darwin-prediction-of-alzheimer-s-disease-0-98-auc.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "caa93aab",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "papermill": {
     "duration": 0.02105,
     "end_time": "2024-09-21T18:25:40.591106",
     "exception": false,
     "start_time": "2024-09-21T18:25:40.570056",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## 📊 Project Overview\n",
    "\n",
    "In this notebook, we aim to detect **Alzheimer's Disease (AD)** through handwriting analysis. \n",
    "\n",
    "\n",
    "# DARWIN Dataset Description\n",
    "\n",
    "The **DARWIN dataset** is a collection of handwriting data designed to aid in the classification of Alzheimer’s disease patients versus healthy individuals. It includes data from **174 participants** and aims to enhance existing machine-learning methodologies for predicting Alzheimer’s disease through handwriting analysis.\n",
    "\n",
    "## Dataset Details\n",
    "\n",
    "- **Creator**: Francesco Fontanella\n",
    "- **Source**: [UCI Machine Learning Repository - DARWIN Dataset](https://archive.ics.uci.edu/dataset/732/darwin)\n",
    "\n",
    "\n",
    "### Citation Requests/Acknowledgements\n",
    "\n",
    "1. N. D. Cilia, C. De Stefano, F. Fontanella, A. S. Di Freca, *An experimental protocol to support cognitive impairment diagnosis by using handwriting analysis*, Procedia Computer Science, 141 (2018), 466–471. [DOI: 10.1016/j.procs.2018.10.141](https://doi.org/10.1016/j.procs.2018.10.141)\n",
    "\n",
    "2. N. D. Cilia, G. De Gregorio, C. De Stefano, F. Fontanella, A. Marcelli, A. Parziale, *Diagnosing Alzheimer’s disease from online handwriting: A novel dataset and performance benchmarking*, Engineering Applications of Artificial Intelligence, Vol. 111 (2022), 104822. [DOI: 10.1016/j.engappai.2022.104822](https://doi.org/10.1016/j.engappai.2022.104822)\n",
    "\n",
    "## Data Overview\n",
    "\n",
    "### `data.csv` \n",
    "\n",
    "- **Format**: CSV\n",
    "- **Rows**: 174 samples\n",
    "- **Columns**:\n",
    "  - **ID**: Unique identifier for each participant.\n",
    "  - **Features**: 450 columns representing handwriting features extracted from 25 tasks (18 features per task).\n",
    "  - **Class**: The target variable indicating if the participant is a 'P' (Patient) or 'H' (Healthy).\n",
    "\n",
    "\n",
    "---\n",
    "\n",
    "The project is structured into **3 key parts**:\n",
    "\n",
    "### 1️⃣ **Exploratory Data Analysis (EDA)**\n",
    "We start with a thorough exploration of the dataset:\n",
    "- 📉 Understand feature distribution\n",
    "- 🔍 Detect missing or skewed data\n",
    "- 📊 Visualize key insights to guide further processing\n",
    "\n",
    "### 2️⃣ **Feature Engineering**\n",
    "In this section, we transform the dataset to enhance model training:\n",
    "- 🛠️ Logarithmic transformations for skewed features\n",
    "- ✂️ Selection of the **10 most important features** per task \n",
    "- 📐 **Dimensionality Reduction** with PCA \n",
    "\n",
    "### 3️⃣ **Model Selection and Evaluation**\n",
    "Finally, we apply machine learning models:\n",
    "- 🧪 **Models**: CatBoost, SVM, Logistic Regression, and a custom Neural Network\n",
    "- 🔍 **Grid Search** for hyperparameter tuning\n",
    "- 📊 **Performance Metrics**: ROC AUC Score for each model, visualized and compared\n",
    "\n",
    "---\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cea92def",
   "metadata": {
    "papermill": {
     "duration": 0.019257,
     "end_time": "2024-09-21T18:25:40.630240",
     "exception": false,
     "start_time": "2024-09-21T18:25:40.610983",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# PART I : EDA "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4cd13904",
   "metadata": {
    "papermill": {
     "duration": 0.018132,
     "end_time": "2024-09-21T18:25:40.667503",
     "exception": false,
     "start_time": "2024-09-21T18:25:40.649371",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Importing Libraries and Loading Data\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c915a91a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:40.705803Z",
     "iopub.status.busy": "2024-09-21T18:25:40.705470Z",
     "iopub.status.idle": "2024-09-21T18:25:48.245736Z",
     "shell.execute_reply": "2024-09-21T18:25:48.244975Z"
    },
    "papermill": {
     "duration": 7.562313,
     "end_time": "2024-09-21T18:25:48.248198",
     "exception": false,
     "start_time": "2024-09-21T18:25:40.685885",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "import pandas as pd \n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from collections import defaultdict\n",
    "from scipy.stats import skew, kurtosis\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "from sklearn.model_selection import train_test_split, LeaveOneOut, GridSearchCV\n",
    "from sklearn.metrics import roc_auc_score, roc_curve, auc , classification_report, confusion_matrix\n",
    "from ipywidgets import interact, interactive, widgets, interact_manual, Dropdown\n",
    "\n",
    "from catboost import CatBoostClassifier\n",
    "\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import DataLoader, TensorDataset\n",
    "\n",
    "import re\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a071f744",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:48.288092Z",
     "iopub.status.busy": "2024-09-21T18:25:48.287624Z",
     "iopub.status.idle": "2024-09-21T18:25:48.389078Z",
     "shell.execute_reply": "2024-09-21T18:25:48.388113Z"
    },
    "papermill": {
     "duration": 0.122993,
     "end_time": "2024-09-21T18:25:48.391041",
     "exception": false,
     "start_time": "2024-09-21T18:25:48.268048",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of the data: (174, 452)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>air_time1</th>\n",
       "      <th>disp_index1</th>\n",
       "      <th>gmrt_in_air1</th>\n",
       "      <th>gmrt_on_paper1</th>\n",
       "      <th>max_x_extension1</th>\n",
       "      <th>max_y_extension1</th>\n",
       "      <th>mean_acc_in_air1</th>\n",
       "      <th>mean_acc_on_paper1</th>\n",
       "      <th>mean_gmrt1</th>\n",
       "      <th>...</th>\n",
       "      <th>mean_jerk_in_air25</th>\n",
       "      <th>mean_jerk_on_paper25</th>\n",
       "      <th>mean_speed_in_air25</th>\n",
       "      <th>mean_speed_on_paper25</th>\n",
       "      <th>num_of_pendown25</th>\n",
       "      <th>paper_time25</th>\n",
       "      <th>pressure_mean25</th>\n",
       "      <th>pressure_var25</th>\n",
       "      <th>total_time25</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>id_1</td>\n",
       "      <td>5160</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>120.804174</td>\n",
       "      <td>86.853334</td>\n",
       "      <td>957</td>\n",
       "      <td>6601</td>\n",
       "      <td>0.361800</td>\n",
       "      <td>0.217459</td>\n",
       "      <td>103.828754</td>\n",
       "      <td>...</td>\n",
       "      <td>0.141434</td>\n",
       "      <td>0.024471</td>\n",
       "      <td>5.596487</td>\n",
       "      <td>3.184589</td>\n",
       "      <td>71</td>\n",
       "      <td>40120</td>\n",
       "      <td>1749.278166</td>\n",
       "      <td>296102.7676</td>\n",
       "      <td>144605</td>\n",
       "      <td>P</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>id_2</td>\n",
       "      <td>51980</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>115.318238</td>\n",
       "      <td>83.448681</td>\n",
       "      <td>1694</td>\n",
       "      <td>6998</td>\n",
       "      <td>0.272513</td>\n",
       "      <td>0.144880</td>\n",
       "      <td>99.383459</td>\n",
       "      <td>...</td>\n",
       "      <td>0.049663</td>\n",
       "      <td>0.018368</td>\n",
       "      <td>1.665973</td>\n",
       "      <td>0.950249</td>\n",
       "      <td>129</td>\n",
       "      <td>126700</td>\n",
       "      <td>1504.768272</td>\n",
       "      <td>278744.2850</td>\n",
       "      <td>298640</td>\n",
       "      <td>P</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>id_3</td>\n",
       "      <td>2600</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>229.933997</td>\n",
       "      <td>172.761858</td>\n",
       "      <td>2333</td>\n",
       "      <td>5802</td>\n",
       "      <td>0.387020</td>\n",
       "      <td>0.181342</td>\n",
       "      <td>201.347928</td>\n",
       "      <td>...</td>\n",
       "      <td>0.178194</td>\n",
       "      <td>0.017174</td>\n",
       "      <td>4.000781</td>\n",
       "      <td>2.392521</td>\n",
       "      <td>74</td>\n",
       "      <td>45480</td>\n",
       "      <td>1431.443492</td>\n",
       "      <td>144411.7055</td>\n",
       "      <td>79025</td>\n",
       "      <td>P</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>id_4</td>\n",
       "      <td>2130</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>369.403342</td>\n",
       "      <td>183.193104</td>\n",
       "      <td>1756</td>\n",
       "      <td>8159</td>\n",
       "      <td>0.556879</td>\n",
       "      <td>0.164502</td>\n",
       "      <td>276.298223</td>\n",
       "      <td>...</td>\n",
       "      <td>0.113905</td>\n",
       "      <td>0.019860</td>\n",
       "      <td>4.206746</td>\n",
       "      <td>1.613522</td>\n",
       "      <td>123</td>\n",
       "      <td>67945</td>\n",
       "      <td>1465.843329</td>\n",
       "      <td>230184.7154</td>\n",
       "      <td>181220</td>\n",
       "      <td>P</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>id_5</td>\n",
       "      <td>2310</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>257.997131</td>\n",
       "      <td>111.275889</td>\n",
       "      <td>987</td>\n",
       "      <td>4732</td>\n",
       "      <td>0.266077</td>\n",
       "      <td>0.145104</td>\n",
       "      <td>184.636510</td>\n",
       "      <td>...</td>\n",
       "      <td>0.121782</td>\n",
       "      <td>0.020872</td>\n",
       "      <td>3.319036</td>\n",
       "      <td>1.680629</td>\n",
       "      <td>92</td>\n",
       "      <td>37285</td>\n",
       "      <td>1841.702561</td>\n",
       "      <td>158290.0255</td>\n",
       "      <td>72575</td>\n",
       "      <td>P</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 452 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     ID  air_time1  disp_index1  gmrt_in_air1  gmrt_on_paper1  \\\n",
       "0  id_1       5160     0.000013    120.804174       86.853334   \n",
       "1  id_2      51980     0.000016    115.318238       83.448681   \n",
       "2  id_3       2600     0.000010    229.933997      172.761858   \n",
       "3  id_4       2130     0.000010    369.403342      183.193104   \n",
       "4  id_5       2310     0.000007    257.997131      111.275889   \n",
       "\n",
       "   max_x_extension1  max_y_extension1  mean_acc_in_air1  mean_acc_on_paper1  \\\n",
       "0               957              6601          0.361800            0.217459   \n",
       "1              1694              6998          0.272513            0.144880   \n",
       "2              2333              5802          0.387020            0.181342   \n",
       "3              1756              8159          0.556879            0.164502   \n",
       "4               987              4732          0.266077            0.145104   \n",
       "\n",
       "   mean_gmrt1  ...  mean_jerk_in_air25  mean_jerk_on_paper25  \\\n",
       "0  103.828754  ...            0.141434              0.024471   \n",
       "1   99.383459  ...            0.049663              0.018368   \n",
       "2  201.347928  ...            0.178194              0.017174   \n",
       "3  276.298223  ...            0.113905              0.019860   \n",
       "4  184.636510  ...            0.121782              0.020872   \n",
       "\n",
       "   mean_speed_in_air25  mean_speed_on_paper25  num_of_pendown25  paper_time25  \\\n",
       "0             5.596487               3.184589                71         40120   \n",
       "1             1.665973               0.950249               129        126700   \n",
       "2             4.000781               2.392521                74         45480   \n",
       "3             4.206746               1.613522               123         67945   \n",
       "4             3.319036               1.680629                92         37285   \n",
       "\n",
       "   pressure_mean25  pressure_var25  total_time25  class  \n",
       "0      1749.278166     296102.7676        144605      P  \n",
       "1      1504.768272     278744.2850        298640      P  \n",
       "2      1431.443492     144411.7055         79025      P  \n",
       "3      1465.843329     230184.7154        181220      P  \n",
       "4      1841.702561     158290.0255         72575      P  \n",
       "\n",
       "[5 rows x 452 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('/kaggle/input/handwriting-data-to-detect-alzheimers-disease/data.csv')\n",
    "\n",
    "print(\"Shape of the data:\", data.shape)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4c97963",
   "metadata": {
    "papermill": {
     "duration": 0.018791,
     "end_time": "2024-09-21T18:25:48.428849",
     "exception": false,
     "start_time": "2024-09-21T18:25:48.410058",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Data Overview\n",
    "We'll explore the basic structure of the dataset, including data types, missing values, and duplicates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "04b17624",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:48.468053Z",
     "iopub.status.busy": "2024-09-21T18:25:48.467710Z",
     "iopub.status.idle": "2024-09-21T18:25:48.507612Z",
     "shell.execute_reply": "2024-09-21T18:25:48.506475Z"
    },
    "papermill": {
     "duration": 0.062081,
     "end_time": "2024-09-21T18:25:48.509759",
     "exception": false,
     "start_time": "2024-09-21T18:25:48.447678",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 174 entries, 0 to 173\n",
      "Columns: 452 entries, ID to class\n",
      "dtypes: float64(300), int64(150), object(2)\n",
      "memory usage: 614.6+ KB\n",
      "===\n",
      "Number of numeric variables: 450\n",
      "Number of categorical variables: 2\n"
     ]
    }
   ],
   "source": [
    "data.info()\n",
    "\n",
    "# Corrected dtypes to count numeric and categorical variables\n",
    "numeric_vars = data.select_dtypes(include=['int64', 'float64']).columns\n",
    "categorical_vars = data.select_dtypes(include=['object', 'category']).columns\n",
    "print(\"===\")\n",
    "print(f\"Number of numeric variables: {len(numeric_vars)}\")\n",
    "print(f\"Number of categorical variables: {len(categorical_vars)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2b0e70d7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:48.551536Z",
     "iopub.status.busy": "2024-09-21T18:25:48.550976Z",
     "iopub.status.idle": "2024-09-21T18:25:48.559121Z",
     "shell.execute_reply": "2024-09-21T18:25:48.558345Z"
    },
    "papermill": {
     "duration": 0.03076,
     "end_time": "2024-09-21T18:25:48.561153",
     "exception": false,
     "start_time": "2024-09-21T18:25:48.530393",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Clean the 'class' column by removing any leading/trailing spaces and converting to uppercase\n",
    "data['class'] = data['class'].str.strip().str.upper()\n",
    "\n",
    "# Transform the Class column: 'P' -> 1 and 'H' -> 0\n",
    "data['class'] = data['class'].map({'P': 1, 'H': 0})\n",
    "\n",
    "# Drop the 'ID' column\n",
    "data = data.drop(columns=['ID'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bc0caa49",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:48.603323Z",
     "iopub.status.busy": "2024-09-21T18:25:48.602485Z",
     "iopub.status.idle": "2024-09-21T18:25:48.610299Z",
     "shell.execute_reply": "2024-09-21T18:25:48.609255Z"
    },
    "papermill": {
     "duration": 0.029932,
     "end_time": "2024-09-21T18:25:48.612307",
     "exception": false,
     "start_time": "2024-09-21T18:25:48.582375",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "===\n",
      "Number of numeric variables: 451\n",
      "Number of categorical variables: 0\n"
     ]
    }
   ],
   "source": [
    "numeric_vars = data.select_dtypes(include=['int64', 'float64']).columns\n",
    "categorical_vars = data.select_dtypes(include=['object', 'category']).columns\n",
    "print(\"===\")\n",
    "print(f\"Number of numeric variables: {len(numeric_vars)}\")\n",
    "print(f\"Number of categorical variables: {len(categorical_vars)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f3d87c04",
   "metadata": {
    "papermill": {
     "duration": 0.018826,
     "end_time": "2024-09-21T18:25:48.650663",
     "exception": false,
     "start_time": "2024-09-21T18:25:48.631837",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Summary Statistics\n",
    "We display descriptive statistics for numerical variables and investigate the distribution of values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f8713611",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:48.690175Z",
     "iopub.status.busy": "2024-09-21T18:25:48.689466Z",
     "iopub.status.idle": "2024-09-21T18:25:49.381914Z",
     "shell.execute_reply": "2024-09-21T18:25:49.380961Z"
    },
    "papermill": {
     "duration": 0.714461,
     "end_time": "2024-09-21T18:25:49.384042",
     "exception": false,
     "start_time": "2024-09-21T18:25:48.669581",
     "status": "completed"
    },
    "tags": []
   },
   "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>air_time1</th>\n",
       "      <th>disp_index1</th>\n",
       "      <th>gmrt_in_air1</th>\n",
       "      <th>gmrt_on_paper1</th>\n",
       "      <th>max_x_extension1</th>\n",
       "      <th>max_y_extension1</th>\n",
       "      <th>mean_acc_in_air1</th>\n",
       "      <th>mean_acc_on_paper1</th>\n",
       "      <th>mean_gmrt1</th>\n",
       "      <th>mean_jerk_in_air1</th>\n",
       "      <th>...</th>\n",
       "      <th>mean_jerk_in_air25</th>\n",
       "      <th>mean_jerk_on_paper25</th>\n",
       "      <th>mean_speed_in_air25</th>\n",
       "      <th>mean_speed_on_paper25</th>\n",
       "      <th>num_of_pendown25</th>\n",
       "      <th>paper_time25</th>\n",
       "      <th>pressure_mean25</th>\n",
       "      <th>pressure_var25</th>\n",
       "      <th>total_time25</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>174.000000</td>\n",
       "      <td>1.740000e+02</td>\n",
       "      <td>174.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5664.166667</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>297.666685</td>\n",
       "      <td>200.504413</td>\n",
       "      <td>1977.965517</td>\n",
       "      <td>7323.896552</td>\n",
       "      <td>0.416374</td>\n",
       "      <td>0.179823</td>\n",
       "      <td>249.085549</td>\n",
       "      <td>0.067556</td>\n",
       "      <td>...</td>\n",
       "      <td>0.148286</td>\n",
       "      <td>0.019934</td>\n",
       "      <td>4.472643</td>\n",
       "      <td>2.871613</td>\n",
       "      <td>85.839080</td>\n",
       "      <td>43109.712644</td>\n",
       "      <td>1629.585962</td>\n",
       "      <td>163061.767360</td>\n",
       "      <td>1.642033e+05</td>\n",
       "      <td>0.511494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>12653.772746</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>183.943181</td>\n",
       "      <td>111.629546</td>\n",
       "      <td>1648.306365</td>\n",
       "      <td>2188.290512</td>\n",
       "      <td>0.381837</td>\n",
       "      <td>0.064693</td>\n",
       "      <td>132.698462</td>\n",
       "      <td>0.074776</td>\n",
       "      <td>...</td>\n",
       "      <td>0.062207</td>\n",
       "      <td>0.002388</td>\n",
       "      <td>1.501411</td>\n",
       "      <td>0.852809</td>\n",
       "      <td>27.485518</td>\n",
       "      <td>19092.024337</td>\n",
       "      <td>324.142316</td>\n",
       "      <td>56845.610814</td>\n",
       "      <td>4.969397e+05</td>\n",
       "      <td>0.501310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>65.000000</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>28.734515</td>\n",
       "      <td>29.935835</td>\n",
       "      <td>754.000000</td>\n",
       "      <td>561.000000</td>\n",
       "      <td>0.067748</td>\n",
       "      <td>0.096631</td>\n",
       "      <td>41.199445</td>\n",
       "      <td>0.011861</td>\n",
       "      <td>...</td>\n",
       "      <td>0.030169</td>\n",
       "      <td>0.014987</td>\n",
       "      <td>1.323565</td>\n",
       "      <td>0.950249</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>15930.000000</td>\n",
       "      <td>474.049462</td>\n",
       "      <td>26984.926660</td>\n",
       "      <td>2.998000e+04</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1697.500000</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>174.153023</td>\n",
       "      <td>136.524742</td>\n",
       "      <td>1362.500000</td>\n",
       "      <td>6124.000000</td>\n",
       "      <td>0.218209</td>\n",
       "      <td>0.146647</td>\n",
       "      <td>161.136182</td>\n",
       "      <td>0.029523</td>\n",
       "      <td>...</td>\n",
       "      <td>0.107732</td>\n",
       "      <td>0.018301</td>\n",
       "      <td>3.485934</td>\n",
       "      <td>2.401199</td>\n",
       "      <td>66.000000</td>\n",
       "      <td>32803.750000</td>\n",
       "      <td>1499.112088</td>\n",
       "      <td>120099.046800</td>\n",
       "      <td>5.917500e+04</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2890.000000</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>255.791452</td>\n",
       "      <td>176.494494</td>\n",
       "      <td>1681.000000</td>\n",
       "      <td>6975.500000</td>\n",
       "      <td>0.275184</td>\n",
       "      <td>0.163659</td>\n",
       "      <td>224.445268</td>\n",
       "      <td>0.039233</td>\n",
       "      <td>...</td>\n",
       "      <td>0.140483</td>\n",
       "      <td>0.019488</td>\n",
       "      <td>4.510578</td>\n",
       "      <td>2.830672</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>37312.500000</td>\n",
       "      <td>1729.385010</td>\n",
       "      <td>158236.771800</td>\n",
       "      <td>7.611500e+04</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4931.250000</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>358.917885</td>\n",
       "      <td>234.052560</td>\n",
       "      <td>2082.750000</td>\n",
       "      <td>8298.500000</td>\n",
       "      <td>0.442706</td>\n",
       "      <td>0.188879</td>\n",
       "      <td>294.392298</td>\n",
       "      <td>0.071057</td>\n",
       "      <td>...</td>\n",
       "      <td>0.199168</td>\n",
       "      <td>0.021134</td>\n",
       "      <td>5.212794</td>\n",
       "      <td>3.335828</td>\n",
       "      <td>101.500000</td>\n",
       "      <td>46533.750000</td>\n",
       "      <td>1865.626974</td>\n",
       "      <td>200921.078475</td>\n",
       "      <td>1.275425e+05</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>109965.000000</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>1168.328276</td>\n",
       "      <td>865.210522</td>\n",
       "      <td>18602.000000</td>\n",
       "      <td>15783.000000</td>\n",
       "      <td>2.772566</td>\n",
       "      <td>0.627350</td>\n",
       "      <td>836.784702</td>\n",
       "      <td>0.543199</td>\n",
       "      <td>...</td>\n",
       "      <td>0.375078</td>\n",
       "      <td>0.029227</td>\n",
       "      <td>10.416715</td>\n",
       "      <td>5.602909</td>\n",
       "      <td>209.000000</td>\n",
       "      <td>139575.000000</td>\n",
       "      <td>1999.775983</td>\n",
       "      <td>352981.850000</td>\n",
       "      <td>5.704200e+06</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 451 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           air_time1  disp_index1  gmrt_in_air1  gmrt_on_paper1  \\\n",
       "count     174.000000   174.000000    174.000000      174.000000   \n",
       "mean     5664.166667     0.000010    297.666685      200.504413   \n",
       "std     12653.772746     0.000003    183.943181      111.629546   \n",
       "min        65.000000     0.000002     28.734515       29.935835   \n",
       "25%      1697.500000     0.000008    174.153023      136.524742   \n",
       "50%      2890.000000     0.000009    255.791452      176.494494   \n",
       "75%      4931.250000     0.000011    358.917885      234.052560   \n",
       "max    109965.000000     0.000028   1168.328276      865.210522   \n",
       "\n",
       "       max_x_extension1  max_y_extension1  mean_acc_in_air1  \\\n",
       "count        174.000000        174.000000        174.000000   \n",
       "mean        1977.965517       7323.896552          0.416374   \n",
       "std         1648.306365       2188.290512          0.381837   \n",
       "min          754.000000        561.000000          0.067748   \n",
       "25%         1362.500000       6124.000000          0.218209   \n",
       "50%         1681.000000       6975.500000          0.275184   \n",
       "75%         2082.750000       8298.500000          0.442706   \n",
       "max        18602.000000      15783.000000          2.772566   \n",
       "\n",
       "       mean_acc_on_paper1  mean_gmrt1  mean_jerk_in_air1  ...  \\\n",
       "count          174.000000  174.000000         174.000000  ...   \n",
       "mean             0.179823  249.085549           0.067556  ...   \n",
       "std              0.064693  132.698462           0.074776  ...   \n",
       "min              0.096631   41.199445           0.011861  ...   \n",
       "25%              0.146647  161.136182           0.029523  ...   \n",
       "50%              0.163659  224.445268           0.039233  ...   \n",
       "75%              0.188879  294.392298           0.071057  ...   \n",
       "max              0.627350  836.784702           0.543199  ...   \n",
       "\n",
       "       mean_jerk_in_air25  mean_jerk_on_paper25  mean_speed_in_air25  \\\n",
       "count          174.000000            174.000000           174.000000   \n",
       "mean             0.148286              0.019934             4.472643   \n",
       "std              0.062207              0.002388             1.501411   \n",
       "min              0.030169              0.014987             1.323565   \n",
       "25%              0.107732              0.018301             3.485934   \n",
       "50%              0.140483              0.019488             4.510578   \n",
       "75%              0.199168              0.021134             5.212794   \n",
       "max              0.375078              0.029227            10.416715   \n",
       "\n",
       "       mean_speed_on_paper25  num_of_pendown25   paper_time25  \\\n",
       "count             174.000000        174.000000     174.000000   \n",
       "mean                2.871613         85.839080   43109.712644   \n",
       "std                 0.852809         27.485518   19092.024337   \n",
       "min                 0.950249         32.000000   15930.000000   \n",
       "25%                 2.401199         66.000000   32803.750000   \n",
       "50%                 2.830672         81.000000   37312.500000   \n",
       "75%                 3.335828        101.500000   46533.750000   \n",
       "max                 5.602909        209.000000  139575.000000   \n",
       "\n",
       "       pressure_mean25  pressure_var25  total_time25       class  \n",
       "count       174.000000      174.000000  1.740000e+02  174.000000  \n",
       "mean       1629.585962   163061.767360  1.642033e+05    0.511494  \n",
       "std         324.142316    56845.610814  4.969397e+05    0.501310  \n",
       "min         474.049462    26984.926660  2.998000e+04    0.000000  \n",
       "25%        1499.112088   120099.046800  5.917500e+04    0.000000  \n",
       "50%        1729.385010   158236.771800  7.611500e+04    1.000000  \n",
