi4yNvbW7/99pumT58uV1dXhYSE6Pnnn9edO3ce6j6RkZH69ddf5eDgYHn9+uuvlq1GXF1dMz3/Qcft7OxkNput2tKr0d3d3er9qVOn1Lp1a/n7+2vJkiXat2+fpk+fbnX+g+5ds2ZN1ahRQ/PmzdPPP/+sgwcPqk+fPpme86jYZgQAAAAAAADAU6VWrVr67bffMt2yw9XVVe3atVO7du302muvqVKlSjp48KBq1aolJycnpaSkZHqPgwcPau/evdq0aZPVyuorV67o+eef16FDh1S+fHm5urpq/fr16t+/f5prVK9eXXPmzNGff/6Z7ursokWL6tChQ1ZtsbGxcnR0zLS2vXv36u7du5o4caLs7O6td168eHGae69fv15jx47N8Dr9+/fX5MmTdebMGTVt2lTe3t6Z3vdRsTIbAAAAAAAAwFPlvffe07x58xQWFqZff/1VcXFxio6O1rvvvitJioqKUkREhA4dOqTjx4/rq6++kqurq3x9fSVJpUuX1pYtW3TmzBldvHgx3XtERESoTp06ev755+Xv7295NWjQQPXq1VNERIRcXFz01ltvacSIEZo3b56OHTumXbt2WVZud+vWTcWLF1eHDh20fft2HT9+XEuWLNHOnTslSY0bN9bevXs1b948HTlyRGPGjEkTbqenbNmyunv3rqZNm2YZ36xZs6z6jBo1Snv27FFISIh++eUX/e9//9PMmTOtxvuf//xHZ86c0ezZs9W3b9+H/yIeEiuzAQAAAAAAAGRfWJKtK3hoLVq00MqVKzVu3DhNmDBBjo6OqlSpkmV1dMGCBTV+/HgNHz5cKSkpqlatmr7//nt5enpKksaNG6eBAweqbNmySk5OTrPVx+3bt/X111/rrbfeSvf+nTp1Unh4uD7++GONHj1aDg4Oeu+993T27FmVKFFCgwYNkiQ5OTlp3bp1ev3119W6dWvdvXtXVapUsWwJ0qJFC40ePVojRozQrVu31LdvX/Xq1UsHDx7MdPzPPvusJk2apI8//lijRo3S888/r/DwcPXq1cvSp0KFClq3bp3efvtt1alTR66urnruuefUrVs3S58CBQqoU6dOWrVqlTp06PBwX0I2mMz//KSfcFevXpWHh4eSkpJUoEABW5djUXrkKluXYOWkS3dbl2AtD/6lmNuYQ1nAPMqU0eaQZMB5xBx6IKPNI+ZQ3sMcygLmUaaMNockA84j5tADGW0eMYfyHuZQFhhoHhk1KzKaW7du6cSJEypTpoxcXFxsXQ4MpFmzZqpcubI+/fTTbJ3/MHOLldkAAAAAAAAAgIfy559/at26ddqwYYM+++yzXLknYTYAAAAAAAAA4KHUqlVLly9f1scff6yKFSvmyj0JswEAAAAAAAAAD+XkyZO5fk+7XL8jAAAAAAAAAAAPiTAbAAAAAAAAQJaYzWZbl4AnzMPMKcJsAAAAAAAAAJlydHSUJN24ccPGleBJc/v2bUmSvb39A/uyZzYAAAAAAACATNnb26tgwYI6f/68JMnNzU0mk8nGVSGvS01N1YULF+Tm5iYHhwdH1YTZAAAAAAAAAB6oePHikmQJtIGcYGdnJx8fnyz9coQwGwAAAAAAAMADmUwmlShRQsWKFdOdO3dsXQ6eEE5OTrKzy9pu2ITZAAAAAAAAALLM3t4+S/sbAznN5g+AnDFjhsqUKSMXFxcFBARo69atmfbfvHmzAgIC5OLiIj8/P82aNSuXKgUAAAAAAAAA2IpNw+zo6GgNHTpU77zzjvbv36/g4GC1atVK8fHx6fY/ceKEWrdureDgYO3fv19vv/22QkNDtWTJklyuHAAAAAAAAACQm2waZk+aNEn9+vVT//79VblyZU2ZMkXe3t6aOXNmuv1nzZolHx8fTZkyRZUrV1b//v3Vt29f/fe//83lygEAAAAAAAAAuclme2bfvn1b+/bt08iRI63amzdvrh07dqR7zs6dO9W8eXOrthYtWigiIkJ37tyRo6NjmnOSk5OVnJxseZ+UlCRJunr16qMOIUelJt+wdQlWrprMti