Models_Sequences_data.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "[Colab link ](https://drive.google.com/file/d/1HmvrkYaFIpb9l7d9RgZpcffEBTkbej60/view?usp=sharing)\n"
      ],
      "metadata": {
        "id": "0m3FP0o2mWUE"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xo1CVjZb-oDD"
      },
      "source": [
        "## Importing"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "aGGXquTD-qxk"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "from sklearn.preprocessing import PolynomialFeatures\n",
        "from sklearn import decomposition\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn import decomposition\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "import os\n",
        "\n",
        "from tqdm.auto import tqdm\n",
        "\n",
        "import shutil\n",
        "\n",
        "from itertools import groupby\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from statistics import mean\n",
        "\n",
        "import scipy.stats as stats\n",
        "from scipy.stats import skew\n",
        "from scipy.stats import kurtosis\n",
        "from scipy.fftpack import fft\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "\n",
        "import pickle\n",
        "\n",
        "from keras.models import Sequential, Model\n",
        "from keras.layers import LSTM, Dense, Input, Flatten, Conv1D, MaxPooling1D, Concatenate, Dropout\n",
        "from keras.optimizers import Adam\n",
        "from keras.callbacks import EarlyStopping\n",
        "from keras.utils.vis_utils import plot_model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EflF3KAFXrai",
        "outputId": "ed194b6d-d2b9-4533-f52b-da63c74500da"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yoPlShO_1mD5"
      },
      "source": [
        "## Parameters"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "NJh8EXU71npV"
      },
      "outputs": [],
      "source": [
        "# path of original data folder\n",
        "data_path = \"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Data\"\n",
        "# path of csv files, each file is a 2mins walk for a subject\n",
        "csvD_path = \"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Data/csv_files\"\n",
        "# In this file, the Null values are replaved with 0 in the HY scale - Null values were given to healthy control\n",
        "demographics = \"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Data/Processed_data/demographics_HYprocessed.csv\" \n",
        "dem_df = pd.read_csv(demographics)\n",
        "# parequet folder path\n",
        "parquet_path = \"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Data/parquet_files/\""
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lxD2898AlNMI"
      },
      "source": [
        "## Data Utils"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VLlmkpXwQPt0"
      },
      "source": [
        "### Data Class"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1gA_nqXg_Q0V"
      },
      "source": [
        "Note: \n",
        "Gait cycle = stride cycle\n",
        "\n",
        "\n",
        "- Usually, Vertical Ground React Force VGRF values less than 20N are generated from noise\n",
        "- For gait cycle segmentation, to eliminate the effect of gait initiation and termenation, the first and last 20 seconds of VGRF data was discarded.\n",
        "- Note: Since we are deleting the first and last 20 seconds, some patients that performed less than 2 mins have almost no gait cycles left. For example, the paient with Id `JuCo02` and `JuCo03` walked for almost 40.40 seconds. Thus, after deleting the first and last 20 sec, he has only half gait cycle left. \n",
        "\n",
        "*   stride time variability and swing time variability are important parameters in distinguishing a PD patient’s gait from normal gait.\n",
        "*   As PD patients tend to put less pressure during placing the heel strike and toe off than control subjects, maximum VGRF at heel strike and toe off for each gait cycle was computed. The mean and standard deviation of the VGRF overall gait cycles were taken as features for classification.\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "3uHx0SlZl2aJ"
      },
      "outputs": [],
      "source": [
        "class Data:\n",
        "  def __init__(self, prepare_or_get, data_folder, demographics_file, parquet_path):\n",
        "    '''\n",
        "    :param prepare_or_get:  1: prepare data , 0: load preloaded datas (npy)\n",