       "75%        1865.626974   200921.078475  1.275425e+05    1.000000  \n",
       "max        1999.775983   352981.850000  5.704200e+06    1.000000  \n",
       "\n",
       "[8 rows x 451 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3cf4ee0d",
   "metadata": {
    "papermill": {
     "duration": 0.01953,
     "end_time": "2024-09-21T18:25:49.424076",
     "exception": false,
     "start_time": "2024-09-21T18:25:49.404546",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Missing Values & Duplicate Data\n",
    "We check for missing and duplicated rows in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "8687c875",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:49.465593Z",
     "iopub.status.busy": "2024-09-21T18:25:49.465244Z",
     "iopub.status.idle": "2024-09-21T18:25:49.522900Z",
     "shell.execute_reply": "2024-09-21T18:25:49.521835Z"
    },
    "papermill": {
     "duration": 0.080881,
     "end_time": "2024-09-21T18:25:49.525025",
     "exception": false,
     "start_time": "2024-09-21T18:25:49.444144",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Missing Data  : Series([], dtype: int64)\n",
      "Number of duplicate rows: 0\n"
     ]
    }
   ],
   "source": [
    "missing_data = data.isnull().sum()\n",
    "print(\"Missing Data  :\", missing_data[missing_data > 0])\n",
    "duplicates = data[data.duplicated()]\n",
    "print(f\"Number of duplicate rows: {len(duplicates)}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43685284",
   "metadata": {
    "papermill": {
     "duration": 0.019916,
     "end_time": "2024-09-21T18:25:49.566064",
     "exception": false,
     "start_time": "2024-09-21T18:25:49.546148",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Target Variable Distribution\n",
    "We will now examine the distribution of the target variable (`class`), which indicates whether the participant is a 'Patient' (1) or 'Healthy' (0)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "cb317417",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:49.609476Z",
     "iopub.status.busy": "2024-09-21T18:25:49.609102Z",
     "iopub.status.idle": "2024-09-21T18:25:49.795468Z",
     "shell.execute_reply": "2024-09-21T18:25:49.794403Z"
    },
    "papermill": {
     "duration": 0.210539,
     "end_time": "2024-09-21T18:25:49.797526",
     "exception": false,
     "start_time": "2024-09-21T18:25:49.586987",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "class\n",
       "1    89\n",
       "0    85\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt.figure()\n",
    "sns.countplot(x='class', data=data)\n",
    "plt.title('Distribution of the Target Variable')\n",
    "plt.show()\n",
    "\n",
    "data['class'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7480615",
   "metadata": {
    "papermill": {
     "duration": 0.02127,
     "end_time": "2024-09-21T18:25:49.839753",
     "exception": false,
     "start_time": "2024-09-21T18:25:49.818483",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Skewness"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "17b96024",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:49.885003Z",
     "iopub.status.busy": "2024-09-21T18:25:49.884633Z",
     "iopub.status.idle": "2024-09-21T18:25:49.894980Z",
     "shell.execute_reply": "2024-09-21T18:25:49.893886Z"
    },
    "papermill": {
     "duration": 0.03575,
     "end_time": "2024-09-21T18:25:49.896938",
     "exception": false,
     "start_time": "2024-09-21T18:25:49.861188",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features with skewness > 0.5: 359 out 451\n"
     ]
    }
   ],
   "source": [
    "skewness_values = skew(data, axis=0)\n",
    "threshold = 0.5  \n",
    "skewed_features = [i for i, skew_val in enumerate(skewness_values) if skew_val > threshold]\n",
    "print(f\"Features with skewness > {threshold}: {len(skewed_features)} out {len(data.columns)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d07ba92b",
   "metadata": {
    "papermill": {
     "duration": 0.022245,
     "end_time": "2024-09-21T18:25:49.942129",
     "exception": false,
     "start_time": "2024-09-21T18:25:49.919884",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Interactive Visualization\n",
    "\n",
    "#### 1. Relationship between features\n",
    "This interactive scatter plot allows you to explore the relationships between two numeric variables while color-coding by the target variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d33d8b49",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:49.987998Z",
     "iopub.status.busy": "2024-09-21T18:25:49.987636Z",
     "iopub.status.idle": "2024-09-21T18:25:50.385998Z",
     "shell.execute_reply": "2024-09-21T18:25:50.385034Z"
    },
    "papermill": {
     "duration": 0.423965,
     "end_time": "2024-09-21T18:25:50.387949",
     "exception": false,
     "start_time": "2024-09-21T18:25:49.963984",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "43d8f2b3d75f4f7f85079dbd55bbebdf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(Dropdown(description='X-axis:', options=('air_time1', 'disp_index1', 'gmrt_in_air1', 'gm…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.interactive_scatter(x_col, y_col)>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numeric_columns = data.select_dtypes(include=['float64', 'int64']).columns\n",
    "\n",
    "def interactive_scatter(x_col, y_col):\n",
    "    plt.figure(figsize=(10, 6))\n",
    "    plt.scatter(data[x_col], data[y_col], c=data['class'], cmap='viridis', alpha=0.7)\n",
    "    plt.title(f'Scatter Plot of {x_col} vs. {y_col}')\n",
    "    plt.xlabel(x_col)\n",
    "    plt.ylabel(y_col)\n",
    "    plt.colorbar(label='Class')\n",
    "    plt.show()\n",
    "\n",
    "interact(interactive_scatter, \n",
    "         x_col=Dropdown(options=numeric_columns, description='X-axis:'), \n",
    "         y_col=Dropdown(options=numeric_columns, description='Y-axis:'))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f5e6149",
   "metadata": {
    "papermill": {
     "duration": 0.020797,
     "end_time": "2024-09-21T18:25:50.431276",
     "exception": false,
     "start_time": "2024-09-21T18:25:50.410479",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "#### 2. Distribution of each feature\n",
    "This plot shows the distribution of any selected numeric variable in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2d491f32",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:50.475931Z",
     "iopub.status.busy": "2024-09-21T18:25:50.475173Z",
     "iopub.status.idle": "2024-09-21T18:25:50.481180Z",
     "shell.execute_reply": "2024-09-21T18:25:50.480258Z"
    },
    "papermill": {
     "duration": 0.030505,
     "end_time": "2024-09-21T18:25:50.483167",
     "exception": false,
     "start_time": "2024-09-21T18:25:50.452662",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def interactive_distribution(column):\n",
    "    plt.figure(figsize=(10, 6))\n",
    "    sns.histplot(data[column], kde=True, color='skyblue', bins=30)\n",
    "    plt.title(f'Distribution of {column}')\n",
    "    plt.xlabel(column)\n",
    "    plt.ylabel('Frequency')\n",
    "    plt.grid(True)\n",
    "    plt.show()\n",
    "# Eliminate # to run the code(interactive plot) \n",
    "\n",
    "#interact(interactive_distribution, \n",
    "         #column=Dropdown(options=numeric_columns, description='Feature:'))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb8d7624",
   "metadata": {
    "papermill": {
     "duration": 0.020534,
     "end_time": "2024-09-21T18:25:50.525764",
     "exception": false,
     "start_time": "2024-09-21T18:25:50.505230",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Correlation Analysis\n",
    "We now identify and list features that have a high correlation (greater than 0.9)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "dadf4c36",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:50.568267Z",
     "iopub.status.busy": "2024-09-21T18:25:50.567946Z",
     "iopub.status.idle": "2024-09-21T18:25:54.105449Z",
     "shell.execute_reply": "2024-09-21T18:25:54.102687Z"
    },
    "papermill": {
     "duration": 3.562277,
     "end_time": "2024-09-21T18:25:54.108474",
     "exception": false,
     "start_time": "2024-09-21T18:25:50.546197",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Highly Correlated Features (Correlation > 0.9):\n",
      "total_time1 - air_time1: Correlation = 0.97\n",
      "mean_gmrt1 - gmrt_in_air1: Correlation = 0.94\n",
      "mean_speed_on_paper1 - gmrt_on_paper1: Correlation = 0.99\n",
      "mean_jerk_in_air1 - mean_acc_in_air1: Correlation = 0.99\n",
      "gmrt_in_air1 - mean_gmrt1: Correlation = 0.94\n",
      "mean_acc_in_air1 - mean_jerk_in_air1: Correlation = 0.99\n",
      "gmrt_on_paper1 - mean_speed_on_paper1: Correlation = 0.99\n",
      "air_time1 - total_time1: Correlation = 0.97\n",
      "total_time2 - air_time2: Correlation = 0.96\n",
      "mean_gmrt2 - gmrt_in_air2: Correlation = 0.93\n",
      "mean_jerk_in_air2 - mean_acc_in_air2: Correlation = 0.99\n",
      "gmrt_in_air2 - mean_gmrt2: Correlation = 0.93\n",
      "mean_acc_in_air2 - mean_jerk_in_air2: Correlation = 0.99\n",
      "air_time2 - total_time2: Correlation = 0.96\n",
      "mean_gmrt3 - gmrt_in_air3: Correlation = 0.93\n",
      "mean_jerk_in_air3 - mean_acc_in_air3: Correlation = 1.00\n",
      "gmrt_in_air3 - mean_gmrt3: Correlation = 0.93\n",
      "mean_acc_in_air3 - mean_jerk_in_air3: Correlation = 1.00\n",
      "total_time4 - air_time4: Correlation = 0.94\n",
      "mean_gmrt4 - gmrt_in_air4: Correlation = 0.94\n",
      "mean_speed_on_paper4 - gmrt_on_paper4: Correlation = 0.99\n",
      "mean_jerk_in_air4 - mean_acc_in_air4: Correlation = 1.00\n",
      "gmrt_in_air4 - mean_gmrt4: Correlation = 0.94\n",
      "mean_acc_in_air4 - mean_jerk_in_air4: Correlation = 1.00\n",
      "gmrt_on_paper4 - mean_speed_on_paper4: Correlation = 0.99\n",
      "air_time4 - total_time4: Correlation = 0.94\n",
      "mean_speed_on_paper5 - gmrt_on_paper5: Correlation = 0.99\n",
      "mean_jerk_in_air5 - mean_acc_in_air5: Correlation = 0.97\n",
      "mean_acc_in_air5 - mean_jerk_in_air5: Correlation = 0.97\n",
      "gmrt_on_paper5 - mean_speed_on_paper5: Correlation = 0.99\n",
      "total_time5 - paper_time5: Correlation = 0.91\n",
      "paper_time5 - total_time5: Correlation = 0.91\n",
      "total_time6 - air_time6: Correlation = 0.98\n",
      "mean_speed_in_air6 - gmrt_in_air6: Correlation = 0.95\n",
      "mean_speed_on_paper6 - gmrt_on_paper6: Correlation = 0.97\n",
      "mean_jerk_in_air6 - mean_acc_in_air6: Correlation = 1.00\n",
      "mean_acc_in_air6 - mean_jerk_in_air6: Correlation = 1.00\n",
      "gmrt_in_air6 - mean_speed_in_air6: Correlation = 0.95\n",
      "gmrt_on_paper6 - mean_speed_on_paper6: Correlation = 0.97\n",
      "air_time6 - total_time6: Correlation = 0.98\n",
      "total_time7 - air_time7: Correlation = 1.00\n",
      "mean_speed_in_air7 - gmrt_in_air7: Correlation = 0.99\n",
      "mean_gmrt7 - gmrt_on_paper7: Correlation = 0.93\n",
      "mean_speed_on_paper7 - gmrt_on_paper7: Correlation = 0.96\n",
      "mean_jerk_in_air7 - mean_acc_in_air7: Correlation = 1.00\n",
      "gmrt_on_paper7 - mean_gmrt7: Correlation = 0.93\n",
      "mean_acc_in_air7 - mean_jerk_in_air7: Correlation = 1.00\n",
      "gmrt_in_air7 - mean_speed_in_air7: Correlation = 0.99\n",
      "gmrt_on_paper7 - mean_speed_on_paper7: Correlation = 0.96\n",
      "air_time7 - total_time7: Correlation = 1.00\n",
      "total_time8 - air_time8: Correlation = 0.93\n",
      "mean_gmrt8 - gmrt_in_air8: Correlation = 0.97\n",
      "mean_speed_on_paper8 - gmrt_on_paper8: Correlation = 1.00\n",
      "mean_jerk_in_air8 - mean_acc_in_air8: Correlation = 1.00\n",
      "gmrt_in_air8 - mean_gmrt8: Correlation = 0.97\n",
      "mean_acc_in_air8 - mean_jerk_in_air8: Correlation = 1.00\n",
      "gmrt_on_paper8 - mean_speed_on_paper8: Correlation = 1.00\n",
      "air_time8 - total_time8: Correlation = 0.93\n",
      "total_time9 - air_time9: Correlation = 0.95\n",
      "mean_gmrt9 - gmrt_in_air9: Correlation = 0.98\n",
      "mean_speed_on_paper9 - gmrt_on_paper9: Correlation = 0.99\n",
      "mean_jerk_in_air9 - mean_acc_in_air9: Correlation = 1.00\n",
      "gmrt_in_air9 - mean_gmrt9: Correlation = 0.98\n",
      "mean_acc_in_air9 - mean_jerk_in_air9: Correlation = 1.00\n",
      "gmrt_on_paper9 - mean_speed_on_paper9: Correlation = 0.99\n",
      "air_time9 - total_time9: Correlation = 0.95\n",
      "total_time10 - air_time10: Correlation = 0.98\n",
      "mean_gmrt10 - gmrt_in_air10: Correlation = 0.91\n",
      "mean_speed_on_paper10 - gmrt_on_paper10: Correlation = 0.99\n",
      "mean_jerk_in_air10 - mean_acc_in_air10: Correlation = 0.99\n",
      "gmrt_in_air10 - mean_gmrt10: Correlation = 0.91\n",
      "mean_acc_in_air10 - mean_jerk_in_air10: Correlation = 0.99\n",
      "gmrt_on_paper10 - mean_speed_on_paper10: Correlation = 0.99\n",
      "pressure_mean11 - pressure_mean10: Correlation = 0.93\n",
      "air_time10 - total_time10: Correlation = 0.98\n",
      "total_time11 - air_time11: Correlation = 1.00\n",
      "mean_speed_on_paper11 - gmrt_on_paper11: Correlation = 0.98\n",
      "mean_jerk_in_air11 - mean_acc_in_air11: Correlation = 1.00\n",
      "mean_acc_in_air11 - mean_jerk_in_air11: Correlation = 1.00\n",
      "gmrt_on_paper11 - mean_speed_on_paper11: Correlation = 0.98\n",
      "pressure_mean10 - pressure_mean11: Correlation = 0.93\n",
      "air_time11 - total_time11: Correlation = 1.00\n",
      "total_time12 - air_time12: Correlation = 1.00\n",
      "mean_gmrt12 - gmrt_in_air12: Correlation = 0.98\n",
      "mean_speed_on_paper12 - gmrt_on_paper12: Correlation = 0.99\n",
      "mean_jerk_in_air12 - mean_acc_in_air12: Correlation = 0.99\n",
      "gmrt_in_air12 - mean_gmrt12: Correlation = 0.98\n",
      "mean_acc_in_air12 - mean_jerk_in_air12: Correlation = 0.99\n",
      "gmrt_on_paper12 - mean_speed_on_paper12: Correlation = 0.99\n",
      "air_time12 - total_time12: Correlation = 1.00\n",
      "total_time13 - air_time13: Correlation = 0.91\n",
      "mean_gmrt13 - gmrt_in_air13: Correlation = 0.96\n",
      "mean_speed_on_paper13 - gmrt_on_paper13: Correlation = 0.98\n",
      "mean_jerk_in_air13 - mean_acc_in_air13: Correlation = 1.00\n",
      "gmrt_in_air13 - mean_gmrt13: Correlation = 0.96\n",
      "mean_acc_in_air13 - mean_jerk_in_air13: Correlation = 1.00\n",
      "gmrt_on_paper13 - mean_speed_on_paper13: Correlation = 0.98\n",
      "air_time13 - total_time13: Correlation = 0.91\n",
      "total_time14 - air_time14: Correlation = 1.00\n",
      "mean_gmrt14 - gmrt_in_air14: Correlation = 0.96\n",
      "mean_speed_in_air14 - gmrt_in_air14: Correlation = 0.95\n",
      "mean_speed_on_paper14 - gmrt_on_paper14: Correlation = 0.97\n",
      "mean_jerk_in_air14 - mean_acc_in_air14: Correlation = 1.00\n",
      "gmrt_in_air14 - mean_gmrt14: Correlation = 0.96\n",
      "mean_speed_in_air14 - mean_gmrt14: Correlation = 0.93\n",
      "mean_acc_in_air14 - mean_jerk_in_air14: Correlation = 1.00\n",
      "gmrt_in_air14 - mean_speed_in_air14: Correlation = 0.95\n",
      "mean_gmrt14 - mean_speed_in_air14: Correlation = 0.93\n",
      "gmrt_on_paper14 - mean_speed_on_paper14: Correlation = 0.97\n",
      "air_time14 - total_time14: Correlation = 1.00\n",
      "total_time15 - air_time15: Correlation = 1.00\n",
      "mean_speed_in_air15 - gmrt_in_air15: Correlation = 0.95\n",
      "mean_speed_on_paper15 - gmrt_on_paper15: Correlation = 0.96\n",
      "mean_jerk_in_air15 - mean_acc_in_air15: Correlation = 1.00\n",
      "mean_acc_in_air15 - mean_jerk_in_air15: Correlation = 1.00\n",
      "gmrt_in_air15 - mean_speed_in_air15: Correlation = 0.95\n",
      "gmrt_on_paper15 - mean_speed_on_paper15: Correlation = 0.96\n",
      "air_time15 - total_time15: Correlation = 1.00\n",
      "total_time16 - air_time16: Correlation = 0.97\n",
      "max_x_extension16 - disp_index16: Correlation = 0.92\n",
      "mean_gmrt16 - gmrt_in_air16: Correlation = 0.90\n",
      "mean_speed_on_paper16 - gmrt_on_paper16: Correlation = 0.92\n",
      "disp_index16 - max_x_extension16: Correlation = 0.92\n",
      "mean_jerk_in_air16 - mean_acc_in_air16: Correlation = 1.00\n",
      "gmrt_in_air16 - mean_gmrt16: Correlation = 0.90\n",
      "mean_acc_in_air16 - mean_jerk_in_air16: Correlation = 1.00\n",
      "gmrt_on_paper16 - mean_speed_on_paper16: Correlation = 0.92\n",
      "air_time16 - total_time16: Correlation = 0.97\n",
      "total_time17 - air_time17: Correlation = 0.99\n",
      "mean_acc_in_air17 - gmrt_in_air17: Correlation = 0.92\n",
      "mean_gmrt17 - gmrt_in_air17: Correlation = 0.96\n",
      "mean_jerk_in_air17 - gmrt_in_air17: Correlation = 0.92\n",
      "mean_speed_in_air17 - gmrt_in_air17: Correlation = 0.99\n",
      "mean_speed_on_paper17 - gmrt_on_paper17: Correlation = 0.95\n",
      "gmrt_in_air17 - mean_acc_in_air17: Correlation = 0.92\n",
      "mean_jerk_in_air17 - mean_acc_in_air17: Correlation = 1.00\n",
      "mean_speed_in_air17 - mean_acc_in_air17: Correlation = 0.92\n",
      "gmrt_in_air17 - mean_gmrt17: Correlation = 0.96\n",
      "mean_speed_in_air17 - mean_gmrt17: Correlation = 0.97\n",
      "gmrt_in_air17 - mean_jerk_in_air17: Correlation = 0.92\n",
      "mean_acc_in_air17 - mean_jerk_in_air17: Correlation = 1.00\n",
      "mean_speed_in_air17 - mean_jerk_in_air17: Correlation = 0.92\n",
      "gmrt_in_air17 - mean_speed_in_air17: Correlation = 0.99\n",
      "mean_acc_in_air17 - mean_speed_in_air17: Correlation = 0.92\n",
      "mean_gmrt17 - mean_speed_in_air17: Correlation = 0.97\n",
      "mean_jerk_in_air17 - mean_speed_in_air17: Correlation = 0.92\n",
      "gmrt_on_paper17 - mean_speed_on_paper17: Correlation = 0.95\n",
      "air_time17 - total_time17: Correlation = 0.99\n",
      "num_of_pendown18 - air_time18: Correlation = 0.92\n",
      "total_time18 - air_time18: Correlation = 0.99\n",
      "mean_gmrt18 - gmrt_in_air18: Correlation = 0.94\n",
      "mean_speed_in_air18 - gmrt_in_air18: Correlation = 0.93\n",
      "mean_speed_on_paper18 - gmrt_on_paper18: Correlation = 0.96\n",
      "mean_jerk_in_air18 - mean_acc_in_air18: Correlation = 1.00\n",
      "gmrt_in_air18 - mean_gmrt18: Correlation = 0.94\n",
      "mean_acc_in_air18 - mean_jerk_in_air18: Correlation = 1.00\n",
      "gmrt_in_air18 - mean_speed_in_air18: Correlation = 0.93\n",
      "gmrt_on_paper18 - mean_speed_on_paper18: Correlation = 0.96\n",
      "air_time18 - num_of_pendown18: Correlation = 0.92\n",
      "total_time18 - num_of_pendown18: Correlation = 0.94\n",
      "total_time18 - paper_time18: Correlation = 0.92\n",
      "air_time18 - total_time18: Correlation = 0.99\n",
      "num_of_pendown18 - total_time18: Correlation = 0.94\n",
      "paper_time18 - total_time18: Correlation = 0.92\n",
      "total_time19 - air_time19: Correlation = 1.00\n",
      "mean_speed_in_air19 - gmrt_in_air19: Correlation = 0.99\n",
      "mean_speed_on_paper19 - gmrt_on_paper19: Correlation = 0.96\n",
      "mean_jerk_in_air19 - mean_acc_in_air19: Correlation = 1.00\n",
      "mean_acc_in_air19 - mean_jerk_in_air19: Correlation = 1.00\n",
      "gmrt_in_air19 - mean_speed_in_air19: Correlation = 0.99\n",
      "gmrt_on_paper19 - mean_speed_on_paper19: Correlation = 0.96\n",
      "air_time19 - total_time19: Correlation = 1.00\n",
      "total_time20 - air_time20: Correlation = 0.98\n",
      "mean_gmrt20 - gmrt_in_air20: Correlation = 0.92\n",
      "mean_speed_in_air20 - gmrt_in_air20: Correlation = 0.90\n",
      "mean_speed_on_paper20 - gmrt_on_paper20: Correlation = 0.97\n",
      "mean_jerk_in_air20 - mean_acc_in_air20: Correlation = 1.00\n",
      "gmrt_in_air20 - mean_gmrt20: Correlation = 0.92\n",
      "mean_acc_in_air20 - mean_jerk_in_air20: Correlation = 1.00\n",
      "gmrt_in_air20 - mean_speed_in_air20: Correlation = 0.90\n",
      "gmrt_on_paper20 - mean_speed_on_paper20: Correlation = 0.97\n",
      "air_time20 - total_time20: Correlation = 0.98\n",
      "mean_gmrt21 - gmrt_in_air21: Correlation = 0.99\n",
      "mean_speed_on_paper21 - gmrt_on_paper21: Correlation = 0.98\n",
      "mean_jerk_in_air21 - mean_acc_in_air21: Correlation = 1.00\n",
      "mean_speed_in_air21 - mean_acc_in_air21: Correlation = 0.94\n",
      "mean_jerk_on_paper21 - mean_acc_on_paper21: Correlation = 0.91\n",
      "gmrt_in_air21 - mean_gmrt21: Correlation = 0.99\n",
      "mean_acc_in_air21 - mean_jerk_in_air21: Correlation = 1.00\n",
      "mean_speed_in_air21 - mean_jerk_in_air21: Correlation = 0.93\n",
      "mean_acc_on_paper21 - mean_jerk_on_paper21: Correlation = 0.91\n",
      "mean_acc_in_air21 - mean_speed_in_air21: Correlation = 0.94\n",
      "mean_jerk_in_air21 - mean_speed_in_air21: Correlation = 0.93\n",
      "gmrt_on_paper21 - mean_speed_on_paper21: Correlation = 0.98\n",
      "total_time21 - paper_time21: Correlation = 0.94\n",
      "paper_time21 - total_time21: Correlation = 0.94\n",
      "total_time22 - air_time22: Correlation = 1.00\n",
      "mean_gmrt22 - gmrt_in_air22: Correlation = 0.90\n",
      "mean_speed_in_air22 - gmrt_in_air22: Correlation = 0.96\n",
      "mean_gmrt22 - gmrt_on_paper22: Correlation = 0.90\n",
      "mean_speed_on_paper22 - gmrt_on_paper22: Correlation = 0.98\n",
      "mean_jerk_in_air22 - mean_acc_in_air22: Correlation = 0.99\n",
      "gmrt_in_air22 - mean_gmrt22: Correlation = 0.90\n",
      "gmrt_on_paper22 - mean_gmrt22: Correlation = 0.90\n",
      "mean_acc_in_air22 - mean_jerk_in_air22: Correlation = 0.99\n",
      "gmrt_in_air22 - mean_speed_in_air22: Correlation = 0.96\n",
      "gmrt_on_paper22 - mean_speed_on_paper22: Correlation = 0.98\n",
      "air_time22 - total_time22: Correlation = 1.00\n",
      "total_time23 - air_time23: Correlation = 0.99\n",
      "mean_speed_in_air23 - gmrt_in_air23: Correlation = 0.97\n",
      "mean_gmrt23 - gmrt_on_paper23: Correlation = 0.92\n",
      "mean_speed_on_paper23 - gmrt_on_paper23: Correlation = 0.97\n",
      "mean_jerk_in_air23 - mean_acc_in_air23: Correlation = 0.99\n",
      "gmrt_on_paper23 - mean_gmrt23: Correlation = 0.92\n",
      "mean_speed_on_paper23 - mean_gmrt23: Correlation = 0.91\n",
      "mean_acc_in_air23 - mean_jerk_in_air23: Correlation = 0.99\n",
      "gmrt_in_air23 - mean_speed_in_air23: Correlation = 0.97\n",
      "gmrt_on_paper23 - mean_speed_on_paper23: Correlation = 0.97\n",
      "mean_gmrt23 - mean_speed_on_paper23: Correlation = 0.91\n",
      "air_time23 - total_time23: Correlation = 0.99\n",
      "total_time24 - air_time24: Correlation = 0.99\n",
      "mean_gmrt24 - gmrt_in_air24: Correlation = 0.93\n",
      "mean_speed_in_air24 - gmrt_in_air24: Correlation = 0.98\n",
      "mean_gmrt24 - gmrt_on_paper24: Correlation = 0.95\n",
      "mean_speed_on_paper24 - gmrt_on_paper24: Correlation = 0.92\n",
      "max_y_extension24 - max_x_extension24: Correlation = 0.93\n",
      "max_x_extension24 - max_y_extension24: Correlation = 0.93\n",
      "mean_jerk_in_air24 - mean_acc_in_air24: Correlation = 1.00\n",
      "gmrt_in_air24 - mean_gmrt24: Correlation = 0.93\n",
      "gmrt_on_paper24 - mean_gmrt24: Correlation = 0.95\n",
      "mean_acc_in_air24 - mean_jerk_in_air24: Correlation = 1.00\n",
      "gmrt_in_air24 - mean_speed_in_air24: Correlation = 0.98\n",
      "gmrt_on_paper24 - mean_speed_on_paper24: Correlation = 0.92\n",
      "air_time24 - total_time24: Correlation = 0.99\n",
      "total_time25 - air_time25: Correlation = 1.00\n",
      "mean_gmrt25 - gmrt_in_air25: Correlation = 0.95\n",
      "mean_speed_in_air25 - gmrt_in_air25: Correlation = 0.98\n",
      "mean_speed_on_paper25 - gmrt_on_paper25: Correlation = 0.96\n",
      "mean_jerk_in_air25 - mean_acc_in_air25: Correlation = 1.00\n",
      "gmrt_in_air25 - mean_gmrt25: Correlation = 0.95\n",
      "mean_speed_in_air25 - mean_gmrt25: Correlation = 0.94\n",
      "mean_acc_in_air25 - mean_jerk_in_air25: Correlation = 1.00\n",
      "gmrt_in_air25 - mean_speed_in_air25: Correlation = 0.98\n",
      "mean_gmrt25 - mean_speed_in_air25: Correlation = 0.94\n",
      "gmrt_on_paper25 - mean_speed_on_paper25: Correlation = 0.96\n",
      "air_time25 - total_time25: Correlation = 1.00\n",
      "Number of Highly Correlated Features: 210\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gmrt_in_air8</th>\n",
       "      <th>gmrt_in_air7</th>\n",
       "      <th>gmrt_on_paper14</th>\n",
       "      <th>mean_gmrt12</th>\n",
       "      <th>mean_jerk_in_air17</th>\n",
       "      <th>gmrt_in_air21</th>\n",
       "      <th>mean_acc_in_air8</th>\n",
       "      <th>mean_gmrt24</th>\n",
       "      <th>air_time25</th>\n",
       "      <th>mean_speed_in_air15</th>\n",
       "      <th>...</th>\n",
       "      <th>mean_gmrt17</th>\n",
       "      <th>mean_speed_in_air17</th>\n",
       "      <th>mean_jerk_in_air21</th>\n",
       "      <th>mean_jerk_in_air19</th>\n",
       "      <th>mean_jerk_in_air6</th>\n",
       "      <th>mean_speed_on_paper10</th>\n",
       "      <th>mean_speed_on_paper5</th>\n",
       "      <th>gmrt_on_paper10</th>\n",
       "      <th>mean_acc_in_air25</th>\n",
       "      <th>mean_acc_in_air9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>98.738919</td>\n",
       "      <td>390.168619</td>\n",
       "      <td>130.525720</td>\n",
       "      <td>260.869206</td>\n",
       "      <td>0.192499</td>\n",
       "      <td>251.622971</td>\n",
       "      <td>0.352634</td>\n",
       "      <td>189.016809</td>\n",
       "      <td>104485</td>\n",
       "      <td>2.276176</td>\n",
       "      <td>...</td>\n",
       "      <td>260.639185</td>\n",
       "      <td>4.396491</td>\n",
       "      <td>0.058171</td>\n",
       "      <td>0.124412</td>\n",
       "      <td>0.074448</td>\n",
       "      <td>4.146069</td>\n",
       "      <td>0.293197</td>\n",
       "      <td>212.887096</td>\n",
       "      <td>0.785139</td>\n",
       "      <td>0.706240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>85.904279</td>\n",
       "      <td>133.477130</td>\n",
       "      <td>65.171424</td>\n",
       "      <td>109.542648</td>\n",
       "      <td>0.098941</td>\n",
       "      <td>290.800361</td>\n",
       "      <td>0.342866</td>\n",
       "      <td>60.047652</td>\n",
       "      <td>171940</td>\n",
       "      <td>3.148314</td>\n",
       "      <td>...</td>\n",
       "      <td>99.956691</td>\n",
       "      <td>2.223984</td>\n",
       "      <td>0.077640</td>\n",
       "      <td>0.080506</td>\n",
       "      <td>0.128930</td>\n",
       "      <td>0.978483</td>\n",
       "      <td>1.392394</td>\n",
       "      <td>50.873449</td>\n",
       "      <td>0.332026</td>\n",
       "      <td>0.194774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>936.464720</td>\n",
       "      <td>273.908579</td>\n",
       "      <td>172.657769</td>\n",
       "      <td>133.087773</td>\n",
       "      <td>0.410326</td>\n",
       "      <td>703.155498</td>\n",
       "      <td>6.121764</td>\n",
       "      <td>97.829343</td>\n",
       "      <td>33545</td>\n",
       "      <td>3.922532</td>\n",
       "      <td>...</td>\n",
       "      <td>261.046182</td>\n",
       "      <td>6.947663</td>\n",
       "      <td>0.090782</td>\n",
       "      <td>0.173414</td>\n",
       "      <td>0.022402</td>\n",
       "      <td>2.494940</td>\n",
       "      <td>1.540444</td>\n",
       "      <td>123.418668</td>\n",
       "      <td>0.941396</td>\n",
       "      <td>0.229538</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>121.386016</td>\n",
       "      <td>162.139813</td>\n",
       "      <td>116.981707</td>\n",
       "      <td>193.529969</td>\n",
       "      <td>0.196077</td>\n",
       "      <td>451.279786</td>\n",
       "      <td>0.211286</td>\n",
       "      <td>93.496877</td>\n",
       "      <td>113275</td>\n",
       "      <td>3.857709</td>\n",
       "      <td>...</td>\n",
       "      <td>214.161758</td>\n",
       "      <td>5.219644</td>\n",
       "      <td>0.352951</td>\n",
       "      <td>0.088139</td>\n",
       "      <td>0.208204</td>\n",
       "      <td>1.097212</td>\n",
       "      <td>0.528355</td>\n",
       "      <td>71.648968</td>\n",
       "      <td>0.642378</td>\n",
       "      <td>1.039627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>198.886419</td>\n",