7BmsG+LyNiDmUB8yhTRptDkgHnEXPogYw2j5hDeQ9zKAuYR5ky2hySDDiPmEMPZLR5xBzKe5hDWWCgeXQ/IzKbDfg5AbBiszD74sWLSklJkZeXl1W7l5eXzp07l+45586dS7f/3bt3dfHiRZUoUSLNOeHh4Ro7dmyadm9v70eo/snnYesC/mm84SrCAxjyG2Me5TmG+8aYQ3mO4b4x5lCeY8hvjHmU5xjuG2MO5TmG+8aYQ3mOIb8xA86ja9euycPDeHUB+H9sFmbfZzKZrN6bzeY0bQ/qn177faNGjdLw4cMt71NTU/Xnn3/K09Mz0/s8za5evSpvb28lJCSoQIECti4HeRBzCDmBeYRHxRzCo2IOIScwj/ComEN4VMyhBzObzbp27ZpKlixp61IAPIDNwuwiRYrI3t4+zSrs8+fPp1l9fV/x4sXT7e/g4CBPT890z3F2dpazs7NVW8GCBbNf+FOkQIEC/A8dHglzCDmBeYRHxRzCo2IOIScwj/ComEN4VMyhzLEiG8gbbPYASCcnJwUEBCgmJsaqPSYmRkFBQemeU69evTT9161bp8DAwHT3ywYAAAAAAAAAPBlsFmZL0vDhwzVnzhxFRkYqLi5Ow4YNU3x8vAYNGiTp3hYhvXr1svQfNGiQTp06peHDhysuLk6RkZGKiIjQG2+8YashAAAAAAAAAABygU33zO7SpYsuXbqkcePGKTExUf7+/lq9erV8fX0lSYmJiYqPj7f0L1OmjFavXq1hw4Zp+vTpKlmypD799FN16tTJVkN4Ijk7O2vMmDFptmcBsoo5hJzAPMKjYg7hUTGHkBOYR3hUzCE8KuYQgCeJyXz/CYoAAAAAAAAAABiUTbcZAQAAAAAAAAAgKwizAQAAAAAAAACGR5gNAAAAAAAAADA8wmwAAAAAAAAAgOERZgMAAAAZ4FnpAAAAgHE42LoA2N7p06c1c+ZM7dixQ+fOnZPJZJKXl5eCgoI0aNAgeXt727pEAAAAm3B2dtaBAwdUuXJlW5cCAAAAPPVMZpabPNW2bdumVq1aydvbW82bN5eXl5fMZrPOnz+vmJgYJSQk6IcfflD9+vVtXSrysISEBI0ZM0aRkZG2LgUGdvPmTe3bt0+FCxdWlSpVrI7dunVLixcvVq9evWxUHfKCuLg47dq1S/Xq1VOlSpX0v//9T1OnTlVycrJ69Oihxo0b27pEGNjw4cPTbZ86dap69OghT09PSdKkSZNysyzkcZcvX9aXX36pI0eOqESJEurduzcLRZCp/fv3q2DBgipTpowk6euvv9bMmTMVHx8vX19fDR48WF27drVxlTCyIUOGqHPnzgoODrZ1KQDwWBBmP+Vq166tBg0aaPLkyekeHzZsmLZt26Y9e/bkcmV4khw4cEC1atVSSkqKrUuBQf3+++9q3ry54uPjZTKZFBwcrIULF6pEiRKSpD/++EMlS5ZkDiFDa9asUfv27ZUvXz7duHFDy5YtU69evVSjRg2ZzWZt3rxZa9euJdBGhuzs7FSjRg0VLFjQqn3z5s0KDAyUu7u7TCaTNmzYYJsCkSeULFlSBw8elKenp06cOKGgoCBJUrVq1RQXF6dr165p165dqlSpko0rhVHVqlVLEydO1AsvvKA5c+YoNDRUAwYMUOXKlfXbb79pzpw5mjp1qvr27WvrUmFQdnZ2MplMKlu2rPr166fevXurePHiti4LAHIMYfZTztXVVbGxsapYsWK6x//3v/+pZs2aunnzZi5XhrxkxYoVmR4/fvy4Xn/9dYJIZOill17S3bt3NXfuXF25ckXDhw/XoUOHtGnTJvn4+BBm44GCgoLUuHFjffDBB1q0aJFCQkL06quv6sMPP5QkvfPOO9qzZ4/WrVtn40phVOHh4Zo9e7bmzJlj9UsPR0dHHThwIM2/GAHSY2dnp3PnzqlYsWLq1q2bzp07p1WrVsnNzU3Jycn617/+JRcXF33zzTe2LhUG5e7urri4OPn4+KhWrVoaNGiQXnnlFcvxBQsW6MMPP9Svv/5qwyphZHZ2