        "    :param data_folder: path for a folder containing csv files, file for each 2 mins walk (for indivdiual subject)\n",
        "    :param demographics_file: path for csv demographics file\n",
        "    :save_path: path where to save/read parquet files after preparing them\n",
        "    :\n",
        "    '''\n",
        "\n",
        "    self.data_path = data_folder\n",
        "    self.parquet_path= parquet_path\n",
        "    self.dem_path= demographics_file\n",
        "    self.scale=\"HoehnYahr\"            # fixed for now\n",
        "    self.seconds_to_discard = 20      # fixed for now\n",
        "    self.rows_to_discard = 100 * self.seconds_to_discard\n",
        "    \n",
        "    # data in arrays\n",
        "    self.signals_data = []\n",
        "    self.cycles_data=[]\n",
        "    self.statical_data=[]\n",
        "  \n",
        "\n",
        "    if (prepare_or_get==1):\n",
        "      self.prepare_data()\n",
        "    elif (prepare_or_get==0):\n",
        "      self.get_data()\n",
        "  \n",
        "  def get_data(self):\n",
        "    self.signals_data_df= pd.read_parquet(self.parquet_path+'signals.gzip')\n",
        "    self.cycles_data_df=pd.read_parquet(self.parquet_path+'cyclels.gzip')\n",
        "    self.statical_data_df=pd.read_parquet(self.parquet_path+'statical.gzip')\n",
        "    self.left_stances_df=pd.read_parquet(self.parquet_path+'left_stances.gzip')\n",
        "    self.right_stances_df=pd.read_parquet(self.parquet_path+'right_stances.gzip')\n",
        "\n",
        "  def prepare_data(self):\n",
        "    # loop over files, each file is a 2 min walk for a single individual\n",
        "    # files are expected to be in csv format\n",
        "    self.dem_df = pd.read_csv(self.dem_path)\n",
        "    for name in tqdm((os.listdir(self.data_path))):\n",
        "       \n",
        "      # id of subject\n",
        "      id = name.split('_')[0]\n",
        "\n",
        "      # disease level according to 'self.scale'\n",
        "      level = self.get_pd_level( id)\n",
        "\n",
        "      # parkinon's patient OR healthy control ?\n",
        "      if 'Co' in name:  \n",
        "        parkinson = 0 # Healthy control\n",
        "      elif 'Pt' in name:\n",
        "        parkinson = 1 # Parkinson's Patient\n",
        "\n",
        "      # read and filter one file\n",
        "      one_walk=self.read_filter(name)\n",
        "\n",
        "      # sum of values from all the left sensors\n",
        "      one_walk['Total_Force_Left'] = one_walk[list(one_walk.columns[0:8])].sum(axis=1)\n",
        "     \n",
        "      # sum of values from all the right sensors\n",
        "      one_walk['Total_Force_Right'] = one_walk[list(one_walk.columns[8:16])].sum(axis=1)\n",
        "      \n",
        "      # convert to lists\n",
        "      one_walk_lists=[]*18\n",
        "      for column in one_walk.columns.values.tolist():\n",
        "        one_walk_lists.append(one_walk[column].tolist()) \n",
        "\n",
        "      ## Left foot related ##\n",
        "      l_stances, l_strides_time , l_swings_time, l_stances_time, l_indices = self.segment_signal(one_walk_lists[16])\n",
        "      # Find maximum VGRF at heel strike for each gait cycle \n",
        "      l_peaks_heel = self.get_max_per_gait_cycle(one_walk_lists[0],l_indices) # sensor 0 = left heal sensor\n",
        "      # Find maximum VGRF at toe off for each gait cycle\n",
        "      l_peaks_toe = self.get_max_per_gait_cycle(one_walk_lists[7],l_indices) # sensor 7 = left toe sensor\n",
        "\n",
        "      ## right foot related ##\n",
        "      r_stances, r_strides_time , r_swings_time, r_stances_time, r_indices = self.segment_signal(one_walk_lists[17])\n",
        "      # Find maximum VGRF at heel strike for each gait cycle \n",
        "      r_peaks_heel = self.get_max_per_gait_cycle(one_walk_lists[8],l_indices) # sensor 8 = right heal sensor\n",
        "      # Find maximum VGRF at toe off for each gait cycle\n",
        "      r_peaks_toe = self.get_max_per_gait_cycle(one_walk_lists[15],l_indices) # sensor 15 = right toe sensor\n",
        "\n",
        "\n",
        "      ## Saving data ##\n",
        "\n",
        "      # Raw Signals\n",
        "      self.signals_data.append([id,level,parkinson] +one_walk_lists)\n",
        "\n",
        "      # Statical\n",