       "      <td>420.384148</td>\n",
       "      <td>97.601258</td>\n",
       "      <td>171.398363</td>\n",
       "      <td>0.330048</td>\n",
       "      <td>288.615548</td>\n",
       "      <td>1.309207</td>\n",
       "      <td>85.123242</td>\n",
       "      <td>35290</td>\n",
       "      <td>1.655997</td>\n",
       "      <td>...</td>\n",
       "      <td>262.685522</td>\n",
       "      <td>6.334390</td>\n",
       "      <td>0.120758</td>\n",
       "      <td>0.102956</td>\n",
       "      <td>0.350512</td>\n",
       "      <td>1.432330</td>\n",
       "      <td>0.672337</td>\n",
       "      <td>71.451206</td>\n",
       "      <td>0.680095</td>\n",
       "      <td>0.239087</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 210 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   gmrt_in_air8  gmrt_in_air7  gmrt_on_paper14  mean_gmrt12  \\\n",
       "0     98.738919    390.168619       130.525720   260.869206   \n",
       "1     85.904279    133.477130        65.171424   109.542648   \n",
       "2    936.464720    273.908579       172.657769   133.087773   \n",
       "3    121.386016    162.139813       116.981707   193.529969   \n",
       "4    198.886419    420.384148        97.601258   171.398363   \n",
       "\n",
       "   mean_jerk_in_air17  gmrt_in_air21  mean_acc_in_air8  mean_gmrt24  \\\n",
       "0            0.192499     251.622971          0.352634   189.016809   \n",
       "1            0.098941     290.800361          0.342866    60.047652   \n",
       "2            0.410326     703.155498          6.121764    97.829343   \n",
       "3            0.196077     451.279786          0.211286    93.496877   \n",
       "4            0.330048     288.615548          1.309207    85.123242   \n",
       "\n",
       "   air_time25  mean_speed_in_air15  ...  mean_gmrt17  mean_speed_in_air17  \\\n",
       "0      104485             2.276176  ...   260.639185             4.396491   \n",
       "1      171940             3.148314  ...    99.956691             2.223984   \n",
       "2       33545             3.922532  ...   261.046182             6.947663   \n",
       "3      113275             3.857709  ...   214.161758             5.219644   \n",
       "4       35290             1.655997  ...   262.685522             6.334390   \n",
       "\n",
       "   mean_jerk_in_air21  mean_jerk_in_air19  mean_jerk_in_air6  \\\n",
       "0            0.058171            0.124412           0.074448   \n",
       "1            0.077640            0.080506           0.128930   \n",
       "2            0.090782            0.173414           0.022402   \n",
       "3            0.352951            0.088139           0.208204   \n",
       "4            0.120758            0.102956           0.350512   \n",
       "\n",
       "   mean_speed_on_paper10  mean_speed_on_paper5  gmrt_on_paper10  \\\n",
       "0               4.146069              0.293197       212.887096   \n",
       "1               0.978483              1.392394        50.873449   \n",
       "2               2.494940              1.540444       123.418668   \n",
       "3               1.097212              0.528355        71.648968   \n",
       "4               1.432330              0.672337        71.451206   \n",
       "\n",
       "   mean_acc_in_air25  mean_acc_in_air9  \n",
       "0           0.785139          0.706240  \n",
       "1           0.332026          0.194774  \n",
       "2           0.941396          0.229538  \n",
       "3           0.642378          1.039627  \n",
       "4           0.680095          0.239087  \n",
       "\n",
       "[5 rows x 210 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_matrix = data.corr()\n",
    "\n",
    "high_corr_features = []\n",
    "threshold = 0.9\n",
    "for col in corr_matrix.columns:\n",
    "    for row in corr_matrix.index:\n",
    "        if abs(corr_matrix.loc[row, col]) > threshold and row != col:\n",
    "            high_corr_features.append((row, col, corr_matrix.loc[row, col]))\n",
    "\n",
    "print(\"\\nHighly Correlated Features (Correlation > 0.9):\")\n",
    "for feature in high_corr_features:\n",
    "    print(f\"{feature[0]} - {feature[1]}: Correlation = {feature[2]:.2f}\")\n",
    "\n",
    "unique_columns = set([feature[0] for feature in high_corr_features] + [feature[1] for feature in high_corr_features])\n",
    "df_corr = data[list(unique_columns)]\n",
    "print(f\"Number of Highly Correlated Features: {len(unique_columns)}\")\n",
    "df_corr.head()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bdf9153",
   "metadata": {
    "papermill": {
     "duration": 0.022658,
     "end_time": "2024-09-21T18:25:54.154880",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.132222",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Grouping Features by Task"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "00a57be5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:54.202985Z",
     "iopub.status.busy": "2024-09-21T18:25:54.202162Z",
     "iopub.status.idle": "2024-09-21T18:25:54.211432Z",
     "shell.execute_reply": "2024-09-21T18:25:54.210592Z"
    },
    "papermill": {
     "duration": 0.035085,
     "end_time": "2024-09-21T18:25:54.213764",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.178679",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Number of features: Task Number N1: 18\n",
      "\n",
      "Number of features: Task Number N2: 18\n",
      "\n",
      "Number of features: Task Number N3: 18\n",
      "\n",
      "Number of features: Task Number N4: 18\n",
      "\n",
      "Number of features: Task Number N5: 18\n",
      "\n",
      "Number of features: Task Number N6: 18\n",
      "\n",
      "Number of features: Task Number N7: 18\n",
      "\n",
      "Number of features: Task Number N8: 18\n",
      "\n",
      "Number of features: Task Number N9: 18\n",
      "\n",
      "Number of features: Task Number N10: 18\n",
      "\n",
      "Number of features: Task Number N11: 18\n",
      "\n",
      "Number of features: Task Number N12: 18\n",
      "\n",
      "Number of features: Task Number N13: 18\n",
      "\n",
      "Number of features: Task Number N14: 18\n",
      "\n",
      "Number of features: Task Number N15: 18\n",
      "\n",
      "Number of features: Task Number N16: 18\n",
      "\n",
      "Number of features: Task Number N17: 18\n",
      "\n",
      "Number of features: Task Number N18: 18\n",
      "\n",
      "Number of features: Task Number N19: 18\n",
      "\n",
      "Number of features: Task Number N20: 18\n",
      "\n",
      "Number of features: Task Number N21: 18\n",
      "\n",
      "Number of features: Task Number N22: 18\n",
      "\n",
      "Number of features: Task Number N23: 18\n",
      "\n",
      "Number of features: Task Number N24: 18\n",
      "\n",
      "Number of features: Task Number N25: 18\n"
     ]
    }
   ],
   "source": [
    "features_by_task = defaultdict(list)\n",
    "for col in data.columns:\n",
    "    match = re.search(r'(\\d+)$', col)  \n",
    "    if match:\n",
    "        x_value = match.group(1)  \n",
    "        features_by_task[x_value].append(col)   \n",
    "\n",
    "\n",
    "for x_value, features in features_by_task.items():\n",
    "    print(f\"\\nNumber of features: Task Number N{x_value}: {len(features)}\")\n",
    "df_features_by_task = pd.DataFrame([{'X Value': x_value, 'Features': features}\n",
    "    for x_value, features in features_by_task.items()])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8430d36b",
   "metadata": {
    "papermill": {
     "duration": 0.0219,
     "end_time": "2024-09-21T18:25:54.257425",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.235525",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Feature Distribution per Task"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fe22e499",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:54.303458Z",
     "iopub.status.busy": "2024-09-21T18:25:54.303123Z",
     "iopub.status.idle": "2024-09-21T18:25:54.314496Z",
     "shell.execute_reply": "2024-09-21T18:25:54.313615Z"
    },
    "papermill": {
     "duration": 0.036272,
     "end_time": "2024-09-21T18:25:54.316426",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.280154",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_task_distribution(task_name):\n",
    "    columns = features_by_task.get(task_name, [])\n",
    "    num_cols = 5\n",
    "    num_rows = len(columns) // num_cols + (len(columns) % num_cols > 0)\n",
    "    \n",
    "    plt.figure(figsize=(20, num_rows * 5))\n",
    "    \n",
    "    for i, col in enumerate(columns):\n",
    "        plt.subplot(num_rows, num_cols, i + 1)\n",
    "        sns.histplot(data[col], kde=True)\n",
    "        plt.title(f'Histogramme: {col}')\n",
    "    \n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "task_selector = widgets.Dropdown(\n",
    "    options=list(features_by_task.keys()),\n",
    "    value=list(features_by_task.keys())[0],\n",
    "    description='Task:',)\n",
    "\n",
    "def update_histograms(task_name):\n",
    "    plt.close('all')  \n",
    "    plot_task_distribution(task_name)\n",
    "# Eliminate # to run the code(interactive plot) \n",
    "#display(widgets.interactive(update_histograms, task_name=task_selector))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2490353",
   "metadata": {
    "papermill": {
     "duration": 0.022001,
     "end_time": "2024-09-21T18:25:54.360293",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.338292",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Boxplot for Feature Distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "67b37dcb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:54.406776Z",
     "iopub.status.busy": "2024-09-21T18:25:54.406413Z",
     "iopub.status.idle": "2024-09-21T18:25:54.418121Z",
     "shell.execute_reply": "2024-09-21T18:25:54.417149Z"
    },
    "papermill": {
     "duration": 0.038866,
     "end_time": "2024-09-21T18:25:54.421344",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.382478",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_task_boxplots(task_name):\n",
    "    columns = features_by_task.get(task_name, [])\n",
    "    num_cols = 5\n",
    "    num_rows = len(columns) // num_cols + (len(columns) % num_cols > 0)\n",
    "    \n",
    "    plt.figure(figsize=(20, num_rows * 5))\n",
    "    \n",
    "    for i, col in enumerate(columns):\n",
    "        plt.subplot(num_rows, num_cols, i + 1)\n",
    "        sns.boxplot(x=data[col])\n",
    "        plt.title(f'Boxplot: {col}')\n",
    "    \n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "task_selector = widgets.Dropdown(\n",
    "    options=list(features_by_task.keys()),\n",
    "    value=list(features_by_task.keys())[0],\n",
    "    description='Task:',)\n",
    "\n",
    "def update_boxplots(task_name):\n",
    "    plt.close('all')  \n",
    "    plot_task_boxplots(task_name)\n",
    "\n",
    "# Eliminate # to run the code(interactive plot) \n",
    "#display(widgets.interactive(update_boxplots, task_name=task_selector))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b27d49b2",
   "metadata": {
    "papermill": {
     "duration": 0.022743,
     "end_time": "2024-09-21T18:25:54.467671",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.444928",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Correlation Matrix per Task"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "13084c6c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:54.514580Z",
     "iopub.status.busy": "2024-09-21T18:25:54.514240Z",
     "iopub.status.idle": "2024-09-21T18:25:54.524803Z",
     "shell.execute_reply": "2024-09-21T18:25:54.523881Z"
    },
    "papermill": {
     "duration": 0.03756,
     "end_time": "2024-09-21T18:25:54.528086",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.490526",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_task_heatmaps(task_name):\n",
    "    columns = features_by_task.get(task_name, [])\n",
    "    data_subset = data[columns].dropna()  \n",
    "    correlation_matrix = data_subset.corr()\n",
    "    \n",
    "    plt.figure(figsize=(12, 10))\n",
    "    sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', fmt='.2f')\n",
    "    plt.title(f'Matrice des Corrélations pour la tâche X={task_name}')\n",
    "    plt.show()\n",
    "\n",
    "task_selector = widgets.Dropdown(\n",
    "    options=list(features_by_task.keys()),\n",
    "    value=list(features_by_task.keys())[0],\n",
    "    description='Task:',)\n",
    "\n",
    "def update_heatmaps(task_name):\n",
    "    plt.close('all')  \n",
    "    plot_task_heatmaps(task_name)\n",
    "# Eliminate # to run the code(interactive plot) \n",
    "\n",
    "#display(widgets.interactive(update_heatmaps, task_name=task_selector))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fcb7f10",
   "metadata": {
    "papermill": {
     "duration": 0.064919,
     "end_time": "2024-09-21T18:25:54.615580",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.550661",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Feature Importance Ranking Using RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a61c4765",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:54.661687Z",
     "iopub.status.busy": "2024-09-21T18:25:54.660731Z",
     "iopub.status.idle": "2024-09-21T18:25:54.684617Z",
     "shell.execute_reply": "2024-09-21T18:25:54.683722Z"
    },
    "papermill": {
     "duration": 0.048838,
     "end_time": "2024-09-21T18:25:54.686780",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.637942",
     "status": "completed"
    },
    "tags": []
   },
   "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>air_time1</th>\n",
       "      <th>disp_index1</th>\n",
       "      <th>gmrt_in_air1</th>\n",
       "      <th>gmrt_on_paper1</th>\n",
       "      <th>max_x_extension1</th>\n",
       "      <th>max_y_extension1</th>\n",
       "      <th>mean_acc_in_air1</th>\n",
       "      <th>mean_acc_on_paper1</th>\n",
       "      <th>mean_gmrt1</th>\n",
       "      <th>mean_jerk_in_air1</th>\n",
       "      <th>...</th>\n",
       "      <th>mean_jerk_in_air25</th>\n",
       "      <th>mean_jerk_on_paper25</th>\n",
       "      <th>mean_speed_in_air25</th>\n",
       "      <th>mean_speed_on_paper25</th>\n",
       "      <th>num_of_pendown25</th>\n",
       "      <th>paper_time25</th>\n",
       "      <th>pressure_mean25</th>\n",
       "      <th>pressure_var25</th>\n",
       "      <th>total_time25</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5160</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>120.804174</td>\n",
       "      <td>86.853334</td>\n",
       "      <td>957</td>\n",
       "      <td>6601</td>\n",
       "      <td>0.361800</td>\n",
       "      <td>0.217459</td>\n",
       "      <td>103.828754</td>\n",
       "      <td>0.051836</td>\n",
       "      <td>...</td>\n",
       "      <td>0.141434</td>\n",
       "      <td>0.024471</td>\n",
       "      <td>5.596487</td>\n",
       "      <td>3.184589</td>\n",
       "      <td>71</td>\n",
       "      <td>40120</td>\n",
       "      <td>1749.278166</td>\n",
       "      <td>296102.7676</td>\n",
       "      <td>144605</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>51980</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>115.318238</td>\n",
       "      <td>83.448681</td>\n",
       "      <td>1694</td>\n",
       "      <td>6998</td>\n",
       "      <td>0.272513</td>\n",
       "      <td>0.144880</td>\n",
       "      <td>99.383459</td>\n",
       "      <td>0.039827</td>\n",
       "      <td>...</td>\n",
       "      <td>0.049663</td>\n",
       "      <td>0.018368</td>\n",
       "      <td>1.665973</td>\n",
       "      <td>0.950249</td>\n",
       "      <td>129</td>\n",
       "      <td>126700</td>\n",
       "      <td>1504.768272</td>\n",
       "      <td>278744.2850</td>\n",
       "      <td>298640</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2600</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>229.933997</td>\n",
       "      <td>172.761858</td>\n",
       "      <td>2333</td>\n",
       "      <td>5802</td>\n",
       "      <td>0.387020</td>\n",
       "      <td>0.181342</td>\n",
       "      <td>201.347928</td>\n",
       "      <td>0.064220</td>\n",
       "      <td>...</td>\n",
       "      <td>0.178194</td>\n",
       "      <td>0.017174</td>\n",
       "      <td>4.000781</td>\n",
       "      <td>2.392521</td>\n",
       "      <td>74</td>\n",
       "      <td>45480</td>\n",
       "      <td>1431.443492</td>\n",
       "      <td>144411.7055</td>\n",
       "      <td>79025</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2130</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>369.403342</td>\n",
       "      <td>183.193104</td>\n",
       "      <td>1756</td>\n",
       "      <td>8159</td>\n",
       "      <td>0.556879</td>\n",
       "      <td>0.164502</td>\n",
       "      <td>276.298223</td>\n",
       "      <td>0.090408</td>\n",
       "      <td>...</td>\n",
       "      <td>0.113905</td>\n",
       "      <td>0.019860</td>\n",
       "      <td>4.206746</td>\n",
       "      <td>1.613522</td>\n",
       "      <td>123</td>\n",
       "      <td>67945</td>\n",
       "      <td>1465.843329</td>\n",
       "      <td>230184.7154</td>\n",
       "      <td>181220</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2310</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>257.997131</td>\n",
       "      <td>111.275889</td>\n",
       "      <td>987</td>\n",
       "      <td>4732</td>\n",
       "      <td>0.266077</td>\n",
       "      <td>0.145104</td>\n",
       "      <td>184.636510</td>\n",
       "      <td>0.037528</td>\n",
       "      <td>...</td>\n",
       "      <td>0.121782</td>\n",
       "      <td>0.020872</td>\n",
       "      <td>3.319036</td>\n",
       "      <td>1.680629</td>\n",
       "      <td>92</td>\n",
       "      <td>37285</td>\n",
       "      <td>1841.702561</td>\n",
       "      <td>158290.0255</td>\n",
       "      <td>72575</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 451 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   air_time1  disp_index1  gmrt_in_air1  gmrt_on_paper1  max_x_extension1  \\\n",
       "0       5160     0.000013    120.804174       86.853334               957   \n",
       "1      51980     0.000016    115.318238       83.448681              1694   \n",
       "2       2600     0.000010    229.933997      172.761858              2333   \n",
       "3       2130     0.000010    369.403342      183.193104              1756   \n",
       "4       2310     0.000007    257.997131      111.275889               987   \n",
       "\n",
       "   max_y_extension1  mean_acc_in_air1  mean_acc_on_paper1  mean_gmrt1  \\\n",
       "0              6601          0.361800            0.217459  103.828754   \n",
       "1              6998          0.272513            0.144880   99.383459   \n",
       "2              5802          0.387020            0.181342  201.347928   \n",
       "3              8159          0.556879            0.164502  276.298223   \n",
       "4              4732          0.266077            0.145104  184.636510   \n",
       "\n",
       "   mean_jerk_in_air1  ...  mean_jerk_in_air25  mean_jerk_on_paper25  \\\n",
       "0           0.051836  ...            0.141434              0.024471   \n",
       "1           0.039827  ...            0.049663              0.018368   \n",
       "2           0.064220  ...            0.178194              0.017174   \n",
       "3           0.090408  ...            0.113905              0.019860   \n",
       "4           0.037528  ...            0.121782              0.020872   \n",
       "\n",
       "   mean_speed_in_air25  mean_speed_on_paper25  num_of_pendown25  paper_time25  \\\n",
       "0             5.596487               3.184589                71         40120   \n",
       "1             1.665973               0.950249               129        126700   \n",
       "2             4.000781               2.392521                74         45480   \n",
       "3             4.206746               1.613522               123         67945   \n",
       "4             3.319036               1.680629                92         37285   \n",
       "\n",
       "   pressure_mean25  pressure_var25  total_time25  class  \n",
       "0      1749.278166     296102.7676        144605      1  \n",
       "1      1504.768272     278744.2850        298640      1  \n",
       "2      1431.443492     144411.7055         79025      1  \n",
       "3      1465.843329     230184.7154        181220      1  \n",
       "4      1841.702561     158290.0255         72575      1  \n",
       "\n",
       "[5 rows x 451 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "78e3890d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:54.733033Z",
     "iopub.status.busy": "2024-09-21T18:25:54.732611Z",
     "iopub.status.idle": "2024-09-21T18:25:55.029311Z",
     "shell.execute_reply": "2024-09-21T18:25:55.028360Z"
    },
    "papermill": {
     "duration": 0.321949,
     "end_time": "2024-09-21T18:25:55.031497",
     "exception": false,
     "start_time": "2024-09-21T18:25:54.709548",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(random_state=42)</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(random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomForestClassifier(random_state=42)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = data.drop(['class'], axis=1) \n",
    "y = data['class']  \n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "rf = RandomForestClassifier(random_state=42)\n",
    "rf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c547b960",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:55.078321Z",
     "iopub.status.busy": "2024-09-21T18:25:55.077707Z",
     "iopub.status.idle": "2024-09-21T18:25:55.099766Z",
     "shell.execute_reply": "2024-09-21T18:25:55.098859Z"
    },
    "papermill": {
     "duration": 0.049838,
     "end_time": "2024-09-21T18:25:55.103876",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.054038",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Feature importance rank:\n",
      "total_time23: 0.041878\n",
      "total_time15: 0.0370254\n",
      "air_time15: 0.0308567\n",
      "total_time6: 0.0238327\n",
      "air_time23: 0.0213317\n",
      "air_time22: 0.0203271\n",
      "total_time9: 0.0193132\n",
      "total_time22: 0.016894\n",
      "total_time17: 0.0157858\n",
      "total_time13: 0.0155757\n",
      "pressure_var19: 0.0135534\n",
      "total_time7: 0.0134772\n",
      "air_time16: 0.0120298\n",
      "num_of_pendown19: 0.0113966\n",
      "total_time3: 0.0112034\n",
      "air_time17: 0.0108622\n",
      "paper_time17: 0.0105935\n",
      "pressure_mean4: 0.0095681\n",
      "air_time6: 0.0092513\n",
      "pressure_mean5: 0.0090991\n",
      "air_time5: 0.0087059\n",
      "total_time8: 0.0081809\n",
      "mean_jerk_in_air3: 0.0080426\n",
      "air_time7: 0.0079574\n",
      "paper_time20: 0.0078961\n",
      "paper_time15: 0.0075927\n",
      "mean_gmrt17: 0.0074134\n",
      "total_time11: 0.0072496\n",
      "gmrt_in_air25: 0.0065404\n",
      "mean_speed_on_paper22: 0.0064702\n",
      "max_y_extension20: 0.0064495\n",
      "mean_speed_on_paper10: 0.0061687\n",
      "pressure_var5: 0.0059514\n",
      "mean_speed_on_paper23: 0.0059418\n",
      "air_time1: 0.0058801\n",
      "total_time12: 0.0058484\n",
      "total_time2: 0.0057877\n",
      "gmrt_on_paper1: 0.0057343\n",
      "paper_time9: 0.0056385\n",
      "air_time2: 0.0056254\n",
      "max_y_extension3: 0.005603\n",
      "paper_time8: 0.0054455\n",
      "air_time13: 0.0054211\n",
      "pressure_mean21: 0.005408\n",
      "pressure_var8: 0.0053711\n",
      "pressure_mean8: 0.005354\n",
      "mean_speed_in_air23: 0.0052334\n",
      "pressure_mean19: 0.0050218\n",
      "disp_index7: 0.0050057\n",
      "paper_time6: 0.0049962\n",
      "disp_index17: 0.0047243\n",
      "air_time24: 0.0046846\n",
      "paper_time14: 0.0046841\n",
      "mean_speed_on_paper7: 0.0044899\n",
      "pressure_mean9: 0.0044182\n",
      "mean_gmrt7: 0.0043973\n",
      "mean_gmrt15: 0.0043315\n",
      "paper_time13: 0.0042274\n",
      "disp_index8: 0.0041923\n",
      "air_time9: 0.0041707\n",
      "paper_time23: 0.0041271\n",
      "mean_speed_on_paper15: 0.004106\n",
      "total_time20: 0.0040986\n",
      "mean_speed_on_paper2: 0.0040564\n",
      "air_time14: 0.0039954\n",
      "mean_acc_on_paper9: 0.0038327\n",
      "max_x_extension23: 0.0038055\n",
      "mean_gmrt1: 0.0038013\n",
      "gmrt_in_air7: 0.0037339\n",
      "air_time4: 0.0037219\n",
      "mean_gmrt18: 0.0036749\n",
      "mean_acc_on_paper24: 0.0035723\n",
      "gmrt_in_air23: 0.0035684\n",
      "air_time3: 0.0035436\n",
      "disp_index22: 0.0034364\n",
      "mean_jerk_on_paper21: 0.0034277\n",
      "paper_time12: 0.0033695\n",
      "gmrt_on_paper15: 0.0033384\n",
      "paper_time11: 0.0033336\n",
      "max_y_extension19: 0.0032417\n",
      "air_time10: 0.0031602\n",
      "gmrt_on_paper8: 0.0030447\n",
      "mean_jerk_in_air7: 0.0030444\n",
      "mean_speed_in_air7: 0.0030444\n",
      "gmrt_in_air18: 0.0030437\n",
      "pressure_var23: 0.0029745\n",
      "mean_gmrt8: 0.0029427\n",
      "gmrt_in_air3: 0.0029212\n",
      "mean_jerk_on_paper19: 0.002916\n",
      "gmrt_in_air2: 0.0028803\n",
      "air_time21: 0.0028383\n",
      "pressure_var12: 0.0027698\n",
      "total_time16: 0.0027445\n",
      "mean_acc_in_air22: 0.0027011\n",
      "mean_gmrt22: 0.0026492\n",
      "mean_jerk_on_paper15: 0.0026315\n",
      "num_of_pendown22: 0.0025856\n",
      "paper_time7: 0.0025827\n",
      "max_x_extension15: 0.0025685\n",
      "mean_jerk_in_air13: 0.0025469\n",
      "paper_time16: 0.0025317\n",
      "gmrt_on_paper20: 0.0024832\n",
      "max_x_extension20: 0.002482\n",
      "max_x_extension11: 0.0024697\n",
      "pressure_mean18: 0.0023904\n",
      "mean_jerk_in_air23: 0.0023822\n",
      "mean_speed_on_paper13: 0.0023459\n",
      "gmrt_on_paper21: 0.0023026\n",
      "gmrt_on_paper19: 0.0022976\n",
      "mean_jerk_on_paper24: 0.0022798\n",
      "gmrt_in_air20: 0.0022462\n",
      "total_time4: 0.0022114\n",
      "paper_time22: 0.0022045\n",
      "mean_speed_in_air12: 0.0021797\n",
      "paper_time25: 0.0021761\n",
      "mean_jerk_on_paper16: 0.0021684\n",
      "pressure_mean1: 0.0021522\n",
      "paper_time5: 0.0021484\n",
      "mean_gmrt25: 0.0021444\n",
      "pressure_mean10: 0.0020693\n",
      "gmrt_on_paper23: 0.0020593\n",
      "disp_index9: 0.0020377\n",
      "mean_speed_on_paper9: 0.0020322\n",
      "disp_index15: 0.0020245\n",
      "num_of_pendown16: 0.0020201\n",
      "mean_acc_on_paper2: 0.0020181\n",
      "paper_time3: 0.0020022\n",
      "mean_acc_in_air4: 0.0019976\n",
      "paper_time10: 0.0019976\n",
      "disp_index23: 0.0019881\n",
      "mean_acc_in_air18: 0.0019872\n",
      "mean_speed_on_paper12: 0.0019644\n",
      "mean_gmrt19: 0.0019636\n",
      "mean_acc_in_air16: 0.0019442\n",
      "mean_jerk_on_paper14: 0.0019031\n",
      "disp_index3: 0.0018676\n",
      "pressure_var6: 0.00185\n",
      "pressure_mean15: 0.0018491\n",
      "gmrt_on_paper2: 0.0018356\n",
      "max_x_extension22: 0.0017891\n",
      "mean_speed_on_paper6: 0.0017676\n",
      "mean_jerk_in_air15: 0.0017312\n",
      "mean_acc_on_paper22: 0.0017072\n",
      "max_y_extension25: 0.0016637\n",
      "air_time11: 0.0016613\n",
      "num_of_pendown5: 0.0016585\n",
      "gmrt_on_paper22: 0.0016471\n",
      "mean_gmrt5: 0.0016409\n",
      "total_time5: 0.0016386\n",
      "mean_jerk_in_air11: 0.0016322\n",
      "air_time8: 0.0016186\n",
      "max_x_extension19: 0.0015751\n",
      "mean_speed_in_air5: 0.0015586\n",
      "mean_speed_in_air20: 0.001553\n",
      "disp_index4: 0.0015407\n",
      "mean_speed_on_paper1: 0.001523\n",
      "mean_speed_in_air24: 0.0015191\n",
      "mean_jerk_on_paper9: 0.0015058\n",
      "mean_speed_in_air14: 0.0014846\n",
      "mean_jerk_on_paper6: 0.0014754\n",
      "mean_jerk_in_air2: 0.0014609\n",
      "pressure_var20: 0.0014562\n",
      "mean_jerk_on_paper23: 0.0014482\n",
      "mean_jerk_in_air22: 0.0014257\n",