doqJidH333+v+fPnKykpSa1atdKAAQPUunVr2dnx6DQAeRth9lPOz89Po0eP1ssvv5zu8blz5+r999/X8ePHc7ky5CX3f/uf2V8nJpOJIBIZ8vLy0o8//qhq1apZ2l577TWtXLlSGzdulLu7O2E2MuXh4aF9+/apXLlySk1NlbOzs3bv3q1atWpJkg4dOqSmTZvq3LlzNq4URrZnzx716NFDbdu2VXh4uBwdHQmz8VD+Hmb7+fml+eXI7t279a9//UsJCQk2rBJGVqRIEa1du1YBAQHy8vLSunXrVKNGDcvxY8eOqVq1arpx44YNq4SR/f3voTt37mjZsmWKjIzUjz/+KC8vL/Xp00cvv/yyypUrZ+tSASBb+JXcU+6NN97QoEGDNHjwYH333XfatWuXdu/ere+++06DBw/Wq6++qhEjRti6TBhciRIltGTJEqWmpqb7+vnnn21dIgzu5s2bcnCwfibx9OnT1a5dOzVs2FC///67jSpDXmRnZycXFxer7SLy58+vpKQk2xWFPKF27drat2+fLly4oMDAQB08eFAmk8nWZSGPuT9nkpOT5eXlZXXMy8tLFy5csEVZyCNatWqlmTNnSpIaNmyob7/91ur44sWLCSGRZY6OjurcubPWrFmj48ePa8CAAZo/f36G/zIbAPIChwd3wZMsJCREnp6emjx5sj7//HPLqkd7e3sFBARo3rx56ty5s42rhNEFBATo559/VocOHdI9/qBV20ClSpW0d+9eVa5c2ap92rRpMpvNateunY0qQ15RunRpHT161PJ/8Hfu3CkfHx/L8YSEBMse7EBm8uXLpy+//FKLFi1Ss2bN+BcheGhNmjSRg4ODrl69qt9//11Vq1a1HIuPj1eRIkVsWB2M7uOPP1b9+vXVsGFDBQYGauLEidq0aZNlz+xdu3Zp2bJlti4TeZCPj4/CwsI0ZswY/fjjj7YuBwCyjTAb6tKli7p06aI7d+7o4sWLku798zZHR0cbV4a84s0339T169czPF6uXDlt3LgxFytCXvPSSy9p4cKF6tmzZ5pjn332mVJTUzVr1iwbVIa84tVXX7UKHf39/a2O//DDDzz8EQ+la9euatCggfbt2ydfX19bl4M8YsyYMVbv3dzcrN5///33Cg4Ozs2SkMeULFlS+/fv1/jx4/X999/LbDbrp59+UkJCgurXr6/t27crMDDQ1mXCwHx9fWVvb5/hcZPJpGbNmuViRQCQs9gzGwAAAAAAAABgeOyZDQAAAAAAAAAwPMJsAAAAAAAAAIDhEWYDAAAAAAAAAAyPMBsAAAAWmzZtkslk0pUrV7J8TunSpTVlypTHVhMAAAAASITZAAAAeUqfPn1kMpk0aNCgNMdCQkJkMpnUp0+f3C8MAAAAAB4zwmwAAIA8xtvbW4sWLdLNmzctbbdu3dLChQvl4+Njw8oAAAAA4PEhzAYAAMhjatWqJR8fHy1dutTStnTpUnl7e6tmzZqWtuTkZIWGhqpYsWJycXFRgwYNtGfPHqtrrV69WhUqVJCrq6teeOEFnTx5Ms39duzYoeeff16urq7y9vZWaGiorl+//tjGBwAAAADpIcwGAADIg15++WXNnTvX8j4yMlJ9+/a16jNixAgtWbJEX375pX7++WeVK1dOLVq00J9//ilJSkhIUMeOHdW6dWvFxsaqf//+GjlypNU1Dh48qBYtWqhjx4765ZdfFB0drW3btmnw4MGPf5AAAAAA8DeE2QAAAHlQz549tW3bNp08eVKnTp3S9u3b1aNHD8vx69eva+bMmfrkk0/UqlUrValSRbNnz5arq6siIiIkSTNnzpSfn58mT56sihUr6j//+U+a/bY/+eQTde/eXUOHDlX58uUVFBSkTz/9VPPmzdOtW7dyc8gAAAAAnnIOti4AAAAAD69IkSJq06aNvvzyS5nNZrVp00ZFihSxHD927Jju3Lmj+vXrW9ocHR1Vp04dxcXFSZLi4uJUt25dmUwmS5969epZ3Wffvn06evSo5s+fb2kzm81KTU3ViRMnVLly