        "      self.statical_data.append( [ id, level, parkinson \n",
        "                                , mean (l_swings_time) , np.std(l_swings_time), (mean (l_swings_time) / np.std(l_swings_time) *100)\n",
        "                                , mean (l_strides_time) , np.std(l_strides_time) , (mean (l_strides_time) / np.std(l_strides_time)*100)\n",
        "                                , kurtosis(l_strides_time), skew(l_strides_time)\n",
        "                                , mean (l_stances_time) , np.std(l_stances_time) , ( mean (l_stances_time) / np.std(l_stances_time) *100) \n",
        "\n",
        "                                , mean (r_swings_time) , np.std(r_swings_time), (mean (r_swings_time) / np.std(r_swings_time) *100)\n",
        "                                , mean (r_strides_time) , np.std(r_strides_time) , (mean (r_strides_time) / np.std(r_strides_time)*100)\n",
        "                                , kurtosis(r_strides_time), skew(r_strides_time)\n",
        "                                , mean (r_stances_time) , np.std(r_stances_time) , ( mean (r_stances_time) / np.std(r_stances_time) *100)\n",
        "                                \n",
        "                                , mean (l_peaks_heel) , np.std(l_peaks_heel) ,mean (l_peaks_toe) , np.std(l_peaks_toe)\n",
        "                                , mean (r_peaks_heel) , np.std(r_peaks_heel) ,mean (r_peaks_toe) , np.std(r_peaks_toe)\n",
        "                                  ] )\n",
        "      \n",
        "      \n",
        "      # Segmented Signals\n",
        "      \n",
        "      # NEED TO BE MOVED TO SEPERATE FUNCTION\n",
        "      # only for segmented data, for a trial we need each stance phase with the following \n",
        "      # swing phases's time, so here we are adjusting the phases to be equal\n",
        "\n",
        "      dif_l =  len(l_stances)-len(l_swings_time)\n",
        "      if (dif_l > 0):\n",
        "        l_stances=l_stances[:-dif_l]\n",
        "      elif(dif_l < 0):\n",
        "        l_swings_time= l_swings_time[:dif_l]\n",
        "\n",
        "      \n",
        "      dif_r =  len(r_stances)-len(r_swings_time)\n",
        "      if (dif_r>0):\n",
        "        r_stances=r_stances[:-dif_r]\n",
        "      elif(dif_r < 0):\n",
        "        r_swings_time = r_swings_time[:dif_r]\n",
        "       \n",
        "      self.cycles_data.append( [ id, level, parkinson ] \n",
        "                              + [l_stances] + [l_strides_time] + [l_swings_time] + [l_stances_time]\n",
        "                              + [r_stances] + [r_strides_time] + [r_swings_time] + [r_stances_time]\n",
        "                              + [l_peaks_heel] + [l_peaks_toe]\n",
        "                              + [r_peaks_heel] + [r_peaks_toe]\n",
        "                              )\n",
        "      \n",
        "\n",
        "    self.statical_data_to_df()\n",
        "    self.cycles_data_to_df()\n",
        "    self.signal_data_to_df()\n",
        "    ## Get individual stances data frames\n",
        "    self.ind_stances_to_df()\n",
        "    ## writing files ##\n",
        "    self.save_files()\n",
        "\n",
        "  def get_pd_level(self, subject_id):\n",
        "    level = self.dem_df[self.scale][dem_df['ID'] == subject_id ].values[0]\n",
        "    return level\n",
        "\n",
        "  def read_filter(self,name):\n",
        "      # Reading each file, skipping 3 columns (time,total right forces & total left forces)\n",
        "      # For gait cycle segmentation, to eliminate the effect of gait initiation and termenation,\n",
        "      # the first and last N=20 seconds of VGRF data was discarded.\n",
        "\n",
        "      one_walk = pd.read_csv(self.data_path + '/' + name,\n",
        "                             skiprows=list(range(1,self.rows_to_discard+1)) ,\n",
        "                             skipfooter=self.rows_to_discard,\n",
        "                             usecols = np.arange(1,17),\n",
        "                             engine = 'python')  \n",
        "      \n",
        "      # Usually, Vertical Ground React Force VGRF values less than 20N are generated from noise\n",
        "      one_walk.where(one_walk > 20, 0, inplace=True)\n",
        "\n",
        "      return one_walk\n",
        "\n",