      "mean_acc_in_air10: 0.0014187\n",
      "mean_jerk_in_air24: 0.0014179\n",
      "disp_index5: 0.001413\n",
      "mean_gmrt21: 0.0014094\n",
      "disp_index6: 0.0013897\n",
      "mean_jerk_on_paper25: 0.0013894\n",
      "mean_jerk_on_paper2: 0.0013805\n",
      "mean_acc_in_air5: 0.0013759\n",
      "num_of_pendown4: 0.0013631\n",
      "mean_jerk_in_air1: 0.0013578\n",
      "max_x_extension3: 0.0013411\n",
      "disp_index18: 0.0013377\n",
      "air_time25: 0.0013293\n",
      "gmrt_on_paper6: 0.0013252\n",
      "max_x_extension16: 0.0013241\n",
      "gmrt_in_air13: 0.0013064\n",
      "max_y_extension2: 0.0013022\n",
      "mean_acc_on_paper3: 0.0012925\n",
      "mean_gmrt6: 0.001279\n",
      "mean_jerk_in_air18: 0.0012735\n",
      "gmrt_in_air19: 0.0012538\n",
      "disp_index13: 0.0012516\n",
      "mean_jerk_in_air19: 0.0012467\n",
      "num_of_pendown15: 0.0012464\n",
      "mean_speed_in_air25: 0.0012463\n",
      "mean_speed_on_paper8: 0.0012425\n",
      "mean_speed_in_air2: 0.0012372\n",
      "pressure_var11: 0.0012279\n",
      "mean_speed_on_paper14: 0.0012239\n",
      "pressure_mean17: 0.0012121\n",
      "pressure_var4: 0.0012017\n",
      "pressure_mean7: 0.0011943\n",
      "mean_gmrt2: 0.0011709\n",
      "max_y_extension5: 0.001164\n",
      "mean_jerk_in_air25: 0.0011551\n",
      "pressure_var16: 0.0011545\n",
      "pressure_mean23: 0.0011331\n",
      "max_y_extension13: 0.0011133\n",
      "mean_gmrt16: 0.0011087\n",
      "max_x_extension5: 0.0010983\n",
      "mean_gmrt13: 0.001097\n",
      "mean_gmrt10: 0.0010861\n",
      "mean_speed_in_air6: 0.0010855\n",
      "mean_acc_in_air19: 0.0010816\n",
      "max_x_extension9: 0.0010759\n",
      "mean_acc_in_air24: 0.0010726\n",
      "mean_jerk_on_paper7: 0.001071\n",
      "disp_index21: 0.001065\n",
      "num_of_pendown10: 0.0010543\n",
      "mean_speed_in_air11: 0.001039\n",
      "pressure_var21: 0.0010366\n",
      "mean_gmrt11: 0.001034\n",
      "mean_speed_on_paper11: 0.001031\n",
      "gmrt_on_paper9: 0.0010274\n",
      "disp_index19: 0.0010252\n",
      "disp_index24: 0.0010197\n",
      "gmrt_in_air16: 0.0010181\n",
      "mean_speed_in_air9: 0.0010103\n",
      "pressure_var3: 0.0010097\n",
      "mean_acc_in_air20: 0.0010064\n",
      "gmrt_on_paper5: 0.0010012\n",
      "pressure_mean6: 0.0009887\n",
      "max_y_extension7: 0.0009872\n",
      "pressure_var13: 0.0009852\n",
      "mean_jerk_on_paper1: 0.0009816\n",
      "mean_acc_in_air11: 0.000965\n",
      "gmrt_in_air10: 0.0009602\n",
      "mean_gmrt12: 0.0009589\n",
      "mean_jerk_on_paper12: 0.0009578\n",
      "mean_jerk_on_paper20: 0.0009529\n",
      "pressure_mean24: 0.0009508\n",
      "disp_index11: 0.0009501\n",
      "mean_jerk_in_air9: 0.0009396\n",
      "max_y_extension23: 0.0009249\n",
      "paper_time18: 0.0009219\n",
      "gmrt_in_air9: 0.0009124\n",
      "num_of_pendown1: 0.0009067\n",
      "mean_acc_in_air15: 0.0008998\n",
      "mean_speed_in_air13: 0.0008934\n",
      "max_y_extension24: 0.0008857\n",
      "mean_acc_on_paper16: 0.0008839\n",
      "disp_index25: 0.0008772\n",
      "mean_jerk_in_air14: 0.0008771\n",
      "gmrt_in_air5: 0.0008735\n",
      "mean_acc_on_paper15: 0.0008713\n",
      "mean_speed_in_air8: 0.0008702\n",
      "gmrt_on_paper3: 0.0008536\n",
      "pressure_mean3: 0.0008466\n",
      "max_x_extension21: 0.0008404\n",
      "mean_speed_in_air22: 0.0008261\n",
      "mean_acc_on_paper21: 0.0008205\n",
      "gmrt_in_air11: 0.0008197\n",
      "mean_gmrt9: 0.0008164\n",
      "mean_speed_in_air19: 0.0008057\n",
      "mean_speed_on_paper21: 0.0008015\n",
      "gmrt_in_air17: 0.0007953\n",
      "max_x_extension12: 0.0007947\n",
      "max_x_extension25: 0.0007848\n",
      "gmrt_on_paper14: 0.000782\n",
      "gmrt_on_paper12: 0.0007733\n",
      "max_x_extension2: 0.0007653\n",
      "mean_acc_in_air7: 0.0007648\n",
      "max_x_extension13: 0.000764\n",
      "mean_acc_on_paper1: 0.0007571\n",
      "air_time19: 0.000747\n",
      "total_time24: 0.0007338\n",
      "pressure_mean22: 0.0007336\n",
      "mean_jerk_in_air6: 0.0007333\n",
      "total_time10: 0.0007317\n",
      "mean_acc_on_paper7: 0.0007305\n",
      "mean_speed_in_air18: 0.0007055\n",
      "gmrt_in_air1: 0.0007004\n",
      "pressure_mean20: 0.0006996\n",
      "mean_jerk_in_air17: 0.0006935\n",
      "mean_speed_in_air17: 0.0006864\n",
      "mean_gmrt3: 0.0006782\n",
      "gmrt_in_air12: 0.0006543\n",
      "gmrt_in_air6: 0.0006517\n",
      "gmrt_in_air14: 0.0006469\n",
      "mean_acc_on_paper25: 0.0006353\n",
      "mean_gmrt24: 0.0006243\n",
      "gmrt_in_air4: 0.0006242\n",
      "mean_jerk_on_paper17: 0.0006242\n",
      "pressure_var7: 0.0006229\n",
      "max_y_extension17: 0.0006163\n",
      "pressure_mean25: 0.0006141\n",
      "mean_speed_in_air1: 0.0006067\n",
      "pressure_mean13: 0.0006011\n",
      "mean_acc_on_paper20: 0.0005907\n",
      "disp_index1: 0.00058\n",
      "mean_acc_on_paper11: 0.000573\n",
      "total_time19: 0.0005478\n",
      "gmrt_in_air21: 0.0005464\n",
      "mean_acc_in_air14: 0.0005447\n",
      "pressure_mean12: 0.0005409\n",
      "pressure_var25: 0.0005379\n",
      "gmrt_on_paper17: 0.0005368\n",
      "pressure_var22: 0.0005364\n",
      "total_time1: 0.0005363\n",
      "num_of_pendown9: 0.0005353\n",
      "gmrt_on_paper10: 0.000532\n",
      "max_x_extension10: 0.0005289\n",
      "max_y_extension21: 0.0005163\n",
      "mean_jerk_on_paper13: 0.0005118\n",
      "max_y_extension10: 0.000511\n",
      "mean_acc_on_paper19: 0.0005059\n",
      "mean_jerk_in_air4: 0.000498\n",
      "mean_acc_on_paper17: 0.0004947\n",
      "gmrt_on_paper11: 0.000494\n",
      "max_x_extension7: 0.0004896\n",
      "disp_index10: 0.000489\n",
      "air_time18: 0.0004773\n",
      "paper_time1: 0.0004637\n",
      "mean_speed_on_paper17: 0.000461\n",
      "num_of_pendown8: 0.0004562\n",
      "max_y_extension12: 0.0004424\n",
      "mean_speed_on_paper25: 0.0004379\n",
      "mean_acc_in_air23: 0.0004335\n",
      "pressure_var24: 0.0004335\n",
      "gmrt_on_paper25: 0.0004335\n",
      "mean_acc_in_air12: 0.0004322\n",
      "max_x_extension1: 0.0004317\n",
      "mean_speed_in_air10: 0.0004197\n",
      "mean_acc_in_air21: 0.0004113\n",
      "gmrt_on_paper24: 0.0004047\n",
      "mean_speed_on_paper16: 0.0004025\n",
      "mean_speed_on_paper3: 0.000398\n",
      "air_time12: 0.000395\n",
      "disp_index16: 0.000392\n",
      "num_of_pendown20: 0.0003612\n",
      "disp_index2: 0.000354\n",
      "gmrt_in_air8: 0.0003337\n",
      "paper_time2: 0.0003\n",
      "max_y_extension8: 0.0002953\n",
      "mean_acc_on_paper4: 0.0002916\n",
      "total_time18: 0.0002884\n",
      "mean_speed_on_paper24: 0.0002858\n",
      "mean_gmrt20: 0.0002855\n",
      "mean_jerk_on_paper22: 0.0002848\n",
      "gmrt_on_paper13: 0.0002843\n",
      "num_of_pendown11: 0.0002827\n",
      "mean_gmrt14: 0.0002826\n",
      "pressure_var2: 0.000282\n",
      "mean_acc_in_air1: 0.0002816\n",
      "num_of_pendown17: 0.0002808\n",
      "max_x_extension18: 0.0002799\n",
      "max_x_extension4: 0.0002746\n",
      "pressure_var9: 0.0002746\n",
      "mean_gmrt23: 0.0002746\n",
      "mean_jerk_in_air20: 0.000273\n",
      "max_y_extension1: 0.000272\n",
      "max_x_extension14: 0.0002691\n",
      "max_y_extension9: 0.0002686\n",
      "pressure_var17: 0.0002674\n",
      "mean_speed_in_air16: 0.0002641\n",
      "num_of_pendown13: 0.0002638\n",
      "air_time20: 0.0002638\n",
      "max_y_extension16: 0.0002613\n",
      "mean_acc_on_paper18: 0.0002613\n",
      "pressure_mean11: 0.0002607\n",
      "gmrt_on_paper7: 0.0002598\n",
      "mean_jerk_in_air8: 0.000259\n",
      "mean_speed_in_air4: 0.000256\n",
      "num_of_pendown18: 0.0002518\n",
      "max_x_extension8: 0.0002477\n",
      "gmrt_on_paper4: 0.0002438\n",
      "mean_acc_on_paper5: 0.0002436\n",
      "mean_jerk_on_paper3: 0.0002434\n",
      "paper_time24: 0.0002413\n",
      "pressure_var18: 0.0002404\n",
      "mean_acc_on_paper6: 0.0002398\n",
      "mean_acc_in_air8: 0.0002308\n",
      "disp_index12: 0.0002302\n",
      "max_y_extension22: 0.0002161\n",
      "mean_acc_in_air13: 0.0001955\n",
      "pressure_mean2: 0.0001941\n",
      "mean_jerk_in_air12: 0.0001941\n",
      "mean_speed_on_paper5: 0.0001919\n",
      "total_time14: 0.0001919\n",
      "mean_jerk_on_paper8: 0.0001483\n",
      "pressure_var1: 0.0001448\n",
      "mean_jerk_on_paper18: 0.0001448\n",
      "total_time25: 0.000144\n",
      "mean_acc_on_paper8: 0.0001413\n",
      "mean_acc_on_paper23: 0.0001396\n",
      "mean_jerk_on_paper10: 0.0001393\n",
      "pressure_mean14: 0.0001388\n",
      "disp_index20: 6.63e-05\n",
      "mean_acc_in_air2: 0.0\n",
      "num_of_pendown2: 0.0\n",
      "mean_acc_in_air3: 0.0\n",
      "mean_speed_in_air3: 0.0\n",
      "num_of_pendown3: 0.0\n",
      "max_y_extension4: 0.0\n",
      "mean_gmrt4: 0.0\n",
      "mean_jerk_on_paper4: 0.0\n",
      "mean_speed_on_paper4: 0.0\n",
      "paper_time4: 0.0\n",
      "mean_jerk_in_air5: 0.0\n",
      "mean_jerk_on_paper5: 0.0\n",
      "max_x_extension6: 0.0\n",
      "max_y_extension6: 0.0\n",
      "mean_acc_in_air6: 0.0\n",
      "num_of_pendown6: 0.0\n",
      "num_of_pendown7: 0.0\n",
      "mean_acc_in_air9: 0.0\n",
      "mean_acc_on_paper10: 0.0\n",
      "mean_jerk_in_air10: 0.0\n",
      "pressure_var10: 0.0\n",
      "max_y_extension11: 0.0\n",
      "mean_jerk_on_paper11: 0.0\n",
      "mean_acc_on_paper12: 0.0\n",
      "num_of_pendown12: 0.0\n",
      "mean_acc_on_paper13: 0.0\n",
      "disp_index14: 0.0\n",
      "max_y_extension14: 0.0\n",
      "mean_acc_on_paper14: 0.0\n",
      "num_of_pendown14: 0.0\n",
      "pressure_var14: 0.0\n",
      "gmrt_in_air15: 0.0\n",
      "max_y_extension15: 0.0\n",
      "mean_speed_in_air15: 0.0\n",
      "pressure_var15: 0.0\n",
      "gmrt_on_paper16: 0.0\n",
      "mean_jerk_in_air16: 0.0\n",
      "pressure_mean16: 0.0\n",
      "max_x_extension17: 0.0\n",
      "mean_acc_in_air17: 0.0\n",
      "gmrt_on_paper18: 0.0\n",
      "max_y_extension18: 0.0\n",
      "mean_speed_on_paper18: 0.0\n",
      "mean_speed_on_paper19: 0.0\n",
      "paper_time19: 0.0\n",
      "mean_speed_on_paper20: 0.0\n",
      "mean_jerk_in_air21: 0.0\n",
      "mean_speed_in_air21: 0.0\n",
      "num_of_pendown21: 0.0\n",
      "paper_time21: 0.0\n",
      "total_time21: 0.0\n",
      "gmrt_in_air22: 0.0\n",
      "num_of_pendown23: 0.0\n",
      "gmrt_in_air24: 0.0\n",
      "max_x_extension24: 0.0\n",
      "num_of_pendown24: 0.0\n",
      "mean_acc_in_air25: 0.0\n",
      "num_of_pendown25: 0.0\n"
     ]
    }
   ],
   "source": [
    "importances = list(rf.feature_importances_)\n",
    "features = X.columns.tolist()\n",
    "feature_importances = [(feature, round(importance, 7)) for feature, importance in zip(features, importances)]\n",
    "feature_importances = sorted(feature_importances, key=lambda x: x[1], reverse=True)\n",
    "\n",
    "print(\"\\nFeature importance rank:\")\n",
    "for feature, importance in feature_importances:\n",
    "    print(f\"{feature}: {importance}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "42a6d592",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:55.152261Z",
     "iopub.status.busy": "2024-09-21T18:25:55.151861Z",
     "iopub.status.idle": "2024-09-21T18:25:55.164246Z",
     "shell.execute_reply": "2024-09-21T18:25:55.163374Z"
    },
    "papermill": {
     "duration": 0.039388,
     "end_time": "2024-09-21T18:25:55.167122",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.127734",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_task_feature_importances(task_name):\n",
    "    # Filter out important features for the selected task\n",
    "    task_feature_importances = [(feature, importance) for feature, importance in feature_importances \n",
    "                                if feature in features_by_task[task_name] and importance > 0]\n",
    "\n",
    "    # Sort the feature importances\n",
    "    task_feature_importances = sorted(task_feature_importances, key=lambda x: x[1], reverse=True)\n",
    "    \n",
    "    if len(task_feature_importances) == 0:\n",
    "        print(f\"No important features for task {task_name}.\")\n",
    "        return\n",
    "    \n",
    "    # Separate features and their importances\n",
    "    features, importances = zip(*task_feature_importances)\n",
    "    \n",
    "    # Ensure the importances are numeric and features are strings\n",
    "    importances = [float(imp) for imp in importances]  # Convert to float\n",
    "    features = [str(feat) for feat in features]  # Ensure features are strings\n",
    "    \n",
    "    # Create the bar plot\n",
    "    plt.figure(figsize=(12, 8))\n",
    "    sns.barplot(x=importances, y=features, palette=\"viridis\")\n",
    "    plt.title(f'Feature Importance for Task: {task_name}')\n",
    "    plt.xlabel('Importance Score')\n",
    "    plt.ylabel('Features')\n",
    "    plt.show()\n",
    "\n",
    "# Task dropdown for interactive plotting\n",
    "task_selector = Dropdown(\n",
    "    options=list(features_by_task.keys()),\n",
    "    value=list(features_by_task.keys())[0],\n",
    "    description='Task:',\n",
    ")\n",
    "\n",
    "# Eliminate # to run the code(interactive plot) \n",
    "#interactive_plot = interactive(plot_task_feature_importances, task_name=task_selector)\n",
    "\n",
    "# Display the interactive widget\n",
    "#display(interactive_plot)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "caaa96e3",
   "metadata": {
    "papermill": {
     "duration": 0.022563,
     "end_time": "2024-09-21T18:25:55.212555",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.189992",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# PART II : Feature Engineering "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48560f01",
   "metadata": {
    "papermill": {
     "duration": 0.022555,
     "end_time": "2024-09-21T18:25:55.258564",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.236009",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Log Transformation for Skewed Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "0d11470c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:55.309287Z",
     "iopub.status.busy": "2024-09-21T18:25:55.308851Z",
     "iopub.status.idle": "2024-09-21T18:25:55.316067Z",
     "shell.execute_reply": "2024-09-21T18:25:55.315057Z"
    },
    "papermill": {
     "duration": 0.03733,
     "end_time": "2024-09-21T18:25:55.318342",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.281012",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "X = data.drop(['class'], axis=1) \n",
    "y = data['class'] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "cc79a458",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:55.385286Z",
     "iopub.status.busy": "2024-09-21T18:25:55.384931Z",
     "iopub.status.idle": "2024-09-21T18:25:55.707147Z",
     "shell.execute_reply": "2024-09-21T18:25:55.706094Z"
    },
    "papermill": {
     "duration": 0.357563,
     "end_time": "2024-09-21T18:25:55.710229",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.352666",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_log_transformed = X.copy()\n",
    "\n",
    "for i in skewed_features:\n",
    "    data_log_transformed.iloc[:, i] = np.log1p(data.iloc[:, i])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11895690",
   "metadata": {
    "papermill": {
     "duration": 0.027805,
     "end_time": "2024-09-21T18:25:55.769737",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.741932",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Standardization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "76a047da",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:55.828920Z",
     "iopub.status.busy": "2024-09-21T18:25:55.827726Z",
     "iopub.status.idle": "2024-09-21T18:25:55.855876Z",
     "shell.execute_reply": "2024-09-21T18:25:55.854962Z"
    },
    "papermill": {
     "duration": 0.055871,
     "end_time": "2024-09-21T18:25:55.858146",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.802275",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(data_log_transformed)\n",
    "\n",
    "X_scaled = pd.DataFrame(X_scaled, columns=X.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95debd8e",
   "metadata": {
    "papermill": {
     "duration": 0.022447,
     "end_time": "2024-09-21T18:25:55.907763",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.885316",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### RandomForestClassifier : TOP 10 IMPORTANT FEATURES PER TASK"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "7c73a9ab",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:55.958333Z",
     "iopub.status.busy": "2024-09-21T18:25:55.957973Z",
     "iopub.status.idle": "2024-09-21T18:25:56.247224Z",
     "shell.execute_reply": "2024-09-21T18:25:56.246200Z"
    },
    "papermill": {
     "duration": 0.316973,
     "end_time": "2024-09-21T18:25:56.249575",
     "exception": false,
     "start_time": "2024-09-21T18:25:55.932602",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#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(random_state=42)</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(random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomForestClassifier(random_state=42)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.2, random_state=42)\n",
    "\n",
    "rf = RandomForestClassifier(random_state=42)\n",
    "rf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "5fe17401",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:56.298150Z",
     "iopub.status.busy": "2024-09-21T18:25:56.297800Z",
     "iopub.status.idle": "2024-09-21T18:25:56.346324Z",
     "shell.execute_reply": "2024-09-21T18:25:56.345307Z"
    },
    "papermill": {
     "duration": 0.074492,
     "end_time": "2024-09-21T18:25:56.348509",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.274017",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final dataset with top features from each task: (174, 250)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>air_time1</th>\n",
       "      <th>gmrt_on_paper1</th>\n",
       "      <th>mean_gmrt1</th>\n",
       "      <th>pressure_mean1</th>\n",
       "      <th>mean_jerk_in_air1</th>\n",
       "      <th>mean_speed_on_paper1</th>\n",
       "      <th>mean_jerk_on_paper1</th>\n",
       "      <th>num_of_pendown1</th>\n",
       "      <th>mean_acc_on_paper1</th>\n",
       "      <th>mean_acc_in_air1</th>\n",
       "      <th>...</th>\n",
       "      <th>gmrt_in_air25</th>\n",
       "      <th>paper_time25</th>\n",
       "      <th>max_x_extension25</th>\n",
       "      <th>max_y_extension25</th>\n",
       "      <th>mean_gmrt25</th>\n",
       "      <th>mean_jerk_on_paper25</th>\n",
       "      <th>air_time25</th>\n",
       "      <th>mean_speed_in_air25</th>\n",
       "      <th>mean_jerk_in_air25</th>\n",
       "      <th>disp_index25</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.576753</td>\n",
       "      <td>-1.467001</td>\n",
       "      <td>-1.530934</td>\n",
       "      <td>0.003164</td>\n",
       "      <td>-0.202487</td>\n",
       "      <td>-1.467930</td>\n",
       "      <td>0.234572</td>\n",
       "      <td>1.861746</td>\n",
       "      <td>0.646752</td>\n",
       "      <td>-0.071346</td>\n",
       "      <td>...</td>\n",
       "      <td>0.602209</td>\n",
       "      <td>-0.009481</td>\n",
       "      <td>0.978273</td>\n",
       "      <td>-1.291872</td>\n",
       "      <td>0.446195</td>\n",
       "      <td>1.904832</td>\n",
       "      <td>0.809052</td>\n",
       "      <td>0.810916</td>\n",
       "      <td>-0.085634</td>\n",
       "      <td>-0.540180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.852083</td>\n",
       "      <td>-1.548665</td>\n",
       "      <td>-1.619429</td>\n",
       "      <td>0.161252</td>\n",
       "      <td>-0.385295</td>\n",
       "      <td>-1.442870</td>\n",
       "      <td>-0.811943</td>\n",
       "      <td>0.631717</td>\n",
       "      <td>-0.567139</td>\n",
       "      <td>-0.403578</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.641887</td>\n",
       "      <td>3.311314</td>\n",
       "      <td>-0.080431</td>\n",
       "      <td>-0.333294</td>\n",
       "      <td>-2.266524</td>\n",
       "      <td>-0.657524</td>\n",
       "      <td>1.300247</td>\n",
       "      <td>-2.467765</td>\n",
       "      <td>-1.660979</td>\n",
       "      <td>1.560246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.098271</td>\n",
       "      <td>-0.057857</td>\n",
       "      <td>-0.187773</td>\n",
       "      <td>-0.568823</td>\n",
       "      <td>-0.016158</td>\n",
       "      <td>-0.067084</td>\n",
       "      <td>-0.083839</td>\n",
       "      <td>0.467210</td>\n",
       "      <td>0.052018</td>\n",
       "      <td>0.018553</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.118835</td>\n",
       "      <td>0.352637</td>\n",
       "      <td>0.065014</td>\n",
       "      <td>-0.874162</td>\n",
       "      <td>-0.435581</td>\n",
       "      <td>-1.160752</td>\n",
       "      <td>-0.311338</td>\n",
       "      <td>-0.191336</td>\n",
       "      <td>0.510130</td>\n",
       "      <td>0.132375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.294606</td>\n",
       "      <td>0.062597</td>\n",
       "      <td>0.455776</td>\n",
       "      <td>0.847815</td>\n",
       "      <td>0.370817</td>\n",
       "      <td>0.068201</td>\n",
       "      <td>0.145587</td>\n",
       "      <td>0.467210</td>\n",
       "      <td>-0.231527</td>\n",
       "      <td>0.584534</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.220943</td>\n",
       "      <td>1.511863</td>\n",
       "      <td>-0.557915</td>\n",
       "      <td>-0.506091</td>\n",
       "      <td>-0.916895</td>\n",
       "      <td>-0.030022</td>\n",
       "      <td>0.888706</td>\n",
       "      <td>-0.045270</td>\n",
       "      <td>-0.544496</td>\n",
       "      <td>0.328967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.214725</td>\n",
       "      <td>-0.960197</td>\n",
       "      <td>-0.363818</td>\n",
       "      <td>0.506469</td>\n",
       "      <td>-0.420519</td>\n",
       "      <td>-0.844521</td>\n",
       "      <td>-0.428554</td>\n",
       "      <td>0.087814</td>\n",
       "      <td>-0.563269</td>\n",
       "      <td>-0.428421</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.812254</td>\n",
       "      <td>-0.221106</td>\n",
       "      <td>-1.657699</td>\n",
       "      <td>-0.216223</td>\n",
       "      <td>-1.168098</td>\n",
       "      <td>0.395630</td>\n",
       "      <td>-0.261330</td>\n",
       "      <td>-0.721741</td>\n",
       "      <td>-0.412053</td>\n",
       "      <td>-1.264475</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 250 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   air_time1  gmrt_on_paper1  mean_gmrt1  pressure_mean1  mean_jerk_in_air1  \\\n",
       "0   0.576753       -1.467001   -1.530934        0.003164          -0.202487   \n",
       "1   2.852083       -1.548665   -1.619429        0.161252          -0.385295   \n",
       "2  -0.098271       -0.057857   -0.187773       -0.568823          -0.016158   \n",
       "3  -0.294606        0.062597    0.455776        0.847815           0.370817   \n",
       "4  -0.214725       -0.960197   -0.363818        0.506469          -0.420519   \n",
       "\n",
       "   mean_speed_on_paper1  mean_jerk_on_paper1  num_of_pendown1  \\\n",
       "0             -1.467930             0.234572         1.861746   \n",
       "1             -1.442870            -0.811943         0.631717   \n",
       "2             -0.067084            -0.083839         0.467210   \n",
       "3              0.068201             0.145587         0.467210   \n",
       "4             -0.844521            -0.428554         0.087814   \n",
       "\n",
       "   mean_acc_on_paper1  mean_acc_in_air1  ...  gmrt_in_air25  paper_time25  \\\n",
       "0            0.646752         -0.071346  ...       0.602209     -0.009481   \n",
       "1           -0.567139         -0.403578  ...      -2.641887      3.311314   \n",
       "2            0.052018          0.018553  ...      -0.118835      0.352637   \n",
       "3           -0.231527          0.584534  ...      -0.220943      1.511863   \n",
       "4           -0.563269         -0.428421  ...      -0.812254     -0.221106   \n",
       "\n",
       "   max_x_extension25  max_y_extension25  mean_gmrt25  mean_jerk_on_paper25  \\\n",
       "0           0.978273          -1.291872     0.446195              1.904832   \n",
       "1          -0.080431          -0.333294    -2.266524             -0.657524   \n",
       "2           0.065014          -0.874162    -0.435581             -1.160752   \n",
       "3          -0.557915          -0.506091    -0.916895             -0.030022   \n",
       "4          -1.657699          -0.216223    -1.168098              0.395630   \n",
       "\n",
       "   air_time25  mean_speed_in_air25  mean_jerk_in_air25  disp_index25  \n",
       "0    0.809052             0.810916           -0.085634     -0.540180  \n",
       "1    1.300247            -2.467765           -1.660979      1.560246  \n",
       "2   -0.311338            -0.191336            0.510130      0.132375  \n",
       "3    0.888706            -0.045270           -0.544496      0.328967  \n",
       "4   -0.261330            -0.721741           -0.412053     -1.264475  \n",
       "\n",
       "[5 rows x 250 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "importances = rf.feature_importances_\n",
    "features = X.columns.tolist()\n",
    "feature_importances = [(feature, round(importance, 7)) for feature, importance in zip(features, importances)]\n",
    "feature_importances = sorted(feature_importances, key=lambda x: x[1], reverse=True)\n",
    "\n",
    "features_by_task = defaultdict(list)\n",
    "\n",
    "for feature in X.columns:\n",
    "    match = re.search(r'(\\d+)$', feature)  \n",
    "    if match:\n",
    "        task_num = match.group(1)\n",
    "        features_by_task[task_num].append(feature)\n",
    "\n",
    "top_features_by_task = {}\n",
    "for task, task_features in features_by_task.items():\n",
    "    task_importances = [(feature, imp) for feature, imp in feature_importances if feature in task_features]\n",
    "    top_features_by_task[task] = sorted(task_importances, key=lambda x: x[1], reverse=True)[:10]\n",
    "\n",
    "top_features = [feature for task, features in top_features_by_task.items() for feature, _ in features]\n",
    "X_top_features = X_scaled[top_features]\n",
    "\n",
    "\n",
    "print(f\"Final dataset with top features from each task: {X_top_features.shape}\")\n",
    "X_top_features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "f185c859",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:56.398965Z",
     "iopub.status.busy": "2024-09-21T18:25:56.398628Z",
     "iopub.status.idle": "2024-09-21T18:25:56.407442Z",
     "shell.execute_reply": "2024-09-21T18:25:56.406487Z"
    },
    "papermill": {
     "duration": 0.036213,
     "end_time": "2024-09-21T18:25:56.410146",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.373933",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Number of features: Task Number N1: 10\n",
      "\n",
      "Number of features: Task Number N2: 10\n",
      "\n",
      "Number of features: Task Number N3: 10\n",
      "\n",
      "Number of features: Task Number N4: 10\n",
      "\n",
      "Number of features: Task Number N5: 10\n",
      "\n",
      "Number of features: Task Number N6: 10\n",
      "\n",
      "Number of features: Task Number N7: 10\n",
      "\n",
      "Number of features: Task Number N8: 10\n",
      "\n",
      "Number of features: Task Number N9: 10\n",
      "\n",
      "Number of features: Task Number N10: 10\n",
      "\n",
      "Number of features: Task Number N11: 10\n",
      "\n",
      "Number of features: Task Number N12: 10\n",
      "\n",
      "Number of features: Task Number N13: 10\n",
      "\n",
      "Number of features: Task Number N14: 10\n",
      "\n",
      "Number of features: Task Number N15: 10\n",