5cc1RAAAAACwQpgNAACQR/Xt29ey3cf06dOtjpnNZkmyCqrvt99vu98nM6mpqRo4cKBCQ0PTHONhkwAAAAByE9uMAAAA5FEtW7bU7du3dfv2bbVo0cLqWLly5eTk5KRt27ZZ2u7cuaO9e/daVlNXqVJFu3btsjrvn+9r1aqlX3/9VeXKlUvzcnJyekwjAwAAAIC0CLMBAADyKHt7e8XFxSkuLk729vZWx9zd3fXqq6/qzTff1Jo1a3T48GENGDBAN27cUL9+/SRJgwYN0rFjxzR8+HD99ttvWrBggaKioqyu89Zbb2nnzp167bXXFBsbqyNHjmjFihUaMmRIbg0TAAAAACQRZgMAAORpBQoUUIECBdI9Nn78eHXq1Ek9e/ZUrVq1dPToUa1du1aFChWSdG+bkCVLluj7779XjRo1NGvWLH300UdW16hevbo2b96sI0eOKDg4WDVr1tTo0aNVokSJxz42AAAAAPg7kzkrmyUCAAAAAAAAAGBDrMwGAAAAAAAAABgeYTYAAAAAAAAAwPAIswEAAAAAAAAAhkeYDQAAAAAAAAAwPMJsAAAAAAAAAIDhEWYDAAAAAAAAAAyPMBsAAAAAAAAAYHiE2QAAAAAAAAAAwyPMBgAAAAAAAAAYHmE2AAAAAAAAAMDwCLMBAAAAAAAAAIZHmA0AAAAAAAAAMLz/D1KvtAXVNi0yAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "result.plot(kind=\"bar\",figsize=(15,4))\n",
    "plt.title('Train&Test Scores of Classifiers')\n",
    "plt.xlabel('Model')\n",
    "plt.ylabel('Scores')\n",
    "plt.legend(loc=4 , bbox_to_anchor=(1.2, 0))\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23da4bc1",
   "metadata": {},
   "source": [
    "Result:\n",
    "Four types of movement applied which are common in rob0t farward classfied using machine learning algorithms like RandomForest , DecisionTree classification and Logistric Regression Algorithms. we obtained the maximum accuracy with Logistric Regression.we found out thatLogistric Regression Algorithms is most suitable for working with sequence data out of other three algorithms.\n",
    "The models started performing worst after we applied PCA on the resampled data. The reason behind this is, PCA reduces the complexity of the data. It creates components based on giving importance to variables with large variance and also the components which it creates are non collinear in nature which means it takes care of collinearity in large data set. PCA also improves the overall execution time and quality of the models and it is very beneficial when we are working with huge amount of variables.\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}