        "  def segment_signal(self, signal):\n",
        "      # swing - stance phases repeatly\n",
        "      phases = [list(g) for k, g in groupby((signal), lambda x:x>0)]\n",
        "      \n",
        "      # Deleting the first and last phase, since the first and last 20 seconds were deleted,\n",
        "      # there's no garanty that the first and last phase are complete\n",
        "      # Hence, they are being deleted here just for safety\n",
        "\n",
        "      phases = phases[1:-1]\n",
        "\n",
        "      # filtering phases that is shorter than 20 becuase must propably it's noise\n",
        "      phases = [l for l in phases if len(l)>20]\n",
        "\n",
        "      # indices of the gait cycles\n",
        "      indices=[0]\n",
        "      for i in range(1, len(phases),2):\n",
        "        indices.append(indices[-1]+len(phases[i]) + len(phases[i-1]))\n",
        "\n",
        "      # stances phases\n",
        "      stances = [l for l in phases if any(l)]\n",
        "      \n",
        "      # strides times \n",
        "      strides_time =  [((len(phases[i]) + len(phases[i-1]) -1)*0.01 ) for i in range(1,len(phases),2)]\n",
        "\n",
        "      # swings time\n",
        "      swings_time = [(len(l)-1)*0.01 for l in phases if not all (l)]\n",
        "\n",
        "      # stance time \n",
        "      stances_time = [(len(l)-1)*0.01 for l in stances]\n",
        "\n",
        "      return stances, strides_time , swings_time, stances_time, indices\n",
        "\n",
        "  def get_max_per_gait_cycle(self, signal,indices):\n",
        "    gait_cycles = self.get_gait_cycles(signal, indices)\n",
        "    peaks = [max(cycle) for cycle in gait_cycles]\n",
        "    return peaks\n",
        "\n",
        "  def get_gait_cycles(self, signal, indices):\n",
        "    # get gait cycles\n",
        "    gait_cycles= [signal[indices[i]:indices[i+1]] for i in range(len(indices)-1)]\n",
        "    return gait_cycles\n",
        "\n",
        "  def signal_data_to_df(self):\n",
        "    self.signals_data_df = pd.DataFrame(self.signals_data, columns =[\"ID\", \"level\", \"y\", \"L1\", \"L2\",\"L3\", \"L4\",\"L5\",\"L6\",\"L7\",\"L8\",\n",
        "                                       \"R1\", \"R2\",\"R3\", \"R4\",\"R5\",\"R6\",\"R7\",\"R8\",\"Total_Force_Right\",\"Total_Force_Left\" ])\n",
        "     \n",
        "    \n",
        "  def cycles_data_to_df(self):\n",
        "    self.cycles_data_df = pd.DataFrame(self.cycles_data, columns =[\"ID\", \"level\", \"y\", \n",
        "                                                                   \"l_stances\", \"l_strides_time\", \"l_swings_time\" , \"l_stances_time\"\n",
        "                                                                  , \"r_stances\" , \"r_strides_time\" , \"r_swings_time\" , \"r_stances_time\"\n",
        "                                                                  , \"l_peaks_heel\" , \"l_peaks_toe\"\n",
        "                                                                  , \"r_peaks_heel\" , \"r_peaks_toe\" ])\n",
        "    \n",
        "\n",
        "  def statical_data_to_df(self):\n",
        "    self.statical_data_df = pd.DataFrame(self.statical_data, columns =[\"ID\", \"level\", \"y\"\n",
        "                                                                      , \"mean_left_swings_time\" , \"std_left_swings_time\", \"cv_left_swings_time\"\n",
        "                                                                      , \"mean_left_stride_time\" , \"std_left_stride_time\" ,\"cv_left_stride_time\"\n",
        "                                                                      , \"kurtosis_left_strides_time\", \"skew_left_strides_time\"\n",
        "                                                                      , \"mean_left_stance_time\" , \"std_left_stance_time\", \"cv_left_stance_time\"\n",
        "                                                                      , \"mean_right_swings_time\" , \"std_right_swings_time\", \"cv_right_swings_time\"\n",
        "                                                                      , \"mean_right_stride_time\" , \"std_right_stride_time\", \"cv_right_stride_time\"\n",
        "                                                                      , \"kurtosis_right_strides_time\", \"skew_right_strides_time\"\n",