      "\n",
      "Number of features: Task Number N16: 10\n",
      "\n",
      "Number of features: Task Number N17: 10\n",
      "\n",
      "Number of features: Task Number N18: 10\n",
      "\n",
      "Number of features: Task Number N19: 10\n",
      "\n",
      "Number of features: Task Number N20: 10\n",
      "\n",
      "Number of features: Task Number N21: 10\n",
      "\n",
      "Number of features: Task Number N22: 10\n",
      "\n",
      "Number of features: Task Number N23: 10\n",
      "\n",
      "Number of features: Task Number N24: 10\n",
      "\n",
      "Number of features: Task Number N25: 10\n"
     ]
    }
   ],
   "source": [
    "features_by_task = defaultdict(list)\n",
    "for col in X_top_features.columns:\n",
    "    match = re.search(r'(\\d+)$', col)  \n",
    "    if match:\n",
    "        x_value = match.group(1)  \n",
    "        features_by_task[x_value].append(col)   \n",
    "\n",
    "\n",
    "for x_value, features in features_by_task.items():\n",
    "    print(f\"\\nNumber of features: Task Number N{x_value}: {len(features)}\")\n",
    "df_features_by_task = pd.DataFrame([{'X Value': x_value, 'Features': features}\n",
    "    for x_value, features in features_by_task.items()])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8da68d95",
   "metadata": {
    "papermill": {
     "duration": 0.024529,
     "end_time": "2024-09-21T18:25:56.459715",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.435186",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Dimensionality Reduction using PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "1e2dc62e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:56.510607Z",
     "iopub.status.busy": "2024-09-21T18:25:56.510259Z",
     "iopub.status.idle": "2024-09-21T18:25:56.704625Z",
     "shell.execute_reply": "2024-09-21T18:25:56.703736Z"
    },
    "papermill": {
     "duration": 0.222555,
     "end_time": "2024-09-21T18:25:56.707233",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.484678",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PCA reduced the feature set to 64 components.\n"
     ]
    }
   ],
   "source": [
    "pca = PCA(n_components=0.9)  # Retain 90% of teh variance\n",
    "X_pca = pca.fit_transform(X_top_features)\n",
    "\n",
    "print(f\"PCA reduced the feature set to {X_pca.shape[1]} components.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c86c46ef",
   "metadata": {
    "papermill": {
     "duration": 0.031071,
     "end_time": "2024-09-21T18:25:56.770585",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.739514",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Save the Transformed Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "da4419f3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:56.834782Z",
     "iopub.status.busy": "2024-09-21T18:25:56.834428Z",
     "iopub.status.idle": "2024-09-21T18:25:56.865766Z",
     "shell.execute_reply": "2024-09-21T18:25:56.864845Z"
    },
    "papermill": {
     "duration": 0.065707,
     "end_time": "2024-09-21T18:25:56.867611",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.801904",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transformed data saved to 'transformed_data.csv'.\n"
     ]
    }
   ],
   "source": [
    "transformed_data = pd.DataFrame(X_pca)\n",
    "transformed_data['Class'] = y.values  \n",
    "\n",
    "transformed_data.to_csv('transformed_data.csv', index=False)\n",
    "\n",
    "print(\"Transformed data saved to 'transformed_data.csv'.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "b98ea1e3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:56.917300Z",
     "iopub.status.busy": "2024-09-21T18:25:56.916701Z",
     "iopub.status.idle": "2024-09-21T18:25:56.921237Z",
     "shell.execute_reply": "2024-09-21T18:25:56.920396Z"
    },
    "papermill": {
     "duration": 0.030881,
     "end_time": "2024-09-21T18:25:56.923391",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.892510",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape trasnformed from (174, 452) TO (174, 65)\n"
     ]
    }
   ],
   "source": [
    "print(f\"Shape trasnformed from (174, 452) TO {transformed_data.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90ea3b9b",
   "metadata": {
    "papermill": {
     "duration": 0.023708,
     "end_time": "2024-09-21T18:25:56.972878",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.949170",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# PART III : Models Training and Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "ae589b0f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:57.020761Z",
     "iopub.status.busy": "2024-09-21T18:25:57.020457Z",
     "iopub.status.idle": "2024-09-21T18:25:57.034237Z",
     "shell.execute_reply": "2024-09-21T18:25:57.033514Z"
    },
    "papermill": {
     "duration": 0.039962,
     "end_time": "2024-09-21T18:25:57.036154",
     "exception": false,
     "start_time": "2024-09-21T18:25:56.996192",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "data = pd.read_csv('/kaggle/working/transformed_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "29b49afc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:57.085049Z",
     "iopub.status.busy": "2024-09-21T18:25:57.084723Z",
     "iopub.status.idle": "2024-09-21T18:25:57.107642Z",
     "shell.execute_reply": "2024-09-21T18:25:57.106760Z"
    },
    "papermill": {
     "duration": 0.050174,
     "end_time": "2024-09-21T18:25:57.109758",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.059584",
     "status": "completed"
    },
    "tags": []
   },
   "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>55</th>\n",
       "      <th>56</th>\n",
       "      <th>57</th>\n",
       "      <th>58</th>\n",
       "      <th>59</th>\n",
       "      <th>60</th>\n",
       "      <th>61</th>\n",
       "      <th>62</th>\n",
       "      <th>63</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.709241</td>\n",
       "      <td>2.054359</td>\n",
       "      <td>1.795539</td>\n",
       "      <td>-5.193219</td>\n",
       "      <td>1.590595</td>\n",
       "      <td>-3.336288</td>\n",
       "      <td>-1.690712</td>\n",
       "      <td>-2.567750</td>\n",
       "      <td>-1.256762</td>\n",
       "      <td>-3.714133</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.503506</td>\n",
       "      <td>0.599565</td>\n",
       "      <td>0.565926</td>\n",
       "      <td>-0.091839</td>\n",
       "      <td>-1.737411</td>\n",
       "      <td>-0.636425</td>\n",
       "      <td>0.940795</td>\n",
       "      <td>-0.370968</td>\n",
       "      <td>-0.626035</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>16.529104</td>\n",
       "      <td>0.443856</td>\n",
       "      <td>1.122190</td>\n",
       "      <td>-1.967386</td>\n",
       "      <td>-1.603015</td>\n",
       "      <td>-1.879084</td>\n",
       "      <td>0.333950</td>\n",
       "      <td>0.486657</td>\n",
       "      <td>-5.408185</td>\n",
       "      <td>0.309898</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.838162</td>\n",
       "      <td>-0.404962</td>\n",
       "      <td>-0.104504</td>\n",
       "      <td>-2.377410</td>\n",
       "      <td>-1.958903</td>\n",
       "      <td>0.686784</td>\n",
       "      <td>1.008078</td>\n",
       "      <td>-0.083973</td>\n",
       "      <td>0.107189</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.875827</td>\n",
       "      <td>2.439685</td>\n",
       "      <td>-1.116889</td>\n",
       "      <td>-3.022485</td>\n",
       "      <td>-4.075029</td>\n",
       "      <td>-2.063375</td>\n",
       "      <td>1.424632</td>\n",
       "      <td>-0.926506</td>\n",
       "      <td>2.173161</td>\n",
       "      <td>2.667305</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.932707</td>\n",
       "      <td>-1.430546</td>\n",
       "      <td>0.271952</td>\n",
       "      <td>0.529939</td>\n",
       "      <td>1.798347</td>\n",
       "      <td>1.280374</td>\n",
       "      <td>-0.097461</td>\n",
       "      <td>0.351761</td>\n",
       "      <td>2.120903</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7.379832</td>\n",
       "      <td>-2.111653</td>\n",
       "      <td>0.316131</td>\n",
       "      <td>-3.849107</td>\n",
       "      <td>2.911028</td>\n",
       "      <td>1.935967</td>\n",
       "      <td>-0.891683</td>\n",
       "      <td>4.633672</td>\n",
       "      <td>-3.390017</td>\n",
       "      <td>-1.366942</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.780635</td>\n",
       "      <td>0.386296</td>\n",
       "      <td>-0.199818</td>\n",
       "      <td>-0.188223</td>\n",
       "      <td>1.255045</td>\n",
       "      <td>1.358196</td>\n",
       "      <td>0.179196</td>\n",
       "      <td>0.394425</td>\n",
       "      <td>0.073721</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.667963</td>\n",
       "      <td>2.139225</td>\n",
       "      <td>-1.170553</td>\n",
       "      <td>-7.324933</td>\n",
       "      <td>0.049968</td>\n",
       "      <td>1.113142</td>\n",
       "      <td>3.625093</td>\n",
       "      <td>-0.442725</td>\n",
       "      <td>-0.653732</td>\n",
       "      <td>-1.052910</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.055535</td>\n",
       "      <td>-0.097720</td>\n",
       "      <td>-1.540164</td>\n",
       "      <td>1.286678</td>\n",
       "      <td>0.908733</td>\n",
       "      <td>-0.174959</td>\n",
       "      <td>1.358588</td>\n",
       "      <td>-0.746898</td>\n",
       "      <td>-0.465557</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           0         1         2         3         4         5         6  \\\n",
       "0   2.709241  2.054359  1.795539 -5.193219  1.590595 -3.336288 -1.690712   \n",
       "1  16.529104  0.443856  1.122190 -1.967386 -1.603015 -1.879084  0.333950   \n",
       "2  -0.875827  2.439685 -1.116889 -3.022485 -4.075029 -2.063375  1.424632   \n",
       "3   7.379832 -2.111653  0.316131 -3.849107  2.911028  1.935967 -0.891683   \n",
       "4   2.667963  2.139225 -1.170553 -7.324933  0.049968  1.113142  3.625093   \n",
       "\n",
       "          7         8         9  ...        55        56        57        58  \\\n",
       "0 -2.567750 -1.256762 -3.714133  ... -1.503506  0.599565  0.565926 -0.091839   \n",
       "1  0.486657 -5.408185  0.309898  ... -0.838162 -0.404962 -0.104504 -2.377410   \n",
       "2 -0.926506  2.173161  2.667305  ... -1.932707 -1.430546  0.271952  0.529939   \n",
       "3  4.633672 -3.390017 -1.366942  ... -0.780635  0.386296 -0.199818 -0.188223   \n",
       "4 -0.442725 -0.653732 -1.052910  ... -1.055535 -0.097720 -1.540164  1.286678   \n",
       "\n",
       "         59        60        61        62        63  Class  \n",
       "0 -1.737411 -0.636425  0.940795 -0.370968 -0.626035      1  \n",
       "1 -1.958903  0.686784  1.008078 -0.083973  0.107189      1  \n",
       "2  1.798347  1.280374 -0.097461  0.351761  2.120903      1  \n",
       "3  1.255045  1.358196  0.179196  0.394425  0.073721      1  \n",
       "4  0.908733 -0.174959  1.358588 -0.746898 -0.465557      1  \n",
       "\n",
       "[5 rows x 65 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "37d63780",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:57.163074Z",
     "iopub.status.busy": "2024-09-21T18:25:57.162769Z",
     "iopub.status.idle": "2024-09-21T18:25:57.170240Z",
     "shell.execute_reply": "2024-09-21T18:25:57.169428Z"
    },
    "papermill": {
     "duration": 0.035261,
     "end_time": "2024-09-21T18:25:57.172369",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.137108",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Split the dataset into features and target\n",
    "y = data['Class']\n",
    "X = data.drop(columns=['Class'])\n",
    "\n",
    "# Train-test split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "bea422b6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:57.222662Z",
     "iopub.status.busy": "2024-09-21T18:25:57.222031Z",
     "iopub.status.idle": "2024-09-21T18:25:57.227491Z",
     "shell.execute_reply": "2024-09-21T18:25:57.226677Z"
    },
    "papermill": {
     "duration": 0.032335,
     "end_time": "2024-09-21T18:25:57.229455",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.197120",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(174, 65)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "501cf160",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:57.278719Z",
     "iopub.status.busy": "2024-09-21T18:25:57.278448Z",
     "iopub.status.idle": "2024-09-21T18:25:57.283549Z",
     "shell.execute_reply": "2024-09-21T18:25:57.282559Z"
    },
    "papermill": {
     "duration": 0.032106,
     "end_time": "2024-09-21T18:25:57.285540",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.253434",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training samples: 156, Test samples: 18\n",
      "Number of features: 64\n"
     ]
    }
   ],
   "source": [
    "print(f\"Training samples: {X_train.shape[0]}, Test samples: {X_test.shape[0]}\")\n",
    "print(f\"Number of features: {X_train.shape[1]}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd78796e",
   "metadata": {
    "papermill": {
     "duration": 0.023794,
     "end_time": "2024-09-21T18:25:57.333211",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.309417",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "fd790111",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:57.383458Z",
     "iopub.status.busy": "2024-09-21T18:25:57.383157Z",
     "iopub.status.idle": "2024-09-21T18:25:57.391258Z",
     "shell.execute_reply": "2024-09-21T18:25:57.390491Z"
    },
    "papermill": {
     "duration": 0.035739,
     "end_time": "2024-09-21T18:25:57.393191",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.357452",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Leave-One-Out CV\n",
    "loocv = LeaveOneOut()\n",
    "\n",
    "# Models\n",
    "catboost = CatBoostClassifier(learning_rate=0.1, depth=6, iterations=100, verbose=0, task_type='GPU')\n",
    "svm = SVC(probability=True)\n",
    "logreg = LogisticRegression(max_iter=1000)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c046cca",
   "metadata": {
    "papermill": {
     "duration": 0.024143,
     "end_time": "2024-09-21T18:25:57.441666",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.417523",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Grid Search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "e5ba8150",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:57.491514Z",
     "iopub.status.busy": "2024-09-21T18:25:57.490925Z",
     "iopub.status.idle": "2024-09-21T18:25:57.496066Z",
     "shell.execute_reply": "2024-09-21T18:25:57.495262Z"
    },
    "papermill": {
     "duration": 0.032288,
     "end_time": "2024-09-21T18:25:57.497866",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.465578",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "param_grid_catboost = {\n",
    "    'depth': [6, 8, 10],\n",
    "    'learning_rate': [0.01, 0.1, 0.2],\n",
    "    'iterations': [100, 200, 300]\n",
    "}\n",
    "\n",
    "param_grid_svm = {\n",
    "    'C': [0.1, 1, 10],\n",
    "    'kernel': ['linear', 'rbf']\n",
    "}\n",
    "\n",
    "param_grid_logreg = {\n",
    "    'C': [0.1, 1, 10],\n",
    "    'penalty': ['l2']\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "0401d62b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:25:57.546654Z",
     "iopub.status.busy": "2024-09-21T18:25:57.546392Z",
     "iopub.status.idle": "2024-09-21T18:30:37.879141Z",
     "shell.execute_reply": "2024-09-21T18:30:37.877808Z"
    },
    "papermill": {
     "duration": 280.371472,
     "end_time": "2024-09-21T18:30:37.893218",
     "exception": false,
     "start_time": "2024-09-21T18:25:57.521746",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters for CatBoost: {'depth': 6, 'iterations': 200, 'learning_rate': 0.01}\n",
      "Best parameters for SVM: {'C': 1, 'kernel': 'rbf'}\n",
      "Best parameters for Logistic Regression: {'C': 0.1, 'penalty': 'l2'}\n"
     ]
    }
   ],
   "source": [
    "# Grid Search for Catboost\n",
    "grid_search_catboost = GridSearchCV(estimator=catboost, param_grid=param_grid_catboost, cv=3, scoring='roc_auc')\n",
    "grid_search_catboost.fit(X, y)\n",
    "print(f\"Best parameters for CatBoost: {grid_search_catboost.best_params_}\")\n",
    "\n",
    "# Grid Search for SVM\n",
    "grid_search_svm = GridSearchCV(estimator=svm, param_grid=param_grid_svm, cv=3, scoring='roc_auc')\n",
    "grid_search_svm.fit(X, y)\n",
    "print(f\"Best parameters for SVM: {grid_search_svm.best_params_}\")\n",
    "\n",
    "# Grid Search for LR\n",
    "grid_search_logreg = GridSearchCV(estimator=logreg, param_grid=param_grid_logreg, cv=3, scoring='roc_auc')\n",
    "grid_search_logreg.fit(X, y)\n",
    "print(f\"Best parameters for Logistic Regression: {grid_search_logreg.best_params_}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6865778a",
   "metadata": {
    "papermill": {
     "duration": 0.038427,
     "end_time": "2024-09-21T18:30:38.045157",
     "exception": false,
     "start_time": "2024-09-21T18:30:38.006730",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Fitting using best_params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "65588608",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:30:38.122450Z",
     "iopub.status.busy": "2024-09-21T18:30:38.121971Z",
     "iopub.status.idle": "2024-09-21T18:30:38.131652Z",
     "shell.execute_reply": "2024-09-21T18:30:38.130855Z"
    },
    "papermill": {
     "duration": 0.05062,
     "end_time": "2024-09-21T18:30:38.133894",
     "exception": false,
     "start_time": "2024-09-21T18:30:38.083274",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def evaluate_model(model, model_name):\n",
    "    y_true = []\n",
    "    y_pred_proba_all = []\n",
    "\n",
    "    for train_index, test_index in loocv.split(X):\n",
    "        X_train_loocv, X_test_loocv = X.iloc[train_index], X.iloc[test_index]\n",
    "        y_train_loocv, y_test_loocv = y.iloc[train_index], y.iloc[test_index]\n",
    "\n",
    "        model.fit(X_train_loocv, y_train_loocv)\n",
    "\n",
    "        # Predict probabilities for the left-out sample\n",
    "        y_pred_proba = model.predict_proba(X_test_loocv)[:, 1]\n",
    "\n",
    "        y_true.append(y_test_loocv.values[0])\n",
    "        y_pred_proba_all.append(y_pred_proba[0])\n",
    "\n",
    "    auc_score = roc_auc_score(y_true, y_pred_proba_all)\n",
    "    print(f\"{model_name} ROC AUC Score: {auc_score}\")\n",
    "\n",
    "    fpr, tpr, _ = roc_curve(y_true, y_pred_proba_all)\n",
    "    return fpr, tpr, auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "9ffa8298",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:30:38.188645Z",
     "iopub.status.busy": "2024-09-21T18:30:38.188295Z",
     "iopub.status.idle": "2024-09-21T18:36:24.641459Z",
     "shell.execute_reply": "2024-09-21T18:36:24.640113Z"
    },
    "papermill": {
     "duration": 346.484701,
     "end_time": "2024-09-21T18:36:24.646094",
     "exception": false,
     "start_time": "2024-09-21T18:30:38.161393",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CatBoost ROC AUC Score: 0.9271645736946464\n",
      "SVM ROC AUC Score: 0.9668208856576338\n",
      "Logistic Regression ROC AUC Score: 0.9422339722405816\n"
     ]
    }
   ],
   "source": [
    "roc_curves = {}\n",
    "roc_auc_scores = {}\n",
    "\n",
    "fpr, tpr, auc_score = evaluate_model(grid_search_catboost.best_estimator_, 'CatBoost')\n",
    "roc_curves['CatBoost'] = (fpr, tpr)\n",
    "roc_auc_scores['CatBoost'] = auc_score\n",
    "\n",
    "\n",
    "fpr, tpr, auc_score = evaluate_model(grid_search_svm.best_estimator_, 'SVM')\n",
    "roc_curves['SVM'] = (fpr, tpr)\n",
    "roc_auc_scores['SVM'] = auc_score\n",
    "\n",
    "fpr, tpr, auc_score = evaluate_model(grid_search_logreg.best_estimator_, 'Logistic Regression')\n",
    "roc_curves['Logistic Regression'] = (fpr, tpr)\n",
    "roc_auc_scores['Logistic Regression'] = auc_score"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "223dd3fb",
   "metadata": {
    "papermill": {
     "duration": 0.052257,
     "end_time": "2024-09-21T18:36:24.756436",
     "exception": false,
     "start_time": "2024-09-21T18:36:24.704179",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Custom Neural Network Model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "cca8162e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:24.812605Z",
     "iopub.status.busy": "2024-09-21T18:36:24.812251Z",
     "iopub.status.idle": "2024-09-21T18:36:24.953389Z",
     "shell.execute_reply": "2024-09-21T18:36:24.952624Z"
    },
    "papermill": {
     "duration": 0.171085,
     "end_time": "2024-09-21T18:36:24.955505",
     "exception": false,
     "start_time": "2024-09-21T18:36:24.784420",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "# Convert to tensors and move to GPU\n",
    "X_train_tensor = torch.tensor(X_train.values, dtype=torch.float32).to(device)\n",
    "y_train_tensor = torch.tensor(y_train.values, dtype=torch.float32).to(device)\n",
    "X_test_tensor = torch.tensor(X_test.values, dtype=torch.float32).to(device)\n",
    "y_test_tensor = torch.tensor(y_test.values, dtype=torch.float32).to(device)\n",
    "\n",
    "train_dataset = TensorDataset(X_train_tensor, y_train_tensor)\n",
    "test_dataset = TensorDataset(X_test_tensor, y_test_tensor)\n",
    "\n",
    "batch_size = 128\n",
    "train_loader = DataLoader(dataset=train_dataset, batch_size=batch_size, shuffle=True)\n",
    "test_loader = DataLoader(dataset=test_dataset, batch_size=batch_size, shuffle=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "59a51c05",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:25.006499Z",
     "iopub.status.busy": "2024-09-21T18:36:25.006183Z",
     "iopub.status.idle": "2024-09-21T18:36:25.032754Z",
     "shell.execute_reply": "2024-09-21T18:36:25.031765Z"
    },
    "papermill": {
     "duration": 0.0541,
     "end_time": "2024-09-21T18:36:25.034760",
     "exception": false,
     "start_time": "2024-09-21T18:36:24.980660",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Network(\n",
      "  (fc1): Linear(in_features=64, out_features=16, bias=True)\n",
      "  (relu): ReLU()\n",
      "  (dropout): Dropout(p=0.5, inplace=False)\n",
      "  (fc2): Linear(in_features=16, out_features=1, bias=True)\n",
      "  (sigmoid): Sigmoid()\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "# Architecture\n",
    "class Network(nn.Module):\n",
    "    def __init__(self, input_size, hidden_size, dropout_rate):\n",
    "        super(Network, self).__init__()\n",
    "        self.fc1 = nn.Linear(input_size, hidden_size)\n",
    "        self.relu = nn.ReLU()  \n",
    "        self.dropout = nn.Dropout(p=dropout_rate)\n",
    "        self.fc2 = nn.Linear(hidden_size, 1)\n",
    "        self.sigmoid = nn.Sigmoid()\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.fc1(x)\n",
    "        x = self.relu(x)  \n",
    "        x = self.dropout(x)\n",
    "        x = self.fc2(x)\n",
    "        x = self.sigmoid(x)\n",
    "        return x\n",
    "\n",
    "input_size = X_train.shape[1]\n",
    "hidden_size = 16\n",
    "dropout_rate = 0.5\n",
    "\n",
    "model = Network(input_size, hidden_size, dropout_rate).to(device)  \n",
    "print(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "5ff22d08",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:25.086559Z",
     "iopub.status.busy": "2024-09-21T18:36:25.086220Z",
     "iopub.status.idle": "2024-09-21T18:36:26.672747Z",
     "shell.execute_reply": "2024-09-21T18:36:26.671699Z"
    },
    "papermill": {
     "duration": 1.615443,
     "end_time": "2024-09-21T18:36:26.675068",
     "exception": false,
     "start_time": "2024-09-21T18:36:25.059625",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Optimizer and Loss Function\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=0.01)\n",
    "criterion = nn.BCELoss()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "63ae5a16",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:26.726847Z",
     "iopub.status.busy": "2024-09-21T18:36:26.726368Z",
     "iopub.status.idle": "2024-09-21T18:36:26.737024Z",
     "shell.execute_reply": "2024-09-21T18:36:26.736285Z"
    },
    "papermill": {
     "duration": 0.038278,
     "end_time": "2024-09-21T18:36:26.738965",
     "exception": false,
     "start_time": "2024-09-21T18:36:26.700687",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "#Training and Validation Functions\n",
    "def train(model, epoch, log_interval=200):\n",
    "    model.train()\n",
    "    for batch_idx, (data, target) in enumerate(train_loader):\n",
    "        data, target = data.to(device), target.to(device)  # Move data to GPU\n",
    "        optimizer.zero_grad()\n",
    "        output = model(data).squeeze()  \n",
    "        loss = criterion(output, target)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        if batch_idx % log_interval == 0:\n",
    "            print('Train Epoch: {} [{}/{} ({:.0f}%)]\\tLoss: {:.6f}'.format(\n",
    "                epoch, batch_idx * len(data), len(train_loader.dataset),\n",
    "                100. * batch_idx / len(train_loader), loss.item()))\n",
    "\n",
    "@torch.no_grad()\n",
    "def validate(model, loader, loss_vector, accuracy_vector, step=\"Test\"):\n",
    "    model.eval()\n",
    "    val_loss, correct = 0, 0\n",
    "    num_batches = len(loader)\n",
    "    \n",
    "    for data, target in loader:\n",
    "        data, target = data.to(device), target.to(device)  # Move data to GPU\n",
    "        output = model(data).squeeze()\n",
    "        loss = criterion(output, target)\n",
    "        val_loss += loss.item()\n",
    "        pred = (output > 0.5).float()\n",
    "        correct += torch.eq(pred, target).sum().item()\n",
    "    \n",
    "    val_loss /= num_batches\n",
    "    accuracy = 100 * correct / len(loader.dataset)\n",
    "    \n",
    "    loss_vector.append(val_loss)\n",
    "    accuracy_vector.append(accuracy)\n",
    "    \n",
    "    print('{} set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\\n'.format(\n",
    "        step, val_loss, correct, len(loader.dataset), accuracy))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "45eb295e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:26.788728Z",
     "iopub.status.busy": "2024-09-21T18:36:26.788450Z",
     "iopub.status.idle": "2024-09-21T18:36:27.576321Z",
     "shell.execute_reply": "2024-09-21T18:36:27.575328Z"
    },
    "papermill": {
     "duration": 0.815309,
     "end_time": "2024-09-21T18:36:27.578655",
     "exception": false,
     "start_time": "2024-09-21T18:36:26.763346",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch: 1 [0/156 (0%)]\tLoss: 0.807933\n",
      "Training set: Average loss: 0.6469, Accuracy: 95/156 (61%)\n",
      "\n",
      "Test set: Average loss: 0.8343, Accuracy: 8/18 (44%)\n",
      "\n",
      "Train Epoch: 2 [0/156 (0%)]\tLoss: 0.683650\n",
      "Training set: Average loss: 0.5766, Accuracy: 127/156 (81%)\n",
      "\n",
      "Test set: Average loss: 0.7452, Accuracy: 8/18 (44%)\n",
      "\n",
      "Train Epoch: 3 [0/156 (0%)]\tLoss: 0.577537\n",
      "Training set: Average loss: 0.5060, Accuracy: 140/156 (90%)\n",
      "\n",
      "Test set: Average loss: 0.6670, Accuracy: 9/18 (50%)\n",
      "\n",
      "Train Epoch: 4 [0/156 (0%)]\tLoss: 0.509749\n",
      "Training set: Average loss: 0.4432, Accuracy: 144/156 (92%)\n",
      "\n",
      "Test set: Average loss: 0.5993, Accuracy: 13/18 (72%)\n",
      "\n",
      "Train Epoch: 5 [0/156 (0%)]\tLoss: 0.456265\n",
      "Training set: Average loss: 0.3938, Accuracy: 147/156 (94%)\n",
      "\n",
      "Test set: Average loss: 0.5403, Accuracy: 15/18 (83%)\n",
      "\n",
      "Train Epoch: 6 [0/156 (0%)]\tLoss: 0.448371\n",
      "Training set: Average loss: 0.3495, Accuracy: 148/156 (95%)\n",
      "\n",
      "Test set: Average loss: 0.4865, Accuracy: 16/18 (89%)\n",
      "\n",
      "Train Epoch: 7 [0/156 (0%)]\tLoss: 0.384975\n",
      "Training set: Average loss: 0.2786, Accuracy: 151/156 (97%)\n",
      "\n",
      "Test set: Average loss: 0.4418, Accuracy: 15/18 (83%)\n",
      "\n",
      "Train Epoch: 8 [0/156 (0%)]\tLoss: 0.371804\n",
      "Training set: Average loss: 0.2429, Accuracy: 152/156 (97%)\n",
      "\n",
      "Test set: Average loss: 0.4071, Accuracy: 15/18 (83%)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Training loop \n",
    "epochs = 8\n",
    "\n",
    "loss_training, acc_training = [], []\n",
    "loss_validation, acc_validation = [], []\n",
    "\n",
    "for epoch in range(1, epochs + 1):\n",
    "    train(model, epoch)\n",
    "    validate(model, train_loader, loss_training, acc_training, \"Training\")\n",
    "    validate(model, test_loader, loss_validation, acc_validation, \"Test\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "c1954f23",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:27.631541Z",
     "iopub.status.busy": "2024-09-21T18:36:27.631216Z",
     "iopub.status.idle": "2024-09-21T18:36:28.098690Z",
     "shell.execute_reply": "2024-09-21T18:36:28.097784Z"
    },
    "papermill": {
     "duration": 0.496823,
     "end_time": "2024-09-21T18:36:28.101270",
     "exception": false,
     "start_time": "2024-09-21T18:36:27.604447",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epochs_range = np.arange(1, epochs + 1)\n",
    "\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(epochs_range, loss_training, label='Training Loss')\n",
    "plt.plot(epochs_range, loss_validation, label='Test Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Training and Test Loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(epochs_range, acc_training, label='Training Accuracy')\n",
    "plt.plot(epochs_range, acc_validation, label='Test Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accuracy (%)')\n",
    "plt.title('Training and Test Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e3c0bdd",
   "metadata": {
    "papermill": {
     "duration": 0.026563,
     "end_time": "2024-09-21T18:36:28.154586",
     "exception": false,
     "start_time": "2024-09-21T18:36:28.128023",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "8ec87075",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:28.207622Z",
     "iopub.status.busy": "2024-09-21T18:36:28.207286Z",
     "iopub.status.idle": "2024-09-21T18:36:28.216634Z",
     "shell.execute_reply": "2024-09-21T18:36:28.215722Z"
    },
    "papermill": {
     "duration": 0.038249,
     "end_time": "2024-09-21T18:36:28.218624",
     "exception": false,
     "start_time": "2024-09-21T18:36:28.180375",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC of the neural network: 0.8765\n"
     ]
    }
   ],
   "source": [
    "with torch.no_grad(): \n",
    "    y_pred_prob = model(X_test_tensor).cpu().squeeze().numpy()\n",
    "    \n",
    "y_pred_binary = (y_pred_prob > 0.5).astype(int)  # Assuming 0.5 threshold for binary classification\n",
    "\n",
    "#  tensor-to NumPy\n",
    "y_test_np = y_test_tensor.cpu().numpy()\n",
    "\n",
    "# Calculate ROC AUC\n",
    "roc_auc = roc_auc_score(y_test_np, y_pred_prob)\n",
    "print(f'ROC AUC of the neural network: {roc_auc:.4f}')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23d73e2d",
   "metadata": {
    "papermill": {
     "duration": 0.025362,
     "end_time": "2024-09-21T18:36:28.269490",
     "exception": false,
     "start_time": "2024-09-21T18:36:28.244128",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "#### Roc Curves "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "174cfe03",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:28.322268Z",
     "iopub.status.busy": "2024-09-21T18:36:28.321946Z",
     "iopub.status.idle": "2024-09-21T18:36:28.328837Z",
     "shell.execute_reply": "2024-09-21T18:36:28.327905Z"
    },
    "papermill": {
     "duration": 0.035486,
     "end_time": "2024-09-21T18:36:28.330627",
     "exception": false,
     "start_time": "2024-09-21T18:36:28.295141",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_roc_curves(roc_curves, roc_auc_scores, fpr_additional, tpr_additional, auc_additional):\n",
    "    plt.figure(figsize=(12, 8))\n",
    "\n",
    "    # Plot ROC curves for existing models\n",
    "    for model_name, (fpr, tpr) in roc_curves.items():\n",
    "        plt.plot(fpr, tpr, label=f'{model_name} (AUC = {roc_auc_scores[model_name]:.2f})')\n",
    "\n",
    "    # Plot the ROC curve for the additional model\n",
    "    plt.plot(fpr_additional, tpr_additional, color='m', lw=2, label=f'Custom Neural Network (AUC = {auc_additional:.2f})')\n",
    "\n",
    "    # Plot the random classifier line\n",
    "    plt.plot([0, 1], [0, 1], 'k--', label='Random Classifier')\n",
    "\n",
    "    plt.xlabel('False Positive Rate')\n",
    "    plt.ylabel('True Positive Rate')\n",
    "    plt.title('ROC Curves of Different Models')\n",
    "    plt.legend(loc='lower right')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "ad005915",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:28.383886Z",
     "iopub.status.busy": "2024-09-21T18:36:28.383547Z",
     "iopub.status.idle": "2024-09-21T18:36:28.687172Z",
     "shell.execute_reply": "2024-09-21T18:36:28.686221Z"
    },
    "papermill": {
     "duration": 0.333199,
     "end_time": "2024-09-21T18:36:28.689740",
     "exception": false,
     "start_time": "2024-09-21T18:36:28.356541",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr_additional, tpr_additional, _ = roc_curve(y_test_tensor.cpu(), y_pred_prob)\n",
    "auc_additional = auc(fpr_additional, tpr_additional)\n",
    "\n",
    "plot_roc_curves(roc_curves, roc_auc_scores, fpr_additional, tpr_additional, auc_additional)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2bcc198",
   "metadata": {
    "papermill": {
     "duration": 0.027436,
     "end_time": "2024-09-21T18:36:28.745301",
     "exception": false,
     "start_time": "2024-09-21T18:36:28.717865",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "#### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "ac47ab29",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:28.802133Z",
     "iopub.status.busy": "2024-09-21T18:36:28.801772Z",
     "iopub.status.idle": "2024-09-21T18:36:28.807947Z",
     "shell.execute_reply": "2024-09-21T18:36:28.807040Z"
    },
    "papermill": {
     "duration": 0.036905,
     "end_time": "2024-09-21T18:36:28.810039",
     "exception": false,
     "start_time": "2024-09-21T18:36:28.773134",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(y_true, y_pred, model_name):\n",
    "    cm = confusion_matrix(y_true, y_pred)\n",
    "    plt.figure(figsize=(8, 6))\n",
    "    sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', cbar=False,\n",
    "                xticklabels=['Class 0', 'Class 1'], yticklabels=['Class 0', 'Class 1'])\n",
    "    plt.xlabel('Predicted Label')\n",
    "    plt.ylabel('True Label')\n",
    "    plt.title(f'Confusion Matrix for {model_name}')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "d6142cb1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:28.866218Z",
     "iopub.status.busy": "2024-09-21T18:36:28.865934Z",
     "iopub.status.idle": "2024-09-21T18:36:29.161694Z",
     "shell.execute_reply": "2024-09-21T18:36:29.160851Z"
    },
    "papermill": {
     "duration": 0.325624,
     "end_time": "2024-09-21T18:36:29.163710",
     "exception": false,
     "start_time": "2024-09-21T18:36:28.838086",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(y_test_np, y_pred_binary)\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', xticklabels=['Class 0', 'Class 1'], yticklabels=['Class 0', 'Class 1'])\n",
    "plt.xlabel('Predicted Label')\n",
    "plt.ylabel('True Label')\n",
    "plt.title('Confusion Matrix for Custom Neural Network Model')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "475aaedc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:29.220905Z",
     "iopub.status.busy": "2024-09-21T18:36:29.220546Z",
     "iopub.status.idle": "2024-09-21T18:36:29.226367Z",
     "shell.execute_reply": "2024-09-21T18:36:29.225555Z"
    },
    "papermill": {
     "duration": 0.036191,
     "end_time": "2024-09-21T18:36:29.228200",
     "exception": false,
     "start_time": "2024-09-21T18:36:29.192009",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(y_true, y_pred, model_name):\n",
    "    cm = confusion_matrix(y_true, y_pred)\n",
    "    plt.figure(figsize=(8, 6))\n",
    "    sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', xticklabels=['Class 0', 'Class 1'], yticklabels=['Class 0', 'Class 1'])\n",
    "    plt.xlabel('Predicted Label')\n",
    "    plt.ylabel('True Label')\n",
    "    plt.title(f'Confusion Matrix for {model_name}')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "c5649489",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-09-21T18:36:29.285946Z",
     "iopub.status.busy": "2024-09-21T18:36:29.285616Z",
     "iopub.status.idle": "2024-09-21T18:36:30.163405Z",
     "shell.execute_reply": "2024-09-21T18:36:30.162522Z"
    },
    "papermill": {
     "duration": 0.909752,
     "end_time": "2024-09-21T18:36:30.165609",
     "exception": false,
     "start_time": "2024-09-21T18:36:29.255857",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters for CatBoost: {'depth': 6, 'iterations': 200, 'learning_rate': 0.01}\n",
      "Classification Report for CatBoost:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00         9\n",
      "           1       1.00      1.00      1.00         9\n",
      "\n",
      "    accuracy                           1.00        18\n",
      "   macro avg       1.00      1.00      1.00        18\n",
      "weighted avg       1.00      1.00      1.00        18\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters for SVM: {'C': 1, 'kernel': 'rbf'}\n",
      "Classification Report for SVM:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00         9\n",
      "           1       1.00      1.00      1.00         9\n",
      "\n",
      "    accuracy                           1.00        18\n",
      "   macro avg       1.00      1.00      1.00        18\n",
      "weighted avg       1.00      1.00      1.00        18\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters for Logistic Regression: {'C': 0.1, 'penalty': 'l2'}\n",
      "Classification Report for Logistic Regression:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.90      1.00      0.95         9\n",
      "           1       1.00      0.89      0.94         9\n",
      "\n",
      "    accuracy                           0.94        18\n",
      "   macro avg       0.95      0.94      0.94        18\n",
      "weighted avg       0.95      0.94      0.94        18\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred_catboost = grid_search_catboost.predict(X_test)  # Predict class labels\n",
    "print(f\"Best parameters for CatBoost: {grid_search_catboost.best_params_}\")\n",
    "print(f\"Classification Report for CatBoost:\\n{classification_report(y_test, y_pred_catboost)}\")\n",
    "plot_confusion_matrix(y_test, y_pred_catboost, \"CatBoost\")\n",
    "\n",
    "y_pred_svm = grid_search_svm.predict(X_test)  # Predict class labels\n",
    "print(f\"Best parameters for SVM: {grid_search_svm.best_params_}\")\n",
    "print(f\"Classification Report for SVM:\\n{classification_report(y_test, y_pred_svm)}\")\n",
    "plot_confusion_matrix(y_test, y_pred_svm, \"SVM\")\n",
    "\n",
    "y_pred_logreg = grid_search_logreg.predict(X_test)  # Predict class labels\n",
    "print(f\"Best parameters for Logistic Regression: {grid_search_logreg.best_params_}\")\n",
    "print(f\"Classification Report for Logistic Regression:\\n{classification_report(y_test, y_pred_logreg)}\")\n",
    "plot_confusion_matrix(y_test, y_pred_logreg, \"Logistic Regression\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b594f26",
   "metadata": {
    "papermill": {
     "duration": 0.029942,
     "end_time": "2024-09-21T18:36:30.225792",
     "exception": false,
     "start_time": "2024-09-21T18:36:30.195850",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Conclusion\n",
    "\n",
    "In this notebook, we evaluated multiple machine learning models to classify participants as Alzheimer’s patients or healthy individuals based on handwriting features. After performing a grid search to optimize hyperparameters, the **Support Vector Machine (SVM)** model showed the best performance with an accuracy of **0.98**.\n",
    "\n",
    "### Why SVM Performed Better\n",
    "\n",
    "The SVM model outperformed other models like CatBoost and Logistic Regression for several reasons:\n",
    "\n",
    "1. **Ability to Handle High-Dimensional Data**: The DARWIN dataset contains 450 handwriting features, making it highly multi-dimensional. SVM is well-suited for datasets with many features, especially when the number of samples is relatively small.\n",
    "   \n",
    "2. **Effective in Non-Linear Classification**: SVM with a properly selected kernel (like the RBF kernel) can handle non-linear relationships between the features and the target variable. This likely contributed to its high performance.\n",
    "\n",
    "3. **Robustness Against Overfitting**: SVM models are effective in preventing overfitting"
   ]
  }
 ],
 "metadata": {
  "kaggle": {
   "accelerator": "nvidiaTeslaT4",
   "dataSources": [
    {
     "datasetId": 3609813,
     "sourceId": 6278618,
     "sourceType": "datasetVersion"
    }
   ],
   "dockerImageVersionId": 30762,
   "isGpuEnabled": true,
   "isInternetEnabled": true,
   "language": "python",
   "sourceType": "notebook"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.14"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 654.525485,
   "end_time": "2024-09-21T18:36:31.978154",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2024-09-21T18:25:37.452669",
   "version": "2.6.0"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "01390b7b874246158d7ac3ff9143ee61": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "35f85764500240d8b39afa4ce27c77c2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "43d8f2b3d75f4f7f85079dbd55bbebdf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "VBoxModel",
      "state": {
       "_dom_classes": [
        "widget-interact"
       ],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "VBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "VBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_edbcbe2018b548158fa10b6b67361e70",
        "IPY_MODEL_f509710592774f95a152bd35bb12dcec",
        "IPY_MODEL_7a57baca481a4208a05eb3c774a907dd"
       ],
       "layout": "IPY_MODEL_cac5ae50ac8b4591a3f2cc256465a73e"
      }
     },
     "6d71ec7cdf9e4d8b8647aaec591650ed": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "764305a1ce004e06ab311ef24f8c8797": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "7a57baca481a4208a05eb3c774a907dd": {
      "model_module": "@jupyter-widgets/output",
      "model_module_version": "1.0.0",
      "model_name": "OutputModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/output",
       "_model_module_version": "1.0.0",
       "_model_name": "OutputModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/output",
       "_view_module_version": "1.0.0",
       "_view_name": "OutputView",
       "layout": "IPY_MODEL_9c3706e46fb541fba05080923f3c7b09",
       "msg_id": "",
       "outputs": [
        {
         "data": {
          "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0MAAAIjCAYAAADBQ8ABAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB0oklEQVR4nO3de5xN9f7H8ffac7+Yi8vMGIaRyiVESENSp8mQdJQKOZGkG4XpKhqkUjoVlTh0pFOU9Ct1IiWXFJM7kUsXimjGfYbB3Pb398c0+9jNuMzYM3tv+/U8j/U4M2t993d91l4z2Z/5fL/fZRljjAAAAADAx9jcHQAAAAAAuAPJEAAAAACfRDIEAAAAwCeRDAEAAADwSSRDAAAAAHwSyRAAAAAAn0QyBAAAAMAnkQwBAAAA8EkkQwAAAAB8EskQAJ/266+/yrIsTZ8+3d2hOJk/f76aN2+u4OBgWZalw4cPn1N/S5YskWVZWrJkiUviO53ExETdeeedFX4eX3PnnXcqMTGxws9TmT8rAOBuJEPAeWrjxo265ZZbVLduXQUHB6tWrVq67rrr9Nprr1XYOWfOnKnx48eX2L9nzx6NGjVK69evr7Bz/1XxB7riLSAgQBdccIH69Omj7du3u+Qcy5cv16hRo845UfmrAwcO6LbbblNISIgmTpyod955R2FhYS49x7mqqGs/VytXrtQDDzygli1bKiAgQJZluTskj/XGG2943B8BJGnSpEm69dZbVadOHVmWRWINoEL5uzsAAK63fPlyXXPNNapTp44GDBiguLg47dq1S999950mTJigBx98sELOO3PmTG3atElDhgxx2r9nzx6NHj1aiYmJat68eYWc+1QeeughtW7dWvn5+Vq7dq2mTJmiuXPnauPGjYqPjz+nvpcvX67Ro0frzjvvVFRUlGsClrRq1SodOXJEY8aMUXJyskv6vOqqq3T8+HEFBga6pL/TXfu2bdtks7nnb23z5s3Tm2++qWbNmumCCy7Qjz/+6JY4KsLUqVNlt9td1t8bb7yh6tWrl0g2XP2zUlYvvPCCjhw5ossvv1x//PGHW2IA4DtIhoDz0LPPPqvIyEitWrWqxAfVvXv3uieoCpCTk3PGikn79u11yy23SJL69euniy++WA899JDefvttDRs2rDLCLLPie+TKBMtmsyk4OPiM7Y4dO6bQ0NBzOldQUNA5vf5c3H///Xr88ccVEhKiQYMGnVfJUEBAwBnbFBQUyG63n1Mic7Y/KxXl66+/dlSFwsPD3RYHAN/AMDngPPTLL7/okksuKfXDdExMTIl97777ri6//HKFhoYqOjpaV111lb788kvH8U8++URdunRRfHy8goKCVL9+fY0ZM0aFhYWONldffbXmzp2r3377zTE0LTExUUuWLFHr1q0lFSUjxcdOHp6zYsUKderUSZGRkQoNDVWHDh20bNkypxhHjRoly7K0efNm3X777YqOjtaVV15Z5vfmb3/7myRpx44dp223aNEitW/fXmFhYYqKitLf//53bdmyxSmeRx99VJJUr149x3X9+uuvp+139uzZatmypUJCQlS9enX94x//0O7dux3Hr776avXt21eS1Lp16zMOE/rtt9/0wAMPqEGDBgoJCVG1atV06623loijtHkgV199tZo0aaI1a9boqquuUmhoqJ588snTxn821/7XOUPTp0+XZVn69ttv9dBDD6lGjRqKiorSvffeq7y8PB0+fFh9+vRRdHS0oqOj9dhjj8kY43ROu92u8ePH65JLLlFwcLBiY2N177336tChQ07tYmNjFRIScsZrKE2TJk10zTXXlNhvt9tVq1YtR1ItSe+//75atmypKlWqKCIiQk2bNtWECRPKdd6z+f2SSs4ZKp7v9s9//lPjx49X/fr1FRQUpM2bN5/xnImJifrhhx/09ddfO+7f1VdfLen0Pyvff/+9OnTooNDQUF144YX68MMPJRUlMG3atFFISIgaNGigr776qsQ5d+/erbvuukuxsbEKCgrSJZdcomnTppVoV7duXYY3Aqg0VIaA81DdunWVnp6uTZs2qUmTJqdtO3r0aI0aNUpt27bV008/rcDAQK1YsUKLFi1Sx44dJRV9mA0PD1dqaqrCw8O1aNEipaWlKTs7Wy+++KIkafjw4crKytLvv/+uV155RZIUHh6uRo0a6emnn1ZaWpruuecetW/fXpLUtm1bSUVJR+fOndWyZUuNHDlSNptNb731lv72t7/pm2++0eWXX+4U76233qqLLrpIzz33XIkPzGfjl19+kSRVq1btlG2++uorde7cWRdccIFGjRql48eP67XXXlO7du20du1aJSYm6uabb9aPP/6o9957T6+88oqqV68uSapRo8Yp+50+fbr69eun1q1ba+zYscrMzNSECRO0bNkyrVu3TlFRURo+fLgaNGigKVOm6Omnn1a9evVUv379U/a5atUqLV++XD179lTt2rX166+/atKkSbr66qu1efPmM1Z5Dhw4oM6dO6tnz576xz/+odjY2NO2l1Sua5ekBx98UHFxcRo9erS+++47TZkyRVFRUVq+fLnq1Kmj5557TvPmzdOLL76oJk2aqE+fPo7X3nvvvY7376GHHtKOHTv0+uuva926dVq2bNlZVU3OpEePHho1apQyMjIUFxfn2P/tt99qz5496tmzpyRpwYIF6tWrl6699lq98MILkqQtW7Zo2bJlGjx4cJnPeza/X6fz1ltv6cSJE7rnnnsUFBSkqlWrnvE148eP14MPPqjw8HANHz5cks547w8dOqQbbrhBPXv21K233qpJkyapZ8+emjFjhoYMGaL77rtPt99+u1588UXdcsst2rVrl6pUqSJJyszM1BVXXCHLsjRo0CDVqFFDn3/+ufr376/s7OwSQ2sBoNIYAOedL7/80vj5+Rk/Pz+TlJRkHnvsMfPFF1+YvLw8p3Y//fSTsdls5qabbjKFhYVOx+x2u+PrY8eOlTjHvffea0JDQ82JEycc+7p06WLq1q1bou2qVauMJPPWW2+VOMdFF11kUlJSSpyvXr165rrrrnPsGzlypJFkevXqdVbvweLFi40kM23aNLNv3z6zZ88eM3fuXJOYmGgsyzKrVq0yxhizY8eOErE1b97cxMTEmAMHDjj2bdiwwdhsNtOnTx/HvhdffNFIMjt27DhjPHl5eSYmJsY0adLEHD9+3LH/s88+M5JMWlqaY99bb71lJDliPJ3S7k16erqRZP7zn/+UeD8WL17s2NehQwcjyUyePPmM5/mr01173bp1Td++fR3fF1/PX+9zUlKSsSzL3HfffY59BQUFpnbt2qZDhw6Ofd98842RZGbMmOF0nvnz55e6v9jAgQNNWf6Z27Ztm5FkXnvtNaf9DzzwgAkPD3e814MHDzYRERGmoKDgrPs+nbP9/erbt6/T71fxz25ERITZu3dvmc97ySWXOL3PxU73szJz5kzHvq1btxpJxmazme+++86x/4svvijxO9W/f39Ts2ZNs3//fqdz9ezZ00RGRpb6HhhjTFhYmNPPEgC4GsPkgPPQddddp/T0dN14443asGGDxo0bp5SUFNWqVUuffvqpo92cOXNkt9uVlpZWYsL7ycNUTh52dOTIEe3fv1/t27fXsWPHtHXr1nLHuX79ev3000+6/fbbdeDAAe3fv1/79+9XTk6Orr32Wi1durTEhPH77ruvTOe46667VKNGDcXHx6tLly7KycnR22+/rVatWpXa/o8//tD69et15513Ov2FvVmzZrruuus0b968sl+opNWrV2vv3r164IEHnOZjdOnSRQ0bNtTcuXPL1e/J9yY/P18HDhzQhRdeqKioKK1du/aMrw8KClK/fv3Kde6y6t+/v9PPVZs2bWSMUf/+/R37/Pz81KpVK6cV/2bPnq3IyEhdd911jp+R/fv3q2XLlgoPD9fixYtdEt/FF1+s5s2ba9asWY59hYWF+vDDD9W1a1fHex0VFaWcnBwtWLDAJec919+v7t27n7Eq5wrh4eGO6pgkNWjQQFFRUWrUqJHatGnj2F/8dfE9NMbo//7v/9S1a1cZY5zuYUpKirKyss7qZxUAKgLD5IDzVOvWrfXRRx8pLy9PGzZs0Mcff6xXXnlFt9xyi9avX6/GjRvrl19+kc1mU+PGjU/b1w8//KARI0Zo0aJFys7OdjqWlZVV7hh/+uknSXLMkSlNVlaWoqOjHd/Xq1evTOdIS0tT+/bt5efnp+rVq6tRo0by9z/1f/p+++03SUUf9P6qUaNG+uKLL85q4Yay9NuwYUN9++23Zeqv2PHjxzV27Fi99dZb2r17t9PQwbO5N7Vq1aq0VcPq1Knj9H1kZKQkKSEhocT+k+cC/fTTT8rKyip1vpvk2kVBevTooSeffFK7d+9WrVq1tGTJEu3du1c9evRwtHnggQf0wQcfqHPnzqpVq5Y6duyo2267TZ06dSrXOc/196usvxPlVbt27RJzeSIjI0u9f5Ic93Dfvn06fPiwpkyZoilTppTa9/m0sAsA70IyBJznAgMD1bp1a7Vu3VoXX3yx+vXrp9mzZ2vkyJFn9frDhw+rQ4cOioiI0NNPP6369esrODhYa9eu1eOPP35OS/0Wv/bFF1885ZLbf11NqqyT45s2beqy5ak90YMPPqi33npLQ4YMUVJSkiIjI2VZlnr27HlW96a8iw2Uh5+f31nvPzmps9vtiomJ0YwZM0p9vSurIj169NCwYcM0e/ZsDRkyRB988IEiIyOdEp2YmBitX79eX3zxhT7//HN9/vnneuutt9SnTx+9/fbbZTqfK36/KuseluX+Sf+7h8XX8I9//OOUf/ho1qyZCyIEgLIjGQJ8SPHQsOJnd9SvX192u12bN28+ZTKyZMkSHThwQB999JGuuuoqx/7SVmM71QpQp9pfvDBARESExyQsdevWlVT0rJy/2rp1q6pXr+6oCpVlxauT+y1e0a7Ytm3bHMfL6sMPP1Tfvn310ksvOfadOHGiwh+GWpmrfdWvX19fffWV2rVrV+Ef/OvVq6fLL79cs2bN0qBBg/TRRx+pW7duJZYLDwwMVNeuXdW1a1fZ7XY98MAD+te//qWnnnpKF1544Vmfryy/X65WWfewRo0aqlKligoLCz3m9xwAijFnCDgPLV68uNSV1ornuxQP1erWrZtsNpuefvrpEn+BLn598V99T+4vLy9Pb7zxRon+w8LCSh3WU5w8/PUDesuWLVW/fn3985//1NGjR0u8bt++fae8xopSs2ZNNW/eXG+//bZTvJs2bdKXX36p66+/3rHvVNdVmlatWikmJkaTJ09Wbm6uY//nn3+uLVu2qEuXLuWK18/Pr8S9fu2110osy+xqZbn2c3XbbbepsLBQY8aMKXGsoKDA5TH06NFD3333naZNm6b9+/c7DZGTilbgO5nNZnNUNorvbX5+vrZu3XrGh4aW5ffL1cLCwirl/vn5+al79+76v//7P23atKnEcXf8ngNAMSpDwHnowQcf1LFjx3TTTTepYcOGysvL0/LlyzVr1iwlJiY6JsxfeOGFGj58uMaMGaP27dvr5ptvVlBQkFatWqX4+HiNHTtWbdu2VXR0tPr27auHHnpIlmXpnXfeKTXZatmypWbNmqXU1FS1bt1a4eHh6tq1q+rXr6+oqChNnjxZVapUUVhYmNq0aaN69erpzTffVOfOnXXJJZeoX79+qlWrlnbv3q3FixcrIiJC//3vfyv77dOLL76ozp07KykpSf3793csrR0ZGalRo0Y5Xa9UtKx4z549FRAQoK5du5Y6nyggIEAvvPCC+vXrpw4dOqhXr16OpbUTExM1dOjQcsV6ww036J133lFkZKQaN26s9PR0ffXVV6ddOtwVynLt56pDhw669957NXbsWK1fv14dO3ZUQECAfvrpJ82ePVsTJkxwPAPot99+0zvvvCOpaNEKSXrmmWckFVXn7rjjjjOe77bbbtMjjzyiRx55RFWrVi1Rzbj77rt18OBB/e1vf1Pt2rX122+/6bXXXlPz5s3VqFEjSUXP1GnUqJH69u3r9EytvyrL75ertWzZUpMmTdIzzzyjCy+8UDExMSWqlq7y/PPPa/HixWrTpo0GDBigxo0b6+DBg1q7dq2++uorHTx40NH2v//9rzZs2CCpKKn8/vvvHffwxhtvZEgdANdyyxp2ACrU559/bu666y7TsGFDEx4ebgIDA82FF15oHnzwQZOZmVmi/bRp00yLFi1MUFCQiY6ONh06dDALFixwHF+2bJm54oorTEhIiImPj3cs1a2/LL979OhRc/vtt5uoqCgjyWkZ4E8++cQ0btzY+Pv7l1h2d926debmm2821apVM0FBQaZu3brmtttuMwsXLnS0KV5ae9++fWf1HhQvDzx79uzTtittaW1jjPnqq69Mu3btTEhIiImIiDBdu3Y1mzdvLvH6MWPGmFq1ahmbzXZWy2zPmjXL8V5XrVrV9O7d2/z+++9ObcqytPahQ4dMv379TPXq1U14eLhJSUkxW7duLbG89amWS77kkkvOeI5TOdW1n2pp7b9ez6nuad++fU1YWFiJ802ZMsW0bNnShISEmCpVqpimTZuaxx57zOzZs6fEdZa2lbaM9Km0a9fOSDJ33313iWMffvih6dixo4mJiTGBgYGmTp065t577zV//PGHo03xz9XZLAt9tr9fp1pa+8UXXzzr6zpZRkaG6dKli6lSpYrT+1OWn5W6deuaLl26lNgvyQwcONBpX2Zmphk4cKBJSEgwAQEBJi4uzlx77bVmypQpTu369u17ynv4199TADhXljGV8OcnAAAAAPAwzBkCAAAA4JOYMwQAcDh69Gipi1mcrEaNGqdcThnut2/fvtMuoBEYGOj0QGEA8GUMkwMAOIwaNUqjR48+bZsdO3YoMTGxcgJCmSUmJjoe8luaDh06aMmSJZUXEAB4MJIhAIDD9u3btX379tO2ufLKKxUcHFxJEaGsli1bpuPHj5/yeHR0tGM1QAAor6VLl+rFF1/UmjVr9Mcff+jjjz9Wt27dTvuaJUuWKDU1VT/88IMSEhI0YsQI3XnnnZUS76kwTA4A4HDBBRfoggsucHcYOAft2rVzdwgAfEBOTo4uvfRS3XXXXbr55pvP2H7Hjh3q0qWL7rvvPs2YMUMLFy7U3XffrZo1ayolJaUSIi4dlSEAAAAA5WZZ1hkrQ48//rjmzp3r9PDlnj176vDhw5o/f34lRFk6KkOVyG63a8+ePapSpYosy3J3OAAAAPgLY4yOHDmi+Ph42Wyet/DyiRMnlJeXVyF9G2NKfEYNCgpSUFDQOfednp5e4iHWKSkpGjJkyDn3fS5IhirRnj17lJCQ4O4wAAAAcAa7du1S7dq13R2GkxMnTqheYlVlZJ56XuC5CA8PL7Gi6MiRIzVq1Khz7jsjI0OxsbFO+2JjY5Wdna3jx48rJCTknM9RHiRDlahKlSqSin65IiIi3BwNAAAA/io7O1sJCQmOz22eJC8vTxmZx/Xrxl6KqBLg0r6zj+Qrsel7JT6nuqIq5MlIhipRcdkxIiKCZAgAAMCDefKUhogqAYqICKyYvivoc2pcXJwyMzOd9mVmZioiIsJtVSGJZAgAAADwKnZjZHfxGmiu7u+vkpKSNG/ePKd9CxYsUFJSUoWe90w8b1YYAAAAAI929OhRrV+/XuvXr5dUtHT2+vXrtXPnTknSsGHD1KdPH0f7++67T9u3b9djjz2mrVu36o033tAHH3ygoUOHuiN8BypDAAAAgBexy8guF1eGytjf6tWrdc011zi+T01NlST17dtX06dP1x9//OFIjCSpXr16mjt3roYOHaoJEyaodu3aevPNN936jCGJZAgAAABAGV199dU63eNKp0+fXupr1q1bV4FRlR3JEAAAAOBFzJ//c3Wfvog5QwAAAAB8EpUhAAAAwIvYVQGryfloZYhkCAAAAPAi9j83V/fpixgmBwAAAMAnURkCAAAAvIgnLK19vqAyBAAAAMAnURkCAAAAvAhLa7sOyRAAAAA8krEfkQp3S5a/5FdHlhXo7pBwniEZAgAAgEcx9qMyx96XchdLJluSTfKrKQV3lYKvl2VZ7g7RrZgz5DokQwAAAPAYxn5M5shYKW+VZIVLtmqS7FLhbpmjb0j2A7LC+rg7TJwnSIYAAADgOXKXSHlriipBVvD/9vuFSPYD0vFPZIKukuWf6K4I3c7I9c8F8s26EKvJAQAAwIOY3K8k+TknQsWsqpI5KuUtq/S4PIm9gjZfRDIEAAAAz1GYKVkhpR/7c66QKTxQiQHhfMYwOQAAAHgOW6RUuKv0Y8ZIMrJsVSo1JE9jN0Wbq/v0RVSGAAAA4DGsoGskkyeZgpIHzdGiqlFgm8oPDOclkiEAAAB4juBrJf8Li6pD9iNF1SBTKBUeKFpAIegqyb+Ru6N0K1NBmy8iGQIAAIDHsGxVZUWMkAKTJHNcKvxNKvxdsvykkO6ywgf5/HOG4DrMGQIAAIBHsfxqShGjpMIdUsFvRYmQ/yWy/Kq5OzSPUBGrv/nqanIkQwAAAPA4lmVJ/hcUbUAFIRkCAAAAvAirybkOyRAAAADgReyyZJdr5025uj9vwQIKAAAAAHwSlSEAAADAi1AZch0qQwAAAAB8EpUhAAAAwIsYY8kY11ZyXN2ft6AyBAAAAMAnURkCAAAAvEjhn5ur+/RFVIYAAAAA+CQqQwAAAIAXMbLJ7uKahvHRGgnJEAAAAOBFzJ+bq/v0Rb6ZAgIAAADweVSGAAAAAC/CQ1ddh8oQAAAAAJ9EZQgAAADwInZjye7ih6S6uj9vQWUIAAAAgE+iMgQAAAB4EXsFLK3t6v68hW9eNQAAAACfR2UIAAAA8CKsJuc6VIYAAAAA+CQqQwAAAIAXYTU51yEZAgAAALyIkSXj4mFtru7PWzBMDgAAAIBPojIEAAAAeBEWUHAdKkMAAAAAfBKVIQAAAMCLUBlyHSpDAAAAAHwSlSEAAADAixjZZHdxTcP4aI3ErVe9dOlSde3aVfHx8bIsS3PmzHE6boxRWlqaatasqZCQECUnJ+unn35yanPw4EH17t1bERERioqKUv/+/XX06FGnNt9//73at2+v4OBgJSQkaNy4cSVimT17tho2bKjg4GA1bdpU8+bNK3MsAAAAALyHW5OhnJwcXXrppZo4cWKpx8eNG6dXX31VkydP1ooVKxQWFqaUlBSdOHHC0aZ379764YcftGDBAn322WdaunSp7rnnHsfx7OxsdezYUXXr1tWaNWv04osvatSoUZoyZYqjzfLly9WrVy/1799f69atU7du3dStWzdt2rSpTLEAAAAAFc1u/vfgVddt7r4q97CMMR5x6ZZl6eOPP1a3bt0kFVVi4uPj9fDDD+uRRx6RJGVlZSk2NlbTp09Xz549tWXLFjVu3FirVq1Sq1atJEnz58/X9ddfr99//13x8fGaNGmShg8froyMDAUGBkqSnnjiCc2ZM0dbt26VJPXo0UM5OTn67LPPHPFcccUVat68uSZPnnxWsZyN7OxsRUZGKisrSxERES553wAAAOA6nvx5rTi2+dvuU1iVIJf2nXMkV50aTPbI665IHjs4cMeOHcrIyFBycrJjX2RkpNq0aaP09HRJUnp6uqKiohyJkCQlJyfLZrNpxYoVjjZXXXWVIxGSpJSUFG3btk2HDh1ytDn5PMVtis9zNrGUJjc3V9nZ2U4bAAAAAM/gsclQRkaGJCk2NtZpf2xsrONYRkaGYmJinI77+/uratWqTm1K6+Pkc5yqzcnHzxRLacaOHavIyEjHlpCQcIarBgAAAE7PGKtCNl/kscnQ+WDYsGHKyspybLt27XJ3SAAAAAD+5LHJUFxcnCQpMzPTaX9mZqbjWFxcnPbu3et0vKCgQAcPHnRqU1ofJ5/jVG1OPn6mWEoTFBSkiIgIpw0AAAA4F4WyVcjmizz2quvVq6e4uDgtXLjQsS87O1srVqxQUlKSJCkpKUmHDx/WmjVrHG0WLVoku92uNm3aONosXbpU+fn5jjYLFixQgwYNFB0d7Whz8nmK2xSf52xiAQAAAOBd3JoMHT16VOvXr9f69eslFS1UsH79eu3cuVOWZWnIkCF65pln9Omnn2rjxo3q06eP4uPjHSvONWrUSJ06ddKAAQO0cuVKLVu2TIMGDVLPnj0VHx8vSbr99tsVGBio/v3764cfftCsWbM0YcIEpaamOuIYPHiw5s+fr5deeklbt27VqFGjtHr1ag0aNEiSzioWAAAAoDLYK2jzRf7uPPnq1at1zTXXOL4vTlD69u2r6dOn67HHHlNOTo7uueceHT58WFdeeaXmz5+v4OBgx2tmzJihQYMG6dprr5XNZlP37t316quvOo5HRkbqyy+/1MCBA9WyZUtVr15daWlpTs8iatu2rWbOnKkRI0boySef1EUXXaQ5c+aoSZMmjjZnEwsAAAAA7+ExzxnyBZ68bj0AAAA8+/NacWyfbH2wQp4z9PeGr3nkdVckt1aGAAAAUKQgv0Cb039U5m/7FBgcqEZXXKSYhOruDgseyG4s2V28FLar+/MWJEMAAABu9uOaXzTjmf/Tnp8zZC+0yxijsMhQXdG1pW59+EYFBgeeuRMAZUYyBAAA4Ea//7hH/3rkPzqUeVg1alVTYEig7HajIwePaOG736ggv1B9R/Vwd5jwIHZZssvFlSEX9+ctPHZpbQAAAF+w+P1lOrDnkOIviFNgSFEFyGazFFk9QlWqVdHKuWv1+09/uDlK4PxEMgQAAOAmeSfytPar7xUeFSrLVvIv81Wiw3Ts6Alt+maLG6KDpzJ/zhly5WZ8dM4QyRAAAICb5J3IV0F+ofwDA0o9blmWbDZLJ47lVnJkgG8gGQIAAHCTkCrBiqxeRcePHi/1eGFBoYyRqtWMruTI4MnsslXI5ot886oBAAA8gJ+fn9p1u1z5J/KVdzzP6ZgxRvt3H1TVuEi1uLapmyIEzm+sJgcAAOBGHW5rq03Ltmrz8h8VHBaksIgQFRQUKnv/EYVUCVH3oV0VHhXm7jDhQYwp2lzdpy8iGQIAAHCj0CohemB8P331zlIt/2SljhzMkc3PpmYdGiv5H1epyZWN3B0iPExFDGvz1WFyJEMAAABuFhYRqr8P7KTO/f+mw3uzFBAUoKiYSFmWb67wBVQWkiEAAAAPERgcqJg6NdwdBjwcD111Hd+shwEAAADweVSGAAAAAC9CZch1qAwBAAAA8ElUhgAAAAAvYowlY1xbyXF1f96CyhAAAAAAn0RlCAAAAPAizBlyHZIhAAAAwIvYjWR38bA2u3Fpd16DYXIAAAAAfBKVIQAAAMCL2GWT3cU1DVf35y1886oBAAAA+DySIQAAAMCLGFkVspXVxIkTlZiYqODgYLVp00YrV648bfvx48erQYMGCgkJUUJCgoYOHaoTJ06U921wCZIhAAAAAGUya9YspaamauTIkVq7dq0uvfRSpaSkaO/evaW2nzlzpp544gmNHDlSW7Zs0b///W/NmjVLTz75ZCVH7oxkCAAAAPAixliyu3gr60NXX375ZQ0YMED9+vVT48aNNXnyZIWGhmratGmltl++fLnatWun22+/XYmJierYsaN69ep1xmpSRSMZAgAAACBJys7Odtpyc3NLtMnLy9OaNWuUnJzs2Gez2ZScnKz09PRS+23btq3WrFnjSH62b9+uefPm6frrr6+YCzlLrCYHAAAAeJHiao6r+5SkhIQEp/0jR47UqFGjnPbt379fhYWFio2NddofGxurrVu3ltr/7bffrv379+vKK6+UMUYFBQW677773D5MjmQIAAAA8CJ2WbKXY8GDM/UpSbt27VJERIRjf1BQkEv6X7JkiZ577jm98cYbatOmjX7++WcNHjxYY8aM0VNPPeWSc5QHyRAAAAAASVJERIRTMlSa6tWry8/PT5mZmU77MzMzFRcXV+prnnrqKd1xxx26++67JUlNmzZVTk6O7rnnHg0fPlw2m3tm7zBnCAAAAPAi7l5aOzAwUC1bttTChQsd++x2uxYuXKikpKRSX3Ps2LESCY+fn1/R9RhTjnfBNagMAQAAACiT1NRU9e3bV61atdLll1+u8ePHKycnR/369ZMk9enTR7Vq1dLYsWMlSV27dtXLL7+sFi1aOIbJPfXUU+ratasjKXIHkiEAAADAi1TknKGz1aNHD+3bt09paWnKyMhQ8+bNNX/+fMeiCjt37nSqBI0YMUKWZWnEiBHavXu3atSooa5du+rZZ5916XWUlWXcWZfyMdnZ2YqMjFRWVtYZx2ICAACg8nny57Xi2F7bMEohVYJd2vfxIyf04KWjPPK6KxKVIQAAAMCLVOTS2r6GBRQAAAAA+CQqQwAAAIAXMcaScXElx9X9eQuSIQAAAMCLkAy5DsPkAAAAAPgkKkMAAACAFymUpUIXL63t6v68BZUhAAAAAD6JyhAAAADgRcyfm6v79EVUhgAAAAD4JCpDAAAAgBcxsskY19Y0jI/WSHzzqgEAAAD4PCpDAAAAgBcxkuwV0KcvIhkCAAAAvIjdWLK7+CGpru7PWzBMDgAAAIBPojIEAAAAeBEjm8sXPGABBQAAAADwIVSGAAAAAC9iN0Wbq/v0RVSGAAAAAPgkKkMAAACAFzGSjFy7+puPFoaoDAEAAADwTVSGAAAAAC9ijCXj4ucCubo/b0EyBAAAAHgRuyzZXTxMztX9eQuGyQEAAADwSVSGAAAAAC/CMDnXoTIEAAAAwCdRGQIAAAC8iP3PzdV9+iIqQwAAAAB8EpUhAAAAwIswZ8h1qAwBAAAA8ElUhgAAAAAvQmXIdagMAQAAAPBJVIYAAAAAL2KXJbtcW8lxdX/ewqMrQ4WFhXrqqadUr149hYSEqH79+hozZoyMMY42xhilpaWpZs2aCgkJUXJysn766Senfg4ePKjevXsrIiJCUVFR6t+/v44ePerU5vvvv1f79u0VHByshIQEjRs3rkQ8s2fPVsOGDRUcHKymTZtq3rx5FXPhAAAAwCkYUzGbL/LoZOiFF17QpEmT9Prrr2vLli164YUXNG7cOL322muONuPGjdOrr76qyZMna8WKFQoLC1NKSopOnDjhaNO7d2/98MMPWrBggT777DMtXbpU99xzj+N4dna2OnbsqLp162rNmjV68cUXNWrUKE2ZMsXRZvny5erVq5f69++vdevWqVu3burWrZs2bdpUOW8GAAAAAJeyjPHcPPCGG25QbGys/v3vfzv2de/eXSEhIXr33XdljFF8fLwefvhhPfLII5KkrKwsxcbGavr06erZs6e2bNmixo0ba9WqVWrVqpUkaf78+br++uv1+++/Kz4+XpMmTdLw4cOVkZGhwMBASdITTzyhOXPmaOvWrZKkHj16KCcnR5999pkjliuuuELNmzfX5MmTz+p6srOzFRkZqaysLEVERLjkPQIAAIDrePLnteLYHk8fr6DwEJf2nXv0uF5IGuKR112RPLoy1LZtWy1cuFA//vijJGnDhg369ttv1blzZ0nSjh07lJGRoeTkZMdrIiMj1aZNG6Wnp0uS0tPTFRUV5UiEJCk5OVk2m00rVqxwtLnqqqsciZAkpaSkaNu2bTp06JCjzcnnKW5TfJ7S5ObmKjs722kDAAAA4Bk8egGFJ554QtnZ2WrYsKH8/PxUWFioZ599Vr1795YkZWRkSJJiY2OdXhcbG+s4lpGRoZiYGKfj/v7+qlq1qlObevXqleij+Fh0dLQyMjJOe57SjB07VqNHjy7rZQMAAACnxNLaruPRlaEPPvhAM2bM0MyZM7V27Vq9/fbb+uc//6m3337b3aGdlWHDhikrK8ux7dq1y90hAQAAAPiTR1eGHn30UT3xxBPq2bOnJKlp06b67bffNHbsWPXt21dxcXGSpMzMTNWsWdPxuszMTDVv3lySFBcXp7179zr1W1BQoIMHDzpeHxcXp8zMTKc2xd+fqU3x8dIEBQUpKCiorJcNAAAAnJL5c3N1n77IoytDx44dk83mHKKfn5/sdrskqV69eoqLi9PChQsdx7Ozs7VixQolJSVJkpKSknT48GGtWbPG0WbRokWy2+1q06aNo83SpUuVn5/vaLNgwQI1aNBA0dHRjjYnn6e4TfF5AAAAAHgXj06GunbtqmeffVZz587Vr7/+qo8//lgvv/yybrrpJkmSZVkaMmSInnnmGX366afauHGj+vTpo/j4eHXr1k2S1KhRI3Xq1EkDBgzQypUrtWzZMg0aNEg9e/ZUfHy8JOn2229XYGCg+vfvrx9++EGzZs3ShAkTlJqa6ohl8ODBmj9/vl566SVt3bpVo0aN0urVqzVo0KBKf18AAADgu+zGqpDNF3n0MLnXXntNTz31lB544AHt3btX8fHxuvfee5WWluZo89hjjyknJ0f33HOPDh8+rCuvvFLz589XcHCwo82MGTM0aNAgXXvttbLZbOrevbteffVVx/HIyEh9+eWXGjhwoFq2bKnq1asrLS3N6VlEbdu21cyZMzVixAg9+eSTuuiiizRnzhw1adKkct4MAAAAQCyg4Eoe/Zyh840nr1sPAAAAz/68Vhxb6revVchzhl6+8kGPvO6K5NGVIQAAAADOWEDBdTx6zhAAAAAAVBQqQwAAAIAXMbJk5OI5Qy7uz1tQGQIAAADgk6gMAQAAAF6kIpbC9tWltakMAQAAAPBJVIYAAAAAb2Ksos3VffogkiEAAADAi9hN0ebqPn0Rw+QAAAAA+CQqQwAAAIAXYWlt16EyBAAAAMAnURkCAAAAvIgxloyLFzxwdX/egsoQAAAAAJ9EZQgAAADwIsYUba7u0xdRGQIAAADgk6gMAQAAAF7E/Lm5uk9fRDIEAAAAeBGW1nYdhskBAAAA8ElUhgAAAABvUgFLa4ultQEAAADAd1AZAgAAALwIS2u7DpUhAAAAAD6JyhAAAADgRVha23WoDAEAAADwSVSGAAAAAG/jo6u/uRrJEAAAAOBF7MaS3cXJkKv78xYMkwMAAADgk6gMAQAAAN6EFRRchsoQAAAAAJ9EZQgAAADwIkaWjFw7x8fV/XkLKkMAAAAAfBKVIQAAAMCLGFO0ubpPX0RlCAAAAIBPIhkCAAAA4JMYJgcAAAB4EWMsGRc/JNXV/XkLKkMAAAAAfBLJEAAAAACfRDIEAAAAwCcxZwgAAADwIswZch0qQwAAAADKbOLEiUpMTFRwcLDatGmjlStXnrb94cOHNXDgQNWsWVNBQUG6+OKLNW/evEqKtnRUhgAAAAAv4gkPXZ01a5ZSU1M1efJktWnTRuPHj1dKSoq2bdummJiYEu3z8vJ03XXXKSYmRh9++KFq1aql3377TVFRUa65gHIiGQIAAAAgScrOznb6PigoSEFBQSXavfzyyxowYID69esnSZo8ebLmzp2radOm6YknnijRftq0aTp48KCWL1+ugIAASVJiYqLrL6CMGCYHAAAAQJKUkJCgyMhIxzZ27NgSbfLy8rRmzRolJyc79tlsNiUnJys9Pb3Ufj/99FMlJSVp4MCBio2NVZMmTfTcc8+psLCwwq7lbFAZAgAAALyJsYo2V/cpadeuXYqIiHDsLq0qtH//fhUWFio2NtZpf2xsrLZu3Vpq99u3b9eiRYvUu3dvzZs3Tz///LMeeOAB5efna+TIkS68kLIhGQIAAAAgSYqIiHBKhlzFbrcrJiZGU6ZMkZ+fn1q2bKndu3frxRdfJBkCAAAAcHbcvYBC9erV5efnp8zMTKf9mZmZiouLK/U1NWvWVEBAgPz8/Bz7GjVqpIyMDOXl5SkwMLBccZ8r5gwBAAAAOGuBgYFq2bKlFi5c6Nhnt9u1cOFCJSUllfqadu3a6eeff5bdbnfs+/HHH1WzZk23JUISyRAAAADgXUwFbWWQmpqqqVOn6u2339aWLVt0//33Kycnx7G6XJ8+fTRs2DBH+/vvv18HDx7U4MGD9eOPP2ru3Ll67rnnNHDgwHK+Ca7BMDkAAAAAZdKjRw/t27dPaWlpysjIUPPmzTV//nzHogo7d+6Uzfa/uktCQoK++OILDR06VM2aNVOtWrU0ePBgPf744+66BEmSZYyrRxziVLKzsxUZGamsrKwKmZgGAACAc+PJn9eKY/vHZ9MVGBbq0r7zco7p3Rvu9MjrrkgMkwMAAADgkxgmBwAAAHiTcszxOas+fRDJEAAAAOBNSIZchmFyAAAAAHwSyRAAAAAAn0QyBAAAAMAnMWcIAAAA8CrWn5ur+/Q9VIYAAAAA+CQqQwAAAIA3YTU5l3FpZSgnJ0dLly51ZZcAAAAAUCFcWhn6+eefdc0116iwsNCV3QIAAAAoRmXIZZgzBAAAAMAnlakyVLVq1dMepyIEAAAAVDRWk3OVMiVDubm5uv/++9W0adNSj//2228aPXq0SwIDAAAAUJIxRZur+/RFZUqGmjdvroSEBPXt27fU4xs2bCAZAgAAAOAVypQMdenSRYcPHz7l8apVq6pPnz7nGhMAAACAU2EBBZcp0wIKTz75pEaOHHnK4wkJCXrrrbfOOaiT7d69W//4xz9UrVo1hYSEqGnTplq9erXjuDFGaWlpqlmzpkJCQpScnKyffvrJqY+DBw+qd+/eioiIUFRUlPr376+jR486tfn+++/Vvn17BQcHKyEhQePGjSsRy+zZs9WwYUMFBweradOmmjdvnkuvFQAAAEDl8ejV5A4dOqR27dopICBAn3/+uTZv3qyXXnpJ0dHRjjbjxo3Tq6++qsmTJ2vFihUKCwtTSkqKTpw44WjTu3dv/fDDD1qwYIE+++wzLV26VPfcc4/jeHZ2tjp27Ki6detqzZo1evHFFzVq1ChNmTLF0Wb58uXq1auX+vfvr3Xr1qlbt27q1q2bNm3aVDlvBgAAACBJxqqYzQdZxpRvutQ777yjyZMna8eOHUpPT1fdunU1fvx41atXT3//+99dEtwTTzyhZcuW6Ztvvin1uDFG8fHxevjhh/XII49IkrKyshQbG6vp06erZ8+e2rJlixo3bqxVq1apVatWkqT58+fr+uuv1++//674+HhNmjRJw4cPV0ZGhgIDAx3nnjNnjrZu3SpJ6tGjh3JycvTZZ585zn/FFVeoefPmmjx58lldT3Z2tiIjI5WVlaWIiIhyvy8AAACoGJ78ea04tt7/944Cw0Jd2ndezjHN6H6HR153RSpXZWjSpElKTU3V9ddfr8OHDzuW1I6KitL48eNdFtynn36qVq1a6dZbb1VMTIxatGihqVOnOo7v2LFDGRkZSk5OduyLjIxUmzZtlJ6eLklKT09XVFSUIxGSpOTkZNlsNq1YscLR5qqrrnIkQpKUkpKibdu26dChQ442J5+nuE3xeUqTm5ur7Oxspw0AAAA4F1YFbb6oXMnQa6+9pqlTp2r48OHy8/Nz7G/VqpU2btzosuC2b9+uSZMm6aKLLtIXX3yh+++/Xw899JDefvttSVJGRoYkKTY21ul1sbGxjmMZGRmKiYlxOu7v76+qVas6tSmtj5PPcao2xcdLM3bsWEVGRjq2hISEMl0/AAAAgIpTrmRox44datGiRYn9QUFBysnJOeegitntdl122WV67rnn1KJFC91zzz0aMGDAWQ9Lc7dhw4YpKyvLse3atcvdIQEAAMDbmQrafFC5kqF69epp/fr1JfbPnz9fjRo1OteYHGrWrKnGjRs77WvUqJF27twpSYqLi5MkZWZmOrXJzMx0HIuLi9PevXudjhcUFOjgwYNObUrr4+RznKpN8fHSBAUFKSIiwmkDAAAAzhmJkEuUKxlKTU3VwIEDNWvWLBljtHLlSj377LMaNmyYHnvsMZcF165dO23bts1p348//qi6detKKkrK4uLitHDhQsfx7OxsrVixQklJSZKkpKQkHT58WGvWrHG0WbRokex2u9q0aeNos3TpUuXn5zvaLFiwQA0aNHCsXJeUlOR0nuI2xecBAAAA4F3KlQzdfffdeuGFFzRixAgdO3ZMt99+uyZNmqQJEyaoZ8+eLgtu6NCh+u677/Tcc8/p559/1syZMzVlyhQNHDhQkmRZloYMGaJnnnlGn376qTZu3Kg+ffooPj5e3bp1k1RUSerUqZMGDBiglStXatmyZRo0aJB69uyp+Ph4SdLtt9+uwMBA9e/fXz/88INmzZqlCRMmKDU11RHL4MGDNX/+fL300kvaunWrRo0apdWrV2vQoEEuu14AAAAApZs/f76+/fZbx/cTJ05U8+bNdfvttzsWPSurcj9nqHfv3vrpp5909OhRZWRk6Pfff1f//v3L212pWrdurY8//ljvvfeemjRpojFjxmj8+PHq3bu3o81jjz2mBx98UPfcc49at26to0ePav78+QoODna0mTFjhho2bKhrr71W119/va688kqnZwhFRkbqyy+/1I4dO9SyZUs9/PDDSktLc3oWUdu2bR3J2KWXXqoPP/xQc+bMUZMmTVx6zQAAAABKevTRRx2rM2/cuFEPP/ywrr/+eu3YscOpiFEW5X7OEMrOk9etBwAAgGd/XiuO7R+z31FgqIufM3TsmN691bOfMxQeHq5NmzYpMTFRo0aN0qZNm/Thhx9q7dq1uv7660+7yvOplKsydODAAQ0cOFCNGzdW9erVVbVqVacNAAAAAFwpMDBQx44dkyR99dVX6tixoySpatWq5X6ep395XnTHHXfo559/Vv/+/RUbGyvL8tXHNAEAAACVrCJWgPOCsWJXXnmlUlNT1a5dO61cuVKzZs2SVLTAWu3atcvVZ7mSoW+++UbffvutLr300nKdFAAAAADK4vXXX9cDDzygDz/8UJMmTVKtWrUkSZ9//rk6depUrj7LlQw1bNhQx48fL9cJAQAAAJwL68/N1X16tjp16uizzz4rsf+VV14pd5/lmjP0xhtvaPjw4fr666914MABZWdnO20AAAAAKoirH7jqJQ9eXbt2rTZu3Oj4/pNPPlG3bt305JNPKi8vr1x9lisZioqKUnZ2tv72t78pJiZG0dHRio6OVlRUlOMhpQAAAADgKvfee69+/PFHSdL27dvVs2dPhYaGavbs2XrsscfK1We5hsn17t1bAQEBmjlzJgsoAAAAAJXJRxdQ+PHHH9W8eXNJ0uzZs3XVVVdp5syZWrZsmXr27Knx48eXuc9yJUObNm3SunXr1KBBg/K8HAAAAADKxBgju90uqWhp7RtuuEGSlJCQoP3795erz3INk2vVqpV27dpVrhMCAAAAQFm1atVKzzzzjN555x19/fXX6tKliyRpx44dio2NLVef5aoMPfjggxo8eLAeffRRNW3aVAEBAU7HmzVrVq5gAAAAAKA048ePV+/evTVnzhwNHz5cF154oSTpww8/VNu2bcvVp2WMKfMIQZutZEHJsiwZY2RZlgoLC8sVzPkuOztbkZGRysrKUkREhLvDAQAAwF948ue14tjueO9dBYaGurTvvGPH9E6vf3jkdZ/JiRMn5OfnV6JAczbKVRnasWNHeV4GAPAyB/44pO/+u1rrF29S7vE8JV6SoKQbW6vh5ReyeA4AwCMEBweX+7XlSobq1q1b7hMCALzD9u9/078e+Y/27tqvgEB/+fn76fdte7Ty83XqdNff9PeBnUiIAMAtfPOhq4WFhXrllVf0wQcfaOfOnSWeLXTw4MEy93nWydCnn36qzp07KyAgQJ9++ulp2954441lDgQA4DlOHMvVtOEzte/3A4q/IFY2v/8Nj87an63P31yoOo1q67Jrm7oxSgDwUT66tPbo0aP15ptv6uGHH9aIESM0fPhw/frrr5ozZ47S0tLK1edZJ0PdunVTRkaGYmJi1K1bt1O2Y84QAHi/DUt+0B/b9yqmTnWnREiSIqtHaPcvGfrmw3SSIQBApZkxY4amTp2qLl26aNSoUerVq5fq16+vZs2a6bvvvtNDDz1U5j7Pemltu92umJgYx9en2kiEAMD77dyyW8bYFRBY+t/MwiNDtX3jTuXn5VdyZAAAR2XI1ZuHy8jIUNOmRX+ECw8PV1ZWliTphhtu0Ny5c8vVZ7meM/Sf//xHubm5Jfbn5eXpP//5T7kCAQB4DpvNOu0/jMZuZFmlry4KAEBFqF27tv744w9JUv369fXll19KklatWqWgoKBy9Vmuf8X69evnyMROduTIEfXr169cgQAAPEf95omy+fsp73heiWPGGB3NOqZGV1wsP38/N0QHAL7NqqDN0910001auHChpKLnnj711FO66KKL1KdPH911113l6rNcq8kVP0/or37//XdFRkaWKxAAQOUyBb9Ked9J9izJFikFtpXlX0eSdEm7Bkq8JEG/rN+h2MQYx3A5Y4wO7DmkkPBgXXVLkhujBwD4mueff97xdY8ePVSnTh2lp6froosuUteuXcvVZ5mSoRYtWsiyLFmWpWuvvVb+/v97eWFhoXbs2KFOnTqVKxAAQOUwpkAmZ5p0Yr5kcv534NhsmZAbZYXeoYDAAN39fG/965H/aOeW32WMkc1mU2FBocKjw3XL0BvUqM1F7rsIAPBlPrqa3F8lJSUpKenc/jBXpmSoeBW59evXKyUlReHh4Y5jgYGBSkxMVPfu3c8pIABAxTLH/k86/rFkqyLZ6kqWJRkjmYPS8VlFVaKQbopLjNFj0wdq3cJN+mHZVuWeyFNCg3hd3vky1bwg1t2XAQDwAWd6pM/JyvN4nzIlQyNHjpQkJSYmqkePHmd82ut7772nG2+8UWFhYWUODADgesZ+VMqdK1lBki36fwcsS7KqSYV5Msf/KwV3lmUFKSQ8RG3/3lpt/97afUEDAJz5UGXodI/0OVl5H+9TrgUU+vbte8ZESJLuvfdeZWZmlucUAICKULBVKtwv2aqWftxWVbJnSgU/VW5cAACU4nSP9HHF430qdE1UYzw0xQQAX2UKJNklnWoVOD/J2CXD84MAAJ5h0aJFaty4sbKzs0scy8rK0iWXXKJvvvmmXH3zgAgA8CV+tSUrTDJHSj9ujki2MMmvVuXGBQA4ez720NXx48drwIABioiIKHEsMjJS9957r15++eVy9U0yBAA+xPKvLQW2lOwH/6wSncTkS/bDUmCSLL8Yt8QHAMBfbdiw4bQrVnfs2FFr1qwpV9/les4QAMB7WWF3yRTuKZoXZAVJVrBkjksmT/JvLCu0r7tDBACcjg8toCBJmZmZCggIOOVxf39/7du3r1x9UxkCAB9j+cXJinxaVthdki1Wkk3yi5cVNkBW5GhZftXcHSIAAA61atXSpk2bTnn8+++/V82aNcvVd5krQ4WFhVq2bJmaNWumqKio07atW7fuabM4AIB7WLaqUmgPKeQ2SQWS/GVZlrvDAgCcBcsUba7u01Ndf/31euqpp9SpU6cSK1ofP35cI0eO1A033FCuvsucDPn5+aljx47asmXLGZOh02VwAAD3K0qA+KMVAMBzjRgxQh999JEuvvhiDRo0SA0aNJAkbd26VRMnTlRhYaGGDx9err7LNWeoSZMm2r59u+rVq1eukwIAAADA2YiNjdXy5ct1//33a9iwYY7H91iWpZSUFE2cOFGxsbHl6rtcydAzzzyjRx55RGPGjFHLli0VFhbmdLy0Ze8AAAAAnDtfGyYnFU2/mTdvng4dOqSff/5ZxhhddNFFio6OPqd+y5UMXX/99ZKkG2+80WmMuTFGlmWV+wmwAAAAAHAq0dHRat26tcv6K1cytHjxYpcFAAAAAKAMfGxp7YpUrmSoQ4cOro4DAAAAACrVWSdD33//vZo0aSKbzabvv//+tG2bNWt2zoEBAAAAQEU662SoefPmysjIUExMjJo3by7LshwrOZyMOUMAAAAAvMFZJ0M7duxQjRo1HF8DAAAAqHy+uJpcRTnrZKhu3bolvt68ebN27typvLw8xzHLspzaAgAAAIAnKtcCCtu3b9dNN92kjRs3Og2XK15mm2FyAAAAADydrTwvGjx4sOrVq6e9e/cqNDRUmzZt0tKlS9WqVSstWbLExSECAAAAcDAVtPmgclWG0tPTtWjRIlWvXl02m01+fn668sorNXbsWD300ENat26dq+MEAAAAAJcqV2WosLBQVapUkSRVr15de/bskVQ0l2jbtm2uiw4AAAAAKki5KkNNmjTRhg0bVK9ePbVp00bjxo1TYGCgpkyZogsuuMDVMQIAAACAy5UrGRoxYoRycnIkSU8//bRuuOEGtW/fXtWqVdOsWbNcGiAAAACAk1TEHB/mDJ29lJQUx9cXXnihtm7dqoMHDyo6OtqxohwAAAAAeLJyJUOlqVq1qqu6AgAAAHAqVIZcplwLKAAAAACAt3NZZQgAAABAxbMkWS6u5PjqRBcqQwAAAAB8EskQAAAAAJ/EMDkAAADAi1imAobJsYACAAAAAPgOKkMAAACAN2FpbZehMgQAAACgzCZOnKjExEQFBwerTZs2Wrly5Vm97v3335dlWerWrVvFBngWSIYAAAAAlMmsWbOUmpqqkSNHau3atbr00kuVkpKivXv3nvZ1v/76qx555BG1b9++kiI9PZIhAAAAAGXy8ssva8CAAerXr58aN26syZMnKzQ0VNOmTTvlawoLC9W7d2+NHj1aF1xwQSVGe2okQwAAAIA3MRW0ScrOznbacnNzS5w+Ly9Pa9asUXJysmOfzWZTcnKy0tPTTxn2008/rZiYGPXv3/9crt6lSIYAAAAASJISEhIUGRnp2MaOHVuizf79+1VYWKjY2Fin/bGxscrIyCi132+//Vb//ve/NXXq1AqJu7xYTQ4AAADwJhW4mtyuXbsUERHh2B0UFHTOXR85ckR33HGHpk6dqurVq59zf65EMgQAAABAkhQREeGUDJWmevXq8vPzU2ZmptP+zMxMxcXFlWj/yy+/6Ndff1XXrl0d++x2uyTJ399f27ZtU/369V0QfdkxTA4AAADAWQsMDFTLli21cOFCxz673a6FCxcqKSmpRPuGDRtq48aNWr9+vWO78cYbdc0112j9+vVKSEiozPCdUBkCAAAAvIhlijZX91kWqamp6tu3r1q1aqXLL79c48ePV05Ojvr16ydJ6tOnj2rVqqWxY8cqODhYTZo0cXp9VFSUJJXYX9lIhgAAAACUSY8ePbRv3z6lpaUpIyNDzZs31/z58x2LKuzcuVM2m+cPQiMZAgAAALxJBS6gUBaDBg3SoEGDSj22ZMmS0752+vTpZT9hBfD8dO0kzz//vCzL0pAhQxz7Tpw4oYEDB6patWoKDw9X9+7dS0zm2rlzp7p06aLQ0FDFxMTo0UcfVUFBgVObJUuW6LLLLlNQUJAuvPDCUm/QxIkTlZiYqODgYLVp00YrV66siMsEAAAAUAm8JhlatWqV/vWvf6lZs2ZO+4cOHar//ve/mj17tr7++mvt2bNHN998s+N4YWGhunTpory8PC1fvlxvv/22pk+frrS0NEebHTt2qEuXLo5JXEOGDNHdd9+tL774wtFm1qxZSk1N1ciRI7V27VpdeumlSklJ0d69eyv+4gEAAIA/Fc8ZcvXmi7wiGTp69Kh69+6tqVOnKjo62rE/KytL//73v/Xyyy/rb3/7m1q2bKm33npLy5cv13fffSdJ+vLLL7V582a9++67at68uTp37qwxY8Zo4sSJysvLkyRNnjxZ9erV00svvaRGjRpp0KBBuuWWW/TKK684zvXyyy9rwIAB6tevnxo3bqzJkycrNDRU06ZNq9w3AwAAAIBLeEUyNHDgQHXp0kXJyclO+9esWaP8/Hyn/Q0bNlSdOnWUnp4uSUpPT1fTpk2dnpCbkpKi7Oxs/fDDD442f+07JSXF0UdeXp7WrFnj1MZmsyk5OdnRpjS5ubnKzs522gAAAAB4Bo9fQOH999/X2rVrtWrVqhLHMjIyFBgY6Fiar1hsbKwyMjIcbU5OhIqPFx87XZvs7GwdP35chw4dUmFhYalttm7desrYx44dq9GjR5/dhQIAAABnw0MWUDgfeHRlaNeuXRo8eLBmzJih4OBgd4dTZsOGDVNWVpZj27Vrl7tDAgAAAPAnj06G1qxZo7179+qyyy6Tv7+//P399fXXX+vVV1+Vv7+/YmNjlZeXp8OHDzu9LjMzU3FxcZKkuLi4EqvLFX9/pjYREREKCQlR9erV5efnV2qb4j5KExQUpIiICKcNAAAAgGfw6GTo2muv1caNG7V+/XrH1qpVK/Xu3dvxdUBAgBYuXOh4zbZt27Rz504lJSVJkpKSkrRx40anVd8WLFigiIgINW7c2NHm5D6K2xT3ERgYqJYtWzq1sdvtWrhwoaMNAAAAAO/i0XOGqlSpoiZNmjjtCwsLU7Vq1Rz7+/fvr9TUVFWtWlURERF68MEHlZSUpCuuuEKS1LFjRzVu3Fh33HGHxo0bp4yMDI0YMUIDBw5UUFCQJOm+++7T66+/rscee0x33XWXFi1apA8++EBz5851nDc1NVV9+/ZVq1atdPnll2v8+PHKyclRv379KundAAAAAMScIRfy6GTobLzyyiuy2Wzq3r27cnNzlZKSojfeeMNx3M/PT5999pnuv/9+JSUlKSwsTH379tXTTz/taFOvXj3NnTtXQ4cO1YQJE1S7dm29+eabSklJcbTp0aOH9u3bp7S0NGVkZKh58+aaP39+iUUVAAAAAHgHyxjjo3lg5cvOzlZkZKSysrKYPwQAAOCBPPnzWnFs90yYocCQUJf2nXf8mKYM7u2R112RPHrOEAAAAABUFK8fJgcAAAD4FOYMuQzJEAAAAOBNSIZchmFyAAAAAHwSlSEAAADAi1h/bq7u0xdRGQIAAADgk6gMAQAAAN6EOUMuQ2UIAAAAgE+iMgQAAAB4EyNZVIZcgsoQAAAAAJ9EZQgAAADwJswZchkqQwAAAAB8EskQAAAAAJ/EMDkAAADAmzBMzmWoDAEAAADwSVSGAAAAAG9CZchlqAwBAAAA8ElUhgBUCmPypPwNkv2gZFWRAprLsoW6OywAALyO9efm6j59EckQgApncpfJ5EyXCvdIKpRkk/xipdCeUlBHWZav/icYAAC4E8kQgApl8lbKHHlZMickvxqSFSSZfMm+T+boG7LkJwUnuztMAAC8B3OGXIY5QwAqjDF2mWPvSyZH8qtVlAhJkhUg+cVLKpQ59kHREDoAAHBWLEmWcfHm7otyE5IhABWncLtU8Itkqy6VNhTOVl0q3C3lb6r82AAAgM8jGQJQcew5kimQrMBTNAiUVFBUOQIAAKhkJEMAKo6t2p9zhI6VftwcLzpuq1a5cQEAAIhkCEAFsvxrSwHNi5bTNnbng8ZI9r2S/0WSf0O3xAcAgFcyFbT5IFaTA1ChrLB/yBT+IhX+JtmiJStYMrlFCZKtuqywfrIs/i4DAAAqH59AAFQoy/8CWRGjpaBkSXbJfqhoae3AK2VFpMkKaOLuEAEA8C5UhlyGyhCACmf5J8qKeEymcJ9kPyzZqsjyi3N3WAAAwMeRDAGoNJZfjaIHrwIAgHIrfjaQq/v0RQyTAwAAAOCTSIYAAAAA+CSGyQE4LWM/WrT6m62KrFM+PBUAAFQWhsm5DskQgFKZ/B9lTnwi5a2WTIFki5SCr5OCb5Blq+Lu8AAAAM4ZyRCAEkzeWpkjL0r2A0VJkBUk2Q/I5EyX8tZJEU+REAEA4C4VsRS2j1aGmDMEwIkxeTJHJxctge2XKNmqSrYqkl+c5FdTyt8gHf/E3WECAACcM5IhAM7yVkuFvxclP5blfMwKkqwQmdyvZEyue+IDAABwEZIhAM7sGZKMdKrFEqxwyZ4t2Q9ValgAAACuxpwhAH8RJMlIxi5Zpf29JF+y/E6dLAEAgIplTNHm6j59EMkQAGeBl0m2cMkclqyqzseMkexZUuCVsmxVS305AACoWCyt7ToMkwPgxPKrKQUlS/YjRUPhjL3ogMmXCndLtnBZoTe6N0gAAAAXoDIEoAQr7C4Z2aUTCyX7Lkl/LqTgFysrbICsgGZujQ8AAMAVSIYAH2HsB6Xc7ySTXfTsoMA2pxzqZlmBssIfkAm+UcpfI5njki32z9eEVnLkAAAAFYNkCDjPGWOk4x/JHJ9d9OwgWZKMZIuSQm+Tgm+S9dcltP9k+deW/GtXYrQAAOBMLHvR5uo+fRHJEHC+OzFP5thbkgIkv9pFK8GZQsl+QCZnmiwrVAru5O4oAQAAKh0LKADnMWNyZY5/LMmS/GKKEiGp6P/9YiRZMsc+kjF57gwTAACUhamgzQeRDAHns/ytRQ9RtVUr/bitqmTfIxVsq9y4AAAAPADD5IDzWm7RkLhT/qoHFB03uZUZFAAAOAeWHOu8urRPX0QyBJzPbHGSFSKZHMmKKHncHJWs0KJ2AADAOxhTtLm6Tx/EMDngPGb515ECLpXsB/6sEJ3EFEr2g1Jgi6JV4wAAAHwMlSHgPGeF9ZMp/F0q/E2ywiUrWDIniqpCfomyQu90d4gAAKAsKmLBA98sDJEMAec7y7+OFDlG5vh/pdyvix6gaoVKwZ1lhXSV5ccQOQAA4JtIhgAfYPnFyQofIBN2h2TPkWzhsqwgd4cFAADgViRDgA+xrGDJL9jdYQAAAHgEkiEAAADAi1imaHN1n76I1eQAL5Cfl6+crBwVFhSeuTEAAADOCpUhwINl/LpXS2Yt06r565V/Il9Vqoar7d9bq8NtbRUeFebu8AAAgDuwmpzLkAwBHuq3zbv0xpC3lLlzv8IiQhUQ5K8Dew7p/175TN8v3ayBr96liKpV3B0mAACoZNafm6v79EUMkwM8UGHeTs145k3t3blH8RfEqGpclKpEh6tGQjXF1KmuH1f/os/fXOjuMAEAALwayRDgQUzBTtmzntHP3wzRbxs3qmqNvbKZzTKF+2RMUf06IChAoRGhWjFvrXKyctwcMQAAqHSmgjYfRDIEeAhT+IdM9tNS3lJl/h6g/LwAhYQFSOaEVLhDsu91tA2LDNWxrOPav+eQGyMGAADwbiRDgIcwxz+WCndKfnUUGBIqSbLbbZIVLMmSCnfLmAJJUmF+oWx+NgUGBbgxYgAA4BZUhlyGZAjwAMaeI+V+K1lVJMtPDS7zV5VoS1kHiv/LFCiZfMleVAk6vC9LCQ3jFVcvxn1BAwAAeDmSIcATmCNFw+GsYElSZDWbrrwxUMdzjLIO2B1/rLEX5mvf7wcUEBSg5Ds6yLJ8de0XAAB8mDEVs/kgltYGPIEVLlmBksmVFC5JurF/sPJPSMvm5mnPdrssy5KxjimyRqS6DeqkVh0vdW/MAAAAXo5kCPAAli1cJvBK6cSnkomSLD8FBFrqOTREHW4K1PffHtDxo4GKrtdDlyW3UWT1CHeHDAAA3Mk3CzkuRzIEeAgrpJtM/pqiRRRs1SUrTFK+aiYcUM1eRlZYH1kh17k7TAAAgPOGx88ZGjt2rFq3bq0qVaooJiZG3bp107Zt25zanDhxQgMHDlS1atUUHh6u7t27KzMz06nNzp071aVLF4WGhiomJkaPPvqoCgoKnNosWbJEl112mYKCgnThhRdq+vTpJeKZOHGiEhMTFRwcrDZt2mjlypUuv2b4Jsu/tqyINCmwjWSOS4W7JPt+yRYrK+wBKbibu0MEAAAewDIVs/kij0+Gvv76aw0cOFDfffedFixYoPz8fHXs2FE5Of972OTQoUP13//+V7Nnz9bXX3+tPXv26Oabb3YcLywsVJcuXZSXl6fly5fr7bff1vTp05WWluZos2PHDnXp0kXXXHON1q9fryFDhujuu+/WF1984Wgza9YspaamauTIkVq7dq0uvfRSpaSkaO/e/z3/BTgXln89WRFPy4p6WVbEKFkRz8qKnigr5HoWSwAAAH9ibW1XsYzxrqUj9u3bp5iYGH399de66qqrlJWVpRo1amjmzJm65ZZbJElbt25Vo0aNlJ6eriuuuEKff/65brjhBu3Zs0exsbGSpMmTJ+vxxx/Xvn37FBgYqMcff1xz587Vpk2bHOfq2bOnDh8+rPnz50uS2rRpo9atW+v111+XJNntdiUkJOjBBx/UE088USLW3Nxc5ebmOr7Pzs5WQkKCsrKyFBHBnA8AAABPk52drcjISI/8vFYc29DH31ZQUKhL+87NPaZXXujrkdddkTy+MvRXWVlZkqSqVatKktasWaP8/HwlJyc72jRs2FB16tRRenq6JCk9PV1NmzZ1JEKSlJKSouzsbP3www+ONif3UdymuI+8vDytWbPGqY3NZlNycrKjzV+NHTtWkZGRji0hIeFcLx8AAAA+jmFyruNVyZDdbteQIUPUrl07NWnSRJKUkZGhwMBARUVFObWNjY1VRkaGo83JiVDx8eJjp2uTnZ2t48ePa//+/SosLCy1TXEffzVs2DBlZWU5tl27dpXvwgEAAAAPU5a59FOnTlX79u0VHR2t6OhoJScne8Tce69KhgYOHKhNmzbp/fffd3coZyUoKEgRERFOGwAAAHBOPGDKUFnn0i9ZskS9evXS4sWLlZ6eroSEBHXs2FG7d+8u24ldzGuSoUGDBumzzz7T4sWLVbt2bcf+uLg45eXl6fDhw07tMzMzFRcX52jz19Xlir8/U5uIiAiFhISoevXq8vPzK7VNcR8AAACAL3j55Zc1YMAA9evXT40bN9bkyZMVGhqqadOmldp+xowZeuCBB9S8eXM1bNhQb775pux2uxYuXFjJkTvz+GTIGKNBgwbp448/1qJFi1SvXj2n4y1btlRAQIDTG7lt2zbt3LlTSUlJkqSkpCRt3LjRKVNdsGCBIiIi1LhxY0ebv96MBQsWOPoIDAxUy5YtndoU38DiNgAAAEDFq7jSUHZ2ttN28mJgxcozl/6vjh07pvz8fMc6AO7i8cnQwIED9e6772rmzJmqUqWKMjIylJGRoePHj0uSIiMj1b9/f6Wmpmrx4sVas2aN+vXrp6SkJF1xxRWSpI4dO6px48a64447tGHDBn3xxRcaMWKEBg4cqKCgIEnSfffdp+3bt+uxxx7T1q1b9cYbb+iDDz7Q0KFDHbGkpqZq6tSpevvtt7Vlyxbdf//9ysnJUb9+/Sr/jQEAAABcLCEhwWkBsLFjx5ZoU5659H/1+OOPKz4+vsQCZpXN361nPwuTJk2SJF199dVO+9966y3deeedkqRXXnlFNptN3bt3V25urlJSUvTGG2842vr5+emzzz7T/fffr6SkJIWFhalv3756+umnHW3q1aunuXPnaujQoZowYYJq166tN998UykpKY42PXr00L59+5SWlqaMjAw1b95c8+fPL/GDAAAAAFQY+5+bq/uUtGvXLqd57sWFA1d6/vnn9f7772vJkiUKDg52ef9l4fHJ0Nk8Bik4OFgTJ07UxIkTT9mmbt26mjdv3mn7ufrqq7Vu3brTthk0aJAGDRp0xpgAAACAilARS2EX93c2i36dy1z6f/7zn3r++ef11VdfqVmzZucUsyt4/DA5AAAAAJ6jvHPpx40bpzFjxmj+/Plq1apVZYR6Rh5fGQIAAADgWVJTU9W3b1+1atVKl19+ucaPH+80l75Pnz6qVauWY87RCy+8oLS0NM2cOVOJiYmOuUXh4eEKDw9323WQDAEAAAAokzPNpd+5c6dstv8NQps0aZLy8vJ0yy23OPUzcuRIjRo1qjJDd0IyBAAAAHgTY4o2V/dZRqebS79kyRKn73/99ddyBFXxmDMEAAAAwCdRGQIAAAC8SEWuJudrqAwBAAAA8ElUhgAAAACvYv7cXN2n7yEZAgAAALyIZS/aXN2nL2KYHAAAAACfRDIEAAAAwCeRDAEAAADwScwZAgAAALwJ6ye4DJUhAAAAAD6JyhDOa/a8DdLxjyWTJdlipNBesvknujssAACA8jOmaHN1nz6IZAjnJbv9hHQ4VcpbJqngfweOz5Y9pLusKk/Ksiy3xQcAAAD3IxnC+SlrmJT3tYp+xMMlm02yF0rmuHTsfRlbtKzwB9wdJQAAQNlRGXIZ5gzhvGMv2PVnIuQn2UKLEiFJsvlJVpgku3Rsluz2XHeGCQAAUD6mgjYfRDKE88+JuZLJlRRc8phlSQqS7IekvFWVHRkAAAA8CMPkcP4xx4r+3zpVrm+TZJfM0cqKCAAAwGUsU7S5uk9fRGUI5x//iyXZJJN/igb5kgIk/waVGBQAAAA8DckQzj9BHSVbDUknJLvd+Zi9QFK+FNBEtoB67ogOAADgHDFpyFVIhnDesdkCpSrDJCtU0hHJniPZTxT9v3IkW3UpIs3dYQIAAMDNmDOE85It5DrZbVWko5Okgo2SKZSsICmwg1RlqGz+F7g7RAAAgPJhaW2XIRnCecsWdIUUdIXshYck+2HJL1Y2W6i7wwIAAICHIBnCec/mFy35Rbs7DAAAANeoiCk+vlkYIhmC5zLGSAWbZXK/lQr3SrYoWUFJUkBzWRY/ugAAwDdZkiwXD2uzXNqb9+ATJTySMQUyRydLuQskc0KSn6RCmdwvpMAkKXyoLIa8AQAA4ByQDMHjbP/+N30y4Q39uvFHBQQFqnVypK65JUjV421FK8Llfi1ji5YV/oC7QwUAAKh8DJNzGZIheJS3R87SRxM+U97xE5LlJ0nasvq45v3nhIaMD9OlV4ZJiihKiEJuk+VX3b0BAwAAwGvxnCF4BHthlmaOGqVZ4z5U7rHjCg0vUHiEUVgVyT9Q2rvbaEJqjg78YZesqKLV4Qq2uTtsAACAyle8tLarNx9EMgS32/79Rj3f60HNGLdJ+blGxkjHjtqUd8JIKlRgkBQQJO3bbfTNZ7n63xS/QjdGDQAAAG9HMgS3sdvtmjt1gYaljNW3nx5T3glLf66PIrvd0onjNuXlSlKhAgMku11avShfMtmSFS758eBUAADgi0wFbb6HOUNwC7vdrlkv/J/+75W5ysnKl3+glJ9XfEyy2STJKO+EpYBAu6w/i0H2fLtkPyQFXSvLv7a7wgcAAMB5gGQIlW7vL5/o4/HvacF7uco5apN/gFFgkKXc45aMXZKR7HZLfn6SsUsF+ZZsfnZZstSg+XEpoJms8AHuvgwAAAD3sP+5ubpPH0QyhEr19btj9H+vrtDOnwJ0Iscme6GlwgJL9kKjgAC78nL/HLlpJGMsyTKy2y3lnbApLNKmLvf3kRXRSZYt3L0XAgAAAK9HMoRKceRQtj5+aYI+mbxW9sIA5ecWVYQKZMkYo4J8S34BRn7+RoUFRWPi7IWSLCnvhE0h4SEaMK6fYi68xr0XAgAAgPMGyRAq3J5t32ryw5P1/bfHlHvcJv9Ao4I8yRib/PyNjL2oAlRYYCk0zK78Aj/lHi96rZ+/1PyqGro9bZCatW/s3gsBAADwBBWxFLaPLq1NMoQKY4zR6vlfaOJD/9ahvXYV5EuBwXYFBhWtGXfimKWCAkt+fkUJkTFSQaHkH2BXYYGlarFS37QmSr5ruCwrwN2XAwAAgPMMyRAqxM8bftHTNz+jjN+OytiNilZxN8rPNbIsuwICjfLyjOwFlmRJAUFGebmWCvIsFeb7KTomWA+M76G2N3Vx96UAAAB4FipDLkMyBJfbtnSYHrl+m04cO/kxVkW/YMZYOnbET2GRhQoJs+v40aJFFORnZLMZRVa16aLLL9ODr/VXTJ0a7rkAAAAA+ASSIbhMYUGhPn/tbk1Jy3IkQpataHnskxkj5R6zKSzCruBQuwryLQWHGkXXKNQDE3qp2TW3yc/fzw1XAAAA4AUq4hmpvlkYIhmCa+za9q2Gd35FmTslu/1/FSFHImTJ6ZcsP89SXq6Un2dTQIBRvUb5umdcJ110Ra9KjRsAAAC+i2QI52zZrGf0XL+1yjthO3UjI1nW/4ajWpZkL7Tk7290za1V1XPEQ4qr16xyAgYAAPBmzBlyGZIhlFtB/gm9NuBufTHjuAoLT5MI/enk3zE/f6n2hTZ1uutK3fjgYFmWVYGRAgAAnEdIhlyGZAjlsmP1FI3q8V/t2RGkojFwZ2azSfY/h801bRuie8bdqvqtupIIAQAAwC1IhlBmGevv1IibDmrv7qAyva44EarTsIZGz31FIaHBFRAdAADAeY7KkMuceWwT8Kf8E5na8MmVerrfXu37I7DMr7f5SU2SQvXaipdIhAAAAOB2VIZwVvb/9Jaev+t9/bq1mrIP+ZdYLvtMatbNVerrddS8y8SKCRAAAMBXUBlyGZIhnNErfW7Q/PeCZC+sUq7XJzYs1D8XP6/I2EtcHBkAAABQfiRDOKUTOVm6v/kd+v2X8g1ps2zSFSmWHn9/psKqhLg4OgAAAF/FU1ddhWQIpfp1zb/16PWf6vC+ss8NkqTwyAI9+EoT/e3OZ1wcGQAAAOAaJENwYozRM91v09I5RlJAufqodUGuxn4xRjXrN3dpbAAAAJBk/3NzdZ8+iGQIDj+v+VL3X/4vyZT9uT+WTQqPKFTbzkc1cPL7CqlSrQIiBAAAAMPkXIdkCJKk90berWljDutsH6B6srCIAnW48bD6Dm+gqg0+dnlsAAAAQEUgGfJxhYWFGvP3m7VsXvmGxFmWUcfbDqn/uNEKirratcEBAACgJJbWdhmSIR+28vOFGt7lDZV3bpAkdf7HYd09fq4CQ8NdFxgAAABQCUiGfNSDbfpo66pjKs+wOEmyLOmWBwp0z2sLXRsYAAAATo8pQy5DMuRjCvLzdUPY7SosKP9PfPWauXrm0/tVv+X1LowMAAAAqFw2dweAyrP2iy/VOajHOSVCf+ueo/9sn0UiBAAA4C7GXjGbD6Iy5COeu/0xLX5/u84l/73p/gg9MPFD1wUFAAAAuBHJ0HkuNzdX3av1Uu6xcysCPvavFrpuwHAXRQUAAIByYzU5lyEZOo9tWrZYQ9u/rnOpBoWG5+s/O6Yrslq06wIDAAAAPABzhs5Ts1+aqqHtJ6q8q8VJ0t2jo/VJ9ickQgAAAJ7GuHjzUVSGzjPGGH34yhxNefSLc+rn8bdvUfIdPV0UFQAAAFyGYXIuQ2WojCZOnKjExEQFBwerTZs2WrlypbtDcvLVu0v19lPvn0MPds3eN5VECAAAAOc9kqEymDVrllJTUzVy5EitXbtWl156qVJSUrR37153hyZJOpR5WJ9MnK/cE+VbGjH+giAtsH+kKIbFAQAAeK7iypCrNx9EMlQGL7/8sgYMGKB+/fqpcePGmjx5skJDQzVt2jR3hyZJWvvVRmXvP1KOcZ9GL33VQ2//PKMiwgIAAAA8EnOGzlJeXp7WrFmjYcOGOfbZbDYlJycrPT291Nfk5uYqNzfX8X12dnaFxpi1L1s2P5v8/KXCgrN7jX+gXR8dmqGQkJAKjQ0AAAAuwpwhl6EydJb279+vwsJCxcbGOu2PjY1VRkZGqa8ZO3asIiMjHVtCQkKFxhgaESKbzVJkjYizat8kKVCfn/iIRAgAAAA+iWSoAg0bNkxZWVmObdeuXRV6vkuvvkQhVUIUFROtgCC/07Q0Gr+kt15ZNrNC4wEAAEAFMKqAOUPuvij3IBk6S9WrV5efn58yMzOd9mdmZiouLq7U1wQFBSkiIsJpq0g1L4hV++5XqCCvQPH14xUWFay/Pmfo4pYRmnviPV1y1U0VGgsAAADg6ZgzdJYCAwPVsmVLLVy4UN26dZMk2e12LVy4UIMGDXJvcCe5JfUGBYcFaensdNntRuFRebJkKf7iON33Ul/Vu6SOu0MEAADAuWDOkMuQDJVBamqq+vbtq1atWunyyy/X+PHjlZOTo379+rk7NAf/AH91G9RZ1/Zurx9X/6L83ALF1q2uxCZ1ZFnWmTsAAACAZyMZchmSoTLo0aOH9u3bp7S0NGVkZKh58+aaP39+iUUVPEGV6HC1vO5Sd4cBAAAAeCySoTIaNGiQRw2LAwAAgG8xxsi4uJLj6v68BQsoAAAAAPBJVIYAAAAAb8KcIZehMgQAAADAJ1EZAgAAALwJlSGXoTIEAAAAwCdRGQIAAAC8CZUhlyEZAgAAALyIsRsZu4uX1nZxf96CYXIAAAAAfBKVIQAAAMCrmD83V/fpe6gMAQAAAPBJJEMAAACAN7GbitnKaOLEiUpMTFRwcLDatGmjlStXnrb97Nmz1bBhQwUHB6tp06aaN29eed8BlyEZAgAAAFAms2bNUmpqqkaOHKm1a9fq0ksvVUpKivbu3Vtq++XLl6tXr17q37+/1q1bp27duqlbt27atGlTJUfujGQIAAAA8Cqmgraz9/LLL2vAgAHq16+fGjdurMmTJys0NFTTpk0rtf2ECRPUqVMnPfroo2rUqJHGjBmjyy67TK+//noZr921WEChEpk/12/Pzs52cyQAAAAoTfHnNOPBz93JLcitsD7/+jk1KChIQUFBTvvy8vK0Zs0aDRs2zLHPZrMpOTlZ6enppfafnp6u1NRUp30pKSmaM2eOC6IvP5KhSnTkyBFJUkJCgpsjAQAAwOkcOXJEkZGR7g7DSWBgoOLi4vTqV89XSP/h4eElPqeOHDlSo0aNctq3f/9+FRYWKjY21ml/bGystm7dWmrfGRkZpbbPyMg498DPAclQJYqPj9euXbtUpUoVWZZV4efLzs5WQkKCdu3apYiIiAo/H84O98VzcW88F/fGM3FfPBf3pvyMMTpy5Iji4+PdHUoJwcHB2rFjh/Ly8iqkf2NMic+of60KnW9IhiqRzWZT7dq1K/28ERER/IfQA3FfPBf3xnNxbzwT98VzcW/Kx9MqQicLDg5WcHCwW2OoXr26/Pz8lJmZ6bQ/MzNTcXFxpb4mLi6uTO0rCwsoAAAAADhrgYGBatmypRYuXOjYZ7fbtXDhQiUlJZX6mqSkJKf2krRgwYJTtq8sVIYAAAAAlElqaqr69u2rVq1a6fLLL9f48eOVk5Ojfv36SZL69OmjWrVqaezYsZKkwYMHq0OHDnrppZfUpUsXvf/++1q9erWmTJnizssgGTqfBQUFaeTIkef9WE9vw33xXNwbz8W98UzcF8/FvUFF69Gjh/bt26e0tDRlZGSoefPmmj9/vmORhJ07d8pm+98gtLZt22rmzJkaMWKEnnzySV100UWaM2eOmjRp4q5LkCRZxpPXDQQAAACACsKcIQAAAAA+iWQIAAAAgE8iGQIAAADgk0iGAAAAAPgkkqHz1MSJE5WYmKjg4GC1adNGK1eudHdIXm3s2LFq3bq1qlSpopiYGHXr1k3btm1zanPixAkNHDhQ1apVU3h4uLp3717i4WI7d+5Uly5dFBoaqpiYGD366KMqKChwarNkyRJddtllCgoK0oUXXqjp06eXiIf7W7rnn39elmVpyJAhjn3cF/fZvXu3/vGPf6hatWoKCQlR06ZNtXr1asdxY4zS0tJUs2ZNhYSEKDk5WT/99JNTHwcPHlTv3r0VERGhqKgo9e/fX0ePHnVq8/3336t9+/YKDg5WQkKCxo0bVyKW2bNnq2HDhgoODlbTpk01b968irloD1dYWKinnnpK9erVU0hIiOrXr68xY8bo5LWUuC+VY+nSperatavi4+NlWZbmzJnjdNyT7sPZxAJ4LYPzzvvvv28CAwPNtGnTzA8//GAGDBhgoqKiTGZmprtD81opKSnmrbfeMps2bTLr1683119/valTp445evSoo819991nEhISzMKFC83q1avNFVdcYdq2bes4XlBQYJo0aWKSk5PNunXrzLx580z16tXNsGHDHG22b99uQkNDTWpqqtm8ebN57bXXjJ+fn5k/f76jDfe3dCtXrjSJiYmmWbNmZvDgwY793Bf3OHjwoKlbt6658847zYoVK8z27dvNF198YX7++WdHm+eff95ERkaaOXPmmA0bNpgbb7zR1KtXzxw/ftzRplOnTubSSy813333nfnmm2/MhRdeaHr16uU4npWVZWJjY03v3r3Npk2bzHvvvWdCQkLMv/71L0ebZcuWGT8/PzNu3DizefNmM2LECBMQEGA2btxYOW+GB3n22WdNtWrVzGeffWZ27NhhZs+ebcLDw82ECRMcbbgvlWPevHlm+PDh5qOPPjKSzMcff+x03JPuw9nEAngrkqHz0OWXX24GDhzo+L6wsNDEx8ebsWPHujGq88vevXuNJPP1118bY4w5fPiwCQgIMLNnz3a02bJli5Fk0tPTjTFF//DZbDaTkZHhaDNp0iQTERFhcnNzjTHGPPbYY+aSSy5xOlePHj1MSkqK43vub0lHjhwxF110kVmwYIHp0KGDIxnivrjP448/bq688spTHrfb7SYuLs68+OKLjn2HDx82QUFB5r333jPGGLN582YjyaxatcrR5vPPPzeWZZndu3cbY4x54403THR0tONeFZ+7QYMGju9vu+0206VLF6fzt2nTxtx7773ndpFeqEuXLuauu+5y2nfzzTeb3r17G2O4L+7y12TIk+7D2cQCeDOGyZ1n8vLytGbNGiUnJzv22Ww2JScnKz093Y2RnV+ysrIkSVWrVpUkrVmzRvn5+U7ve8OGDVWnTh3H+56enq6mTZs6HkYmSSkpKcrOztYPP/zgaHNyH8Vtivvg/pZu4MCB6tKlS4n3jvviPp9++qlatWqlW2+9VTExMWrRooWmTp3qOL5jxw5lZGQ4vWeRkZFq06aN072JiopSq1atHG2Sk5Nls9m0YsUKR5urrrpKgYGBjjYpKSnatm2bDh065GhzuvvnS9q2bauFCxfqxx9/lCRt2LBB3377rTp37iyJ++IpPOk+nE0sgDcjGTrP7N+/X4WFhU4f7CQpNjZWGRkZborq/GK32zVkyBC1a9fO8dTkjIwMBQYGKioqyqntye97RkZGqfel+Njp2mRnZ+v48ePc31K8//77Wrt2rcaOHVviGPfFfbZv365Jkybpoosu0hdffKH7779fDz30kN5++21J/3tvT/eeZWRkKCYmxum4v7+/qlat6pL754v35oknnlDPnj3VsGFDBQQEqEWLFhoyZIh69+4tifviKTzpPpxNLIA383d3AIC3GThwoDZt2qRvv/3W3aH4vF27dmnw4MFasGCBgoOD3R0OTmK329WqVSs999xzkqQWLVpo06ZNmjx5svr27evm6HzXBx98oBkzZmjmzJm65JJLtH79eg0ZMkTx8fHcFwA+icrQeaZ69ery8/MrsVpWZmam4uLi3BTV+WPQoEH67LPPtHjxYtWuXduxPy4uTnl5eTp8+LBT+5Pf97i4uFLvS/Gx07WJiIhQSEgI9/cv1qxZo7179+qyyy6Tv7+//P399fXXX+vVV1+Vv7+/YmNjuS9uUrNmTTVu3NhpX6NGjbRz505J/3tvT/eexcXFae/evU7HCwoKdPDgQZfcP1+8N48++qijOtS0aVPdcccdGjp0qKOyyn3xDJ50H84mFsCbkQydZwIDA9WyZUstXLjQsc9ut2vhwoVKSkpyY2TezRijQYMG6eOPP9aiRYtUr149p+MtW7ZUQECA0/u+bds27dy50/G+JyUlaePGjU7/eC1YsEARERGOD41JSUlOfRS3Ke6D++vs2muv1caNG7V+/XrH1qpVK/Xu3dvxNffFPdq1a1di+fkff/xRdevWlSTVq1dPcXFxTu9Zdna2VqxY4XRvDh8+rDVr1jjaLFq0SHa7XW3atHG0Wbp0qfLz8x1tFixYoAYNGig6OtrR5nT3z5ccO3ZMNpvzP/1+fn6y2+2SuC+ewpPuw9nEAng1d6/gANd7//33TVBQkJk+fbrZvHmzueeee0xUVJTTalkom/vvv99ERkaaJUuWmD/++MOxHTt2zNHmvvvuM3Xq1DGLFi0yq1evNklJSSYpKclxvHgJ544dO5r169eb+fPnmxo1apS6hPOjjz5qtmzZYiZOnFjqEs7c31M7eTU5Y7gv7rJy5Urj7+9vnn32WfPTTz+ZGTNmmNDQUPPuu+862jz//PMmKirKfPLJJ+b77783f//730tdOrhFixZmxYoV5ttvvzUXXXSR09LBhw8fNrGxseaOO+4wmzZtMu+//74JDQ0tsXSwv7+/+ec//2m2bNliRo4c6VNLOJ+sb9++platWo6ltT/66CNTvXp189hjjznacF8qx5EjR8y6devMunXrjCTz8ssvm3Xr1pnffvvNGONZ9+FsYgG8FcnQeeq1114zderUMYGBgebyyy833333nbtD8mqSSt3eeustR5vjx4+bBx54wERHR5vQ0FBz0003mT/++MOpn19//dV07tzZhISEmOrVq5uHH37Y5OfnO7VZvHixad68uQkMDDQXXHCB0zmKcX9P7a/JEPfFff773/+aJk2amKCgINOwYUMzZcoUp+N2u9089dRTJjY21gQFBZlrr73WbNu2zanNgQMHTK9evUx4eLiJiIgw/fr1M0eOHHFqs2HDBnPllVeaoKAgU6tWLfP888+XiOWDDz4wF198sQkMDDSXXHKJmTt3rusv2AtkZ2ebwYMHmzp16pjg4GBzwQUXmOHDhzstvcx9qRyLFy8u9d+Vvn37GmM86z6cTSyAt7KMOemx0wAAAADgI5gzBAAAAMAnkQwBAAAA8EkkQwAAAAB8EskQAAAAAJ9EMgQAAADAJ5EMAQAAAPBJJEMAAAAAfBLJEAAAAACfRDIEAOe5X3/9VZZlaf369S7v27IszZkzx+X9AgBQGUiGAOA8l5CQoD/++ENNmjQpdx+jRo1S8+bNS+z/448/1Llz53OI7uw8++yzatu2rUJDQxUVFVXh5wMA+AaSIQA4z/n5+SkuLk7+/v6lHjfGqKCgoFx9x8XFKSgo6FzCOyt5eXm69dZbdf/991f4uQAAvoNkCADOA/Pnz9eVV16pqKgoVatWTTfccIN++eUXSSWHyS1ZskSWZenzzz9Xy5YtFRQUpG+//faUfU+fPl2jR4/Whg0bZFmWLMvS9OnTJTkPkys+zwcffKD27dsrJCRErVu31o8//qhVq1apVatWCg8PV+fOnbVv3z6nc7z55ptq1KiRgoOD1bBhQ73xxhtOx0ePHq2hQ4eqadOmrnnDAACQVPqfCQEAXiUnJ0epqalq1qyZjh49qrS0NN10002nnSf0xBNP6J///KcuuOACRUdHn7Jdjx49tGnTJs2fP19fffWVJCkyMvKU7UeOHKnx48erTp06uuuuu3T77berSpUqmjBhgkJDQ3XbbbcpLS1NkyZNkiTNmDFDaWlpev3119WiRQutW7dOAwYMUFhYmPr27Vu+NwQAgLNAMgQA54Hu3bs7fT9t2jTVqFFDmzdvVnh4eKmvefrpp3Xdddedse+QkBCFh4fL399fcXFxZ2z/yCOPKCUlRZI0ePBg9erVSwsXLlS7du0kSf3793dUlqSi5Omll17SzTffLEmqV6+eNm/erH/9618kQwCACsUwOQA4D/z000/q1auXLrjgAkVERCgxMVGStHPnzlO+plWrVhUSS7NmzRxfx8bGSpLT8LbY2Fjt3btXUlFF65dfflH//v0VHh7u2J555hnHMD8AACoKlSEAOA907dpVdevW1dSpUxUfHy+73a4mTZooLy/vlK8JCwurkFgCAgIcX1uWVeo+u90uSTp69KgkaerUqWrTpo1TP35+fhUSHwAAxUiGAMDLHThwQNu2bdPUqVPVvn17STrtggjlERgYqMLCQpf2KRVVieLj47V9+3b17t3b5f0DAHA6JEMA4OWio6NVrVo1TZkyRTVr1tTOnTv1xBNPuPQciYmJ2rFjh9avX6/atWurSpUqLltSe/To0XrooYcUGRmpTp06KTc3V6tXr9ahQ4eUmpoqqWi438GDB7Vz504VFhY6Foa48MILTzknCgCAM2HOEAB4OZvNpvfff19r1qxRkyZNNHToUL344osuPUf37t3VqVMnXXPNNapRo4bee+89l/V99913680339Rbb72lpk2bqkOHDpo+fbrq1avnaJOWlqYWLVpo5MiROnr0qFq0aKEWLVpo9erVLosDAOB7LGOMcXcQAAAAAFDZqAwBAAAA8EkkQwAAXXLJJU5LW5+8zZgxw93hAQBQIRgmBwDQb7/9pvz8/FKPxcbGqkqVKpUcEQAAFY9kCAAAAIBPYpgcAAAAAJ9EMgQAAADAJ5EMAQAAAPBJJEMAAAAAfBLJEAAAAACfRDIEAAAAwCeRDAEAAADwSf8Po1w0avw2EngAAAAASUVORK5CYII=",
          "text/plain": "<Figure size 1000x600 with 2 Axes>"
         },
         "metadata": {},
         "output_type": "display_data"
        }
       ]
      }
     },
     "7c5c7a918a764a5ba6df382d75fd3d8f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "8fc9275f6ea64e988d88962a8d00c706": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "91b8cd9f02934c6f8c7e15e41d072d57": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "9c3706e46fb541fba05080923f3c7b09": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "9fe699155858424ba05b7d1328ff06f7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "a342fa9a74d24ccab5727b2b6b0b7d6e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DropdownModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DropdownModel",
       "_options_labels": [
        "1",
        "2",
        "3",
        "4",
        "5",
        "6",
        "7",
        "8",
        "9",
        "10",
        "11",
        "12",
        "13",
        "14",
        "15",
        "16",
        "17",
        "18",
        "19",
        "20",
        "21",
        "22",
        "23",
        "24",
        "25"
       ],
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "DropdownView",
       "description": "Task:",
       "description_tooltip": null,
       "disabled": false,
       "index": 0,
       "layout": "IPY_MODEL_01390b7b874246158d7ac3ff9143ee61",
       "style": "IPY_MODEL_764305a1ce004e06ab311ef24f8c8797"
      }
     },