        "                                                                      , \"mean_right_stance_time\" , \"std_right_stance_time\", \"cv_right_stance_time\"\n",
        "                                                                      , \"mean_left_peaks_heel\" , \"std_left_peaks_heel\"\n",
        "                                                                      , \"mean_left_peaks_toe\" , \"std_left_peaks_toe\"\n",
        "                                                                      , \"mean_right_peaks_heel\" , \"std_right_peaks_heel\"\n",
        "                                                                      , \"mean_right_peaks_toe\" , \"std_right_peaks_toe\"\n",
        "                                                                       ])\n",
        "  def ind_stances_to_df(self):\n",
        "    #droping NNOT WORKING HERE\n",
        "    self.left_stances_df = self.cycles_data_df.explode([\"l_stances\",\"l_swings_time\"])\n",
        "    self.left_stances_df = self.left_stances_df.drop(columns=[\"r_stances\" ,\t\"r_strides_time\" ,\t\"r_swings_time\" ,\t\"r_stances_time\", \"r_peaks_heel\" ,\t\"r_peaks_toe\"], axis=1)\n",
        "\n",
        "    self.right_stances_df = self.cycles_data_df.explode([\"r_stances\",\"r_swings_time\"])\n",
        "    self.right_stances_df = self.right_stances_df.drop(columns=[\"l_stances\" ,\t\"l_strides_time\" ,\t\"l_swings_time\" ,\t\"l_stances_time\", \"l_peaks_heel\" ,\t\"l_peaks_toe\"], axis=1)\n",
        "\n",
        "  def save_files(self):\n",
        "    self.signals_data_df.to_parquet(self.parquet_path+'signals.gzip')\n",
        "    self.cycles_data_df.to_parquet(self.parquet_path+'cyclels.gzip')\n",
        "    self.statical_data_df.to_parquet(self.parquet_path+'statical.gzip')\n",
        "    self.left_stances_df.to_parquet(self.parquet_path+'left_stances.gzip')\n",
        "    self.right_stances_df.to_parquet(self.parquet_path+'right_stances.gzip')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H_sk20fPzLuN"
      },
      "source": [
        "### Prepare the data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "76d430b27faf43bdadf157f552f13ae2",
            "29ebc352951d4fa0938beba218a394b8",
            "b6828ee43f97471abf45ccc60c277629",
            "ef49c0063f0244d2aeb3a2d9cab4ef4d",
            "549e36afb1344e5dac211583191e8bdf",
            "bff29d2b3d834c9ca2263759ef64a52e",
            "009a2cc28ae84588a4287bf39386769d",
            "2fabca6f213f46d5ac3f834242ebefa1",
            "884b0b1abc344bf2bdb4e38e6fc2fdbd",
            "0cc9fd76f4c34854b31e0409a13d62f3",
            "7c27d950cf8e4efebf2699e943a9e7d8"
          ]
        },
        "id": "n-MVu6AKJPk5",
        "outputId": "718416fb-5095-44da-8ae3-14cc87416fb1"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "76d430b27faf43bdadf157f552f13ae2",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "  0%|          | 0/304 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "data = Data( 1, csvD_path, demographics,parquet_path)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "IfujtCJCre60"
      },
      "outputs": [],
      "source": [
        "data.cycles_data_df"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6gvkVqJ2zQFA"
      },
      "source": [
        "### Read already prepared data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "XzE-whMYt6OS"
      },
      "outputs": [],
      "source": [
        "data = Data( 0, \"\", \"\",parquet_path)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vnyD09XjuQbC",
        "outputId": "eee8f78e-dc75-4e77-91e7-a87a1460bcd1"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "numpy.ndarray"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ],
      "source": [
        "type(data.signals_data_df[\"L1\"].iloc[-1]) "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gmQ47Qcp6D5S"
      },
      "source": [
        "### Example"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "qzvCeNFa80go",
        "outputId": "9abcaf0e-d474-4933-84e2-b5e1f0c051bc"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f30e86c43d0>]"
            ]
          },
          "execution_count": 90,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# example of one stance phase \n",
        "y = data.left_stances_df['l_stances'].iloc[7]\n",
        "import matplotlib.pyplot as plt\n",
        "plt.plot(y)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "id": "27AQ-HtRaTwA",
        "outputId": "