     "ab84a18029e04dd5a47c59eec100a684": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "bdb75ee12f1a4ecf86fb68f8b52a6863": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "cac5ae50ac8b4591a3f2cc256465a73e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "db274c67441d4cdbb4cf439462312b3b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DropdownModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DropdownModel",
       "_options_labels": [
        "1",
        "2",
        "3",
        "4",
        "5",
        "6",
        "7",
        "8",
        "9",
        "10",
        "11",
        "12",
        "13",
        "14",
        "15",
        "16",
        "17",
        "18",
        "19",
        "20",
        "21",
        "22",
        "23",
        "24",
        "25"
       ],
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "DropdownView",
       "description": "Task:",
       "description_tooltip": null,
       "disabled": false,
       "index": 0,
       "layout": "IPY_MODEL_35f85764500240d8b39afa4ce27c77c2",
       "style": "IPY_MODEL_91b8cd9f02934c6f8c7e15e41d072d57"
      }
     },
     "dc1990c5e4bc455a8d3b63997bdb0785": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "e59407c2f3144be59c18a2214489e22b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DropdownModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DropdownModel",
       "_options_labels": [
        "1",
        "2",
        "3",
        "4",
        "5",
        "6",
        "7",
        "8",
        "9",
        "10",
        "11",
        "12",
        "13",
        "14",
        "15",
        "16",
        "17",
        "18",
        "19",
        "20",
        "21",
        "22",
        "23",
        "24",
        "25"
       ],
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "DropdownView",
       "description": "Task:",
       "description_tooltip": null,
       "disabled": false,
       "index": 0,
       "layout": "IPY_MODEL_9fe699155858424ba05b7d1328ff06f7",
       "style": "IPY_MODEL_ab84a18029e04dd5a47c59eec100a684"
      }
     },
     "edbcbe2018b548158fa10b6b67361e70": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DropdownModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DropdownModel",
       "_options_labels": [
        "air_time1",
        "disp_index1",
        "gmrt_in_air1",
        "gmrt_on_paper1",
        "max_x_extension1",
        "max_y_extension1",
        "mean_acc_in_air1",
        "mean_acc_on_paper1",
        "mean_gmrt1",
        "mean_jerk_in_air1",
        "mean_jerk_on_paper1",
        "mean_speed_in_air1",
        "mean_speed_on_paper1",
        "num_of_pendown1",
        "paper_time1",
        "pressure_mean1",
        "pressure_var1",
        "total_time1",
        "air_time2",
        "disp_index2",
        "gmrt_in_air2",
        "gmrt_on_paper2",
        "max_x_extension2",
        "max_y_extension2",
        "mean_acc_in_air2",
        "mean_acc_on_paper2",
        "mean_gmrt2",
        "mean_jerk_in_air2",
        "mean_jerk_on_paper2",
        "mean_speed_in_air2",
        "mean_speed_on_paper2",
        "num_of_pendown2",
        "paper_time2",
        "pressure_mean2",
        "pressure_var2",
        "total_time2",
        "air_time3",
        "disp_index3",
        "gmrt_in_air3",
        "gmrt_on_paper3",
        "max_x_extension3",
        "max_y_extension3",
        "mean_acc_in_air3",
        "mean_acc_on_paper3",
        "mean_gmrt3",
        "mean_jerk_in_air3",
        "mean_jerk_on_paper3",
        "mean_speed_in_air3",
        "mean_speed_on_paper3",
        "num_of_pendown3",
        "paper_time3",
        "pressure_mean3",
        "pressure_var3",
        "total_time3",
        "air_time4",
        "disp_index4",
        "gmrt_in_air4",
        "gmrt_on_paper4",
        "max_x_extension4",
        "max_y_extension4",
        "mean_acc_in_air4",
        "mean_acc_on_paper4",
        "mean_gmrt4",
        "mean_jerk_in_air4",
        "mean_jerk_on_paper4",
        "mean_speed_in_air4",
        "mean_speed_on_paper4",
        "num_of_pendown4",
        "paper_time4",
        "pressure_mean4",
        "pressure_var4",
        "total_time4",
        "air_time5",
        "disp_index5",
        "gmrt_in_air5",
        "gmrt_on_paper5",
        "max_x_extension5",
        "max_y_extension5",
        "mean_acc_in_air5",
        "mean_acc_on_paper5",
        "mean_gmrt5",
        "mean_jerk_in_air5",
        "mean_jerk_on_paper5",
        "mean_speed_in_air5",
        "mean_speed_on_paper5",
        "num_of_pendown5",
        "paper_time5",
        "pressure_mean5",
        "pressure_var5",
        "total_time5",
        "air_time6",
        "disp_index6",
        "gmrt_in_air6",
        "gmrt_on_paper6",
        "max_x_extension6",
        "max_y_extension6",
        "mean_acc_in_air6",
        "mean_acc_on_paper6",
        "mean_gmrt6",
        "mean_jerk_in_air6",
        "mean_jerk_on_paper6",
        "mean_speed_in_air6",
        "mean_speed_on_paper6",
        "num_of_pendown6",
        "paper_time6",
        "pressure_mean6",
        "pressure_var6",
        "total_time6",
        "air_time7",
        "disp_index7",
        "gmrt_in_air7",
        "gmrt_on_paper7",
        "max_x_extension7",
        "max_y_extension7",
        "mean_acc_in_air7",
        "mean_acc_on_paper7",
        "mean_gmrt7",
        "mean_jerk_in_air7",
        "mean_jerk_on_paper7",
        "mean_speed_in_air7",
        "mean_speed_on_paper7",
        "num_of_pendown7",
        "paper_time7",
        "pressure_mean7",
        "pressure_var7",
        "total_time7",
        "air_time8",
        "disp_index8",
        "gmrt_in_air8",
        "gmrt_on_paper8",
        "max_x_extension8",
        "max_y_extension8",
        "mean_acc_in_air8",
        "mean_acc_on_paper8",
        "mean_gmrt8",
        "mean_jerk_in_air8",
        "mean_jerk_on_paper8",
        "mean_speed_in_air8",
        "mean_speed_on_paper8",
        "num_of_pendown8",
        "paper_time8",
        "pressure_mean8",
        "pressure_var8",
        "total_time8",
        "air_time9",
        "disp_index9",
        "gmrt_in_air9",
        "gmrt_on_paper9",
        "max_x_extension9",
        "max_y_extension9",
        "mean_acc_in_air9",
        "mean_acc_on_paper9",
        "mean_gmrt9",
        "mean_jerk_in_air9",
        "mean_jerk_on_paper9",
        "mean_speed_in_air9",
        "mean_speed_on_paper9",
        "num_of_pendown9",
        "paper_time9",
        "pressure_mean9",
        "pressure_var9",
        "total_time9",
        "air_time10",
        "disp_index10",
        "gmrt_in_air10",
        "gmrt_on_paper10",
        "max_x_extension10",
        "max_y_extension10",
        "mean_acc_in_air10",
        "mean_acc_on_paper10",
        "mean_gmrt10",
        "mean_jerk_in_air10",
        "mean_jerk_on_paper10",
        "mean_speed_in_air10",
        "mean_speed_on_paper10",
        "num_of_pendown10",
        "paper_time10",
        "pressure_mean10",
        "pressure_var10",
        "total_time10",
        "air_time11",
        "disp_index11",
        "gmrt_in_air11",
        "gmrt_on_paper11",
        "max_x_extension11",
        "max_y_extension11",
        "mean_acc_in_air11",
        "mean_acc_on_paper11",
        "mean_gmrt11",
        "mean_jerk_in_air11",
        "mean_jerk_on_paper11",
        "mean_speed_in_air11",
        "mean_speed_on_paper11",
        "num_of_pendown11",
        "paper_time11",
        "pressure_mean11",
        "pressure_var11",
        "total_time11",
        "air_time12",
        "disp_index12",
        "gmrt_in_air12",
        "gmrt_on_paper12",
        "max_x_extension12",
        "max_y_extension12",
        "mean_acc_in_air12",
        "mean_acc_on_paper12",
        "mean_gmrt12",
        "mean_jerk_in_air12",
        "mean_jerk_on_paper12",
        "mean_speed_in_air12",
        "mean_speed_on_paper12",
        "num_of_pendown12",
        "paper_time12",
        "pressure_mean12",
        "pressure_var12",
        "total_time12",
        "air_time13",
        "disp_index13",
        "gmrt_in_air13",
        "gmrt_on_paper13",
        "max_x_extension13",
        "max_y_extension13",
        "mean_acc_in_air13",
        "mean_acc_on_paper13",
        "mean_gmrt13",
        "mean_jerk_in_air13",
        "mean_jerk_on_paper13",
        "mean_speed_in_air13",
        "mean_speed_on_paper13",
        "num_of_pendown13",
        "paper_time13",
        "pressure_mean13",
        "pressure_var13",
        "total_time13",
        "air_time14",
        "disp_index14",
        "gmrt_in_air14",
        "gmrt_on_paper14",
        "max_x_extension14",
        "max_y_extension14",
        "mean_acc_in_air14",
        "mean_acc_on_paper14",
        "mean_gmrt14",
        "mean_jerk_in_air14",
        "mean_jerk_on_paper14",
        "mean_speed_in_air14",
        "mean_speed_on_paper14",
        "num_of_pendown14",
        "paper_time14",
        "pressure_mean14",
        "pressure_var14",
        "total_time14",
        "air_time15",
        "disp_index15",
        "gmrt_in_air15",
        "gmrt_on_paper15",
        "max_x_extension15",
        "max_y_extension15",
        "mean_acc_in_air15",
        "mean_acc_on_paper15",
        "mean_gmrt15",
        "mean_jerk_in_air15",
        "mean_jerk_on_paper15",
        "mean_speed_in_air15",
        "mean_speed_on_paper15",
        "num_of_pendown15",
        "paper_time15",
        "pressure_mean15",
        "pressure_var15",
        "total_time15",
        "air_time16",
        "disp_index16",
        "gmrt_in_air16",
        "gmrt_on_paper16",
        "max_x_extension16",
        "max_y_extension16",
        "mean_acc_in_air16",
        "mean_acc_on_paper16",
        "mean_gmrt16",
        "mean_jerk_in_air16",
        "mean_jerk_on_paper16",
        "mean_speed_in_air16",
        "mean_speed_on_paper16",
        "num_of_pendown16",
        "paper_time16",
        "pressure_mean16",
        "pressure_var16",
        "total_time16",
        "air_time17",
        "disp_index17",
        "gmrt_in_air17",
        "gmrt_on_paper17",
        "max_x_extension17",
        "max_y_extension17",
        "mean_acc_in_air17",
        "mean_acc_on_paper17",
        "mean_gmrt17",
        "mean_jerk_in_air17",
        "mean_jerk_on_paper17",
        "mean_speed_in_air17",
        "mean_speed_on_paper17",
        "num_of_pendown17",
        "paper_time17",
        "pressure_mean17",
        "pressure_var17",
        "total_time17",
        "air_time18",
        "disp_index18",
        "gmrt_in_air18",
        "gmrt_on_paper18",
        "max_x_extension18",
        "max_y_extension18",
        "mean_acc_in_air18",
        "mean_acc_on_paper18",
        "mean_gmrt18",
        "mean_jerk_in_air18",
        "mean_jerk_on_paper18",
        "mean_speed_in_air18",
        "mean_speed_on_paper18",
        "num_of_pendown18",
        "paper_time18",
        "pressure_mean18",
        "pressure_var18",
        "total_time18",
        "air_time19",
        "disp_index19",
        "gmrt_in_air19",
        "gmrt_on_paper19",
        "max_x_extension19",
        "max_y_extension19",
        "mean_acc_in_air19",
        "mean_acc_on_paper19",
        "mean_gmrt19",
        "mean_jerk_in_air19",
        "mean_jerk_on_paper19",
        "mean_speed_in_air19",
        "mean_speed_on_paper19",
        "num_of_pendown19",
        "paper_time19",
        "pressure_mean19",
        "pressure_var19",
        "total_time19",
        "air_time20",
        "disp_index20",
        "gmrt_in_air20",
        "gmrt_on_paper20",
        "max_x_extension20",
        "max_y_extension20",
        "mean_acc_in_air20",
        "mean_acc_on_paper20",
        "mean_gmrt20",
        "mean_jerk_in_air20",
        "mean_jerk_on_paper20",
        "mean_speed_in_air20",
        "mean_speed_on_paper20",
        "num_of_pendown20",
        "paper_time20",
        "pressure_mean20",
        "pressure_var20",
        "total_time20",
        "air_time21",
        "disp_index21",
        "gmrt_in_air21",
        "gmrt_on_paper21",
        "max_x_extension21",
        "max_y_extension21",
        "mean_acc_in_air21",
        "mean_acc_on_paper21",
        "mean_gmrt21",
        "mean_jerk_in_air21",
        "mean_jerk_on_paper21",
        "mean_speed_in_air21",
        "mean_speed_on_paper21",
        "num_of_pendown21",
        "paper_time21",
        "pressure_mean21",
        "pressure_var21",
        "total_time21",
        "air_time22",
        "disp_index22",
        "gmrt_in_air22",
        "gmrt_on_paper22",
        "max_x_extension22",
        "max_y_extension22",
        "mean_acc_in_air22",
        "mean_acc_on_paper22",
        "mean_gmrt22",
        "mean_jerk_in_air22",
        "mean_jerk_on_paper22",
        "mean_speed_in_air22",
        "mean_speed_on_paper22",
        "num_of_pendown22",
        "paper_time22",
        "pressure_mean22",
        "pressure_var22",
        "total_time22",
        "air_time23",
        "disp_index23",
        "gmrt_in_air23",
        "gmrt_on_paper23",
        "max_x_extension23",
        "max_y_extension23",
        "mean_acc_in_air23",
        "mean_acc_on_paper23",
        "mean_gmrt23",
        "mean_jerk_in_air23",
        "mean_jerk_on_paper23",
        "mean_speed_in_air23",
        "mean_speed_on_paper23",
        "num_of_pendown23",
        "paper_time23",
        "pressure_mean23",
        "pressure_var23",
        "total_time23",
        "air_time24",
        "disp_index24",
        "gmrt_in_air24",
        "gmrt_on_paper24",
        "max_x_extension24",
        "max_y_extension24",
        "mean_acc_in_air24",
        "mean_acc_on_paper24",
        "mean_gmrt24",
        "mean_jerk_in_air24",
        "mean_jerk_on_paper24",
        "mean_speed_in_air24",
        "mean_speed_on_paper24",
        "num_of_pendown24",
        "paper_time24",
        "pressure_mean24",
        "pressure_var24",
        "total_time24",
        "air_time25",
        "disp_index25",
        "gmrt_in_air25",
        "gmrt_on_paper25",
        "max_x_extension25",
        "max_y_extension25",
        "mean_acc_in_air25",
        "mean_acc_on_paper25",
        "mean_gmrt25",
        "mean_jerk_in_air25",
        "mean_jerk_on_paper25",
        "mean_speed_in_air25",
        "mean_speed_on_paper25",
        "num_of_pendown25",
        "paper_time25",
        "pressure_mean25",
        "pressure_var25",
        "total_time25",
        "class"
       ],
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "DropdownView",
       "description": "X-axis:",
       "description_tooltip": null,
       "disabled": false,
       "index": 0,
       "layout": "IPY_MODEL_bdb75ee12f1a4ecf86fb68f8b52a6863",
       "style": "IPY_MODEL_dc1990c5e4bc455a8d3b63997bdb0785"
      }
     },
     "edcf84914a2c4246b52f9800f0dc1e09": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "ede3c8ad54784f5a9d870c62b4ef8a02": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DropdownModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DropdownModel",
       "_options_labels": [
        "1",
        "2",
        "3",
        "4",
        "5",
        "6",
        "7",
        "8",
        "9",
        "10",
        "11",
        "12",
        "13",
        "14",
        "15",
        "16",
        "17",
        "18",
        "19",
        "20",
        "21",
        "22",
        "23",
        "24",
        "25"
       ],
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "DropdownView",
       "description": "Task:",
       "description_tooltip": null,
       "disabled": false,
       "index": 0,
       "layout": "IPY_MODEL_8fc9275f6ea64e988d88962a8d00c706",
       "style": "IPY_MODEL_edcf84914a2c4246b52f9800f0dc1e09"
      }
     },
     "f509710592774f95a152bd35bb12dcec": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DropdownModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DropdownModel",
       "_options_labels": [
        "air_time1",
        "disp_index1",
        "gmrt_in_air1",
        "gmrt_on_paper1",
        "max_x_extension1",
        "max_y_extension1",
        "mean_acc_in_air1",
        "mean_acc_on_paper1",
        "mean_gmrt1",
        "mean_jerk_in_air1",
        "mean_jerk_on_paper1",
        "mean_speed_in_air1",
        "mean_speed_on_paper1",
        "num_of_pendown1",
        "paper_time1",
        "pressure_mean1",
        "pressure_var1",
        "total_time1",
        "air_time2",
        "disp_index2",
        "gmrt_in_air2",
        "gmrt_on_paper2",
        "max_x_extension2",
        "max_y_extension2",
        "mean_acc_in_air2",
        "mean_acc_on_paper2",
        "mean_gmrt2",
        "mean_jerk_in_air2",
        "mean_jerk_on_paper2",
        "mean_speed_in_air2",
        "mean_speed_on_paper2",
        "num_of_pendown2",
        "paper_time2",
        "pressure_mean2",
        "pressure_var2",
        "total_time2",
        "air_time3",
        "disp_index3",
        "gmrt_in_air3",
        "gmrt_on_paper3",
        "max_x_extension3",
        "max_y_extension3",
        "mean_acc_in_air3",
        "mean_acc_on_paper3",
        "mean_gmrt3",
        "mean_jerk_in_air3",
        "mean_jerk_on_paper3",
        "mean_speed_in_air3",
        "mean_speed_on_paper3",
        "num_of_pendown3",
        "paper_time3",
        "pressure_mean3",
        "pressure_var3",
        "total_time3",
        "air_time4",
        "disp_index4",
        "gmrt_in_air4",
        "gmrt_on_paper4",
        "max_x_extension4",
        "max_y_extension4",
        "mean_acc_in_air4",
        "mean_acc_on_paper4",
        "mean_gmrt4",
        "mean_jerk_in_air4",
        "mean_jerk_on_paper4",
        "mean_speed_in_air4",
        "mean_speed_on_paper4",
        "num_of_pendown4",
        "paper_time4",
        "pressure_mean4",
        "pressure_var4",
        "total_time4",
        "air_time5",
        "disp_index5",
        "gmrt_in_air5",
        "gmrt_on_paper5",
        "max_x_extension5",
        "max_y_extension5",
        "mean_acc_in_air5",
        "mean_acc_on_paper5",
        "mean_gmrt5",
        "mean_jerk_in_air5",
        "mean_jerk_on_paper5",
        "mean_speed_in_air5",
        "mean_speed_on_paper5",
        "num_of_pendown5",
        "paper_time5",
        "pressure_mean5",
        "pressure_var5",
        "total_time5",
        "air_time6",
        "disp_index6",
        "gmrt_in_air6",
        "gmrt_on_paper6",
        "max_x_extension6",
        "max_y_extension6",
        "mean_acc_in_air6",
        "mean_acc_on_paper6",
        "mean_gmrt6",
        "mean_jerk_in_air6",
        "mean_jerk_on_paper6",
        "mean_speed_in_air6",
        "mean_speed_on_paper6",
        "num_of_pendown6",
        "paper_time6",
        "pressure_mean6",
        "pressure_var6",
        "total_time6",
        "air_time7",
        "disp_index7",
        "gmrt_in_air7",
        "gmrt_on_paper7",
        "max_x_extension7",
        "max_y_extension7",
        "mean_acc_in_air7",
        "mean_acc_on_paper7",
        "mean_gmrt7",
        "mean_jerk_in_air7",
        "mean_jerk_on_paper7",
        "mean_speed_in_air7",
        "mean_speed_on_paper7",
        "num_of_pendown7",
        "paper_time7",
        "pressure_mean7",
        "pressure_var7",
        "total_time7",
        "air_time8",
        "disp_index8",
        "gmrt_in_air8",
        "gmrt_on_paper8",
        "max_x_extension8",
        "max_y_extension8",
        "mean_acc_in_air8",
        "mean_acc_on_paper8",
        "mean_gmrt8",
        "mean_jerk_in_air8",
        "mean_jerk_on_paper8",
        "mean_speed_in_air8",
        "mean_speed_on_paper8",
        "num_of_pendown8",
        "paper_time8",
        "pressure_mean8",
        "pressure_var8",
        "total_time8",
        "air_time9",
        "disp_index9",
        "gmrt_in_air9",
        "gmrt_on_paper9",
        "max_x_extension9",
        "max_y_extension9",
        "mean_acc_in_air9",
        "mean_acc_on_paper9",
        "mean_gmrt9",
        "mean_jerk_in_air9",
        "mean_jerk_on_paper9",
        "mean_speed_in_air9",
        "mean_speed_on_paper9",
        "num_of_pendown9",
        "paper_time9",
        "pressure_mean9",
        "pressure_var9",
        "total_time9",
        "air_time10",
        "disp_index10",
        "gmrt_in_air10",
        "gmrt_on_paper10",
        "max_x_extension10",
        "max_y_extension10",
        "mean_acc_in_air10",
        "mean_acc_on_paper10",
        "mean_gmrt10",
        "mean_jerk_in_air10",
        "mean_jerk_on_paper10",
        "mean_speed_in_air10",
        "mean_speed_on_paper10",
        "num_of_pendown10",
        "paper_time10",
        "pressure_mean10",
        "pressure_var10",
        "total_time10",
        "air_time11",
        "disp_index11",
        "gmrt_in_air11",
        "gmrt_on_paper11",
        "max_x_extension11",
        "max_y_extension11",
        "mean_acc_in_air11",
        "mean_acc_on_paper11",
        "mean_gmrt11",
        "mean_jerk_in_air11",
        "mean_jerk_on_paper11",
        "mean_speed_in_air11",
        "mean_speed_on_paper11",
        "num_of_pendown11",
        "paper_time11",
        "pressure_mean11",
        "pressure_var11",
        "total_time11",
        "air_time12",
        "disp_index12",
        "gmrt_in_air12",
        "gmrt_on_paper12",
        "max_x_extension12",
        "max_y_extension12",
        "mean_acc_in_air12",
        "mean_acc_on_paper12",
        "mean_gmrt12",
        "mean_jerk_in_air12",
        "mean_jerk_on_paper12",
        "mean_speed_in_air12",
        "mean_speed_on_paper12",
        "num_of_pendown12",
        "paper_time12",
        "pressure_mean12",
        "pressure_var12",
        "total_time12",
        "air_time13",
        "disp_index13",
        "gmrt_in_air13",
        "gmrt_on_paper13",
        "max_x_extension13",
        "max_y_extension13",
        "mean_acc_in_air13",
        "mean_acc_on_paper13",
        "mean_gmrt13",
        "mean_jerk_in_air13",
        "mean_jerk_on_paper13",
        "mean_speed_in_air13",
        "mean_speed_on_paper13",
        "num_of_pendown13",
        "paper_time13",
        "pressure_mean13",
        "pressure_var13",
        "total_time13",
        "air_time14",
        "disp_index14",
        "gmrt_in_air14",
        "gmrt_on_paper14",
        "max_x_extension14",
        "max_y_extension14",
        "mean_acc_in_air14",
        "mean_acc_on_paper14",
        "mean_gmrt14",
        "mean_jerk_in_air14",
        "mean_jerk_on_paper14",
        "mean_speed_in_air14",
        "mean_speed_on_paper14",
        "num_of_pendown14",
        "paper_time14",
        "pressure_mean14",
        "pressure_var14",
        "total_time14",
        "air_time15",
        "disp_index15",
        "gmrt_in_air15",
        "gmrt_on_paper15",
        "max_x_extension15",
        "max_y_extension15",
        "mean_acc_in_air15",
        "mean_acc_on_paper15",
        "mean_gmrt15",
        "mean_jerk_in_air15",
        "mean_jerk_on_paper15",
        "mean_speed_in_air15",
        "mean_speed_on_paper15",
        "num_of_pendown15",
        "paper_time15",
        "pressure_mean15",
        "pressure_var15",
        "total_time15",
        "air_time16",
        "disp_index16",
        "gmrt_in_air16",
        "gmrt_on_paper16",
        "max_x_extension16",
        "max_y_extension16",
        "mean_acc_in_air16",
        "mean_acc_on_paper16",
        "mean_gmrt16",
        "mean_jerk_in_air16",
        "mean_jerk_on_paper16",
        "mean_speed_in_air16",
        "mean_speed_on_paper16",
        "num_of_pendown16",
        "paper_time16",
        "pressure_mean16",
        "pressure_var16",
        "total_time16",
        "air_time17",
        "disp_index17",
        "gmrt_in_air17",
        "gmrt_on_paper17",
        "max_x_extension17",
        "max_y_extension17",
        "mean_acc_in_air17",
        "mean_acc_on_paper17",
        "mean_gmrt17",
        "mean_jerk_in_air17",
        "mean_jerk_on_paper17",
        "mean_speed_in_air17",
        "mean_speed_on_paper17",
        "num_of_pendown17",
        "paper_time17",
        "pressure_mean17",
        "pressure_var17",
        "total_time17",
        "air_time18",
        "disp_index18",
        "gmrt_in_air18",
        "gmrt_on_paper18",
        "max_x_extension18",
        "max_y_extension18",
        "mean_acc_in_air18",
        "mean_acc_on_paper18",
        "mean_gmrt18",
        "mean_jerk_in_air18",
        "mean_jerk_on_paper18",
        "mean_speed_in_air18",
        "mean_speed_on_paper18",
        "num_of_pendown18",
        "paper_time18",
        "pressure_mean18",
        "pressure_var18",
        "total_time18",
        "air_time19",
        "disp_index19",
        "gmrt_in_air19",
        "gmrt_on_paper19",
        "max_x_extension19",
        "max_y_extension19",
        "mean_acc_in_air19",
        "mean_acc_on_paper19",
        "mean_gmrt19",
        "mean_jerk_in_air19",
        "mean_jerk_on_paper19",
        "mean_speed_in_air19",
        "mean_speed_on_paper19",
        "num_of_pendown19",
        "paper_time19",
        "pressure_mean19",
        "pressure_var19",
        "total_time19",
        "air_time20",
        "disp_index20",
        "gmrt_in_air20",
        "gmrt_on_paper20",
        "max_x_extension20",
        "max_y_extension20",
        "mean_acc_in_air20",
        "mean_acc_on_paper20",
        "mean_gmrt20",
        "mean_jerk_in_air20",
        "mean_jerk_on_paper20",
        "mean_speed_in_air20",
        "mean_speed_on_paper20",
        "num_of_pendown20",
        "paper_time20",
        "pressure_mean20",
        "pressure_var20",
        "total_time20",
        "air_time21",
        "disp_index21",
        "gmrt_in_air21",
        "gmrt_on_paper21",
        "max_x_extension21",
        "max_y_extension21",
        "mean_acc_in_air21",
        "mean_acc_on_paper21",
        "mean_gmrt21",
        "mean_jerk_in_air21",
        "mean_jerk_on_paper21",
        "mean_speed_in_air21",
        "mean_speed_on_paper21",
        "num_of_pendown21",
        "paper_time21",
        "pressure_mean21",
        "pressure_var21",
        "total_time21",
        "air_time22",
        "disp_index22",
        "gmrt_in_air22",
        "gmrt_on_paper22",
        "max_x_extension22",
        "max_y_extension22",
        "mean_acc_in_air22",
        "mean_acc_on_paper22",
        "mean_gmrt22",
        "mean_jerk_in_air22",
        "mean_jerk_on_paper22",
        "mean_speed_in_air22",
        "mean_speed_on_paper22",
        "num_of_pendown22",
        "paper_time22",
        "pressure_mean22",
        "pressure_var22",
        "total_time22",
        "air_time23",
        "disp_index23",
        "gmrt_in_air23",
        "gmrt_on_paper23",
        "max_x_extension23",
        "max_y_extension23",
        "mean_acc_in_air23",
        "mean_acc_on_paper23",
        "mean_gmrt23",
        "mean_jerk_in_air23",
        "mean_jerk_on_paper23",
        "mean_speed_in_air23",
        "mean_speed_on_paper23",
        "num_of_pendown23",
        "paper_time23",
        "pressure_mean23",
        "pressure_var23",
        "total_time23",
        "air_time24",
        "disp_index24",
        "gmrt_in_air24",
        "gmrt_on_paper24",
        "max_x_extension24",
        "max_y_extension24",
        "mean_acc_in_air24",
        "mean_acc_on_paper24",
        "mean_gmrt24",
        "mean_jerk_in_air24",
        "mean_jerk_on_paper24",
        "mean_speed_in_air24",
        "mean_speed_on_paper24",
        "num_of_pendown24",
        "paper_time24",
        "pressure_mean24",
        "pressure_var24",
        "total_time24",
        "air_time25",
        "disp_index25",
        "gmrt_in_air25",
        "gmrt_on_paper25",
        "max_x_extension25",
        "max_y_extension25",
        "mean_acc_in_air25",
        "mean_acc_on_paper25",
        "mean_gmrt25",
        "mean_jerk_in_air25",
        "mean_jerk_on_paper25",
        "mean_speed_in_air25",
        "mean_speed_on_paper25",
        "num_of_pendown25",
        "paper_time25",
        "pressure_mean25",
        "pressure_var25",
        "total_time25",
        "class"
       ],
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "DropdownView",
       "description": "Y-axis:",
       "description_tooltip": null,
       "disabled": false,
       "index": 0,
       "layout": "IPY_MODEL_7c5c7a918a764a5ba6df382d75fd3d8f",
       "style": "IPY_MODEL_6d71ec7cdf9e4d8b8647aaec591650ed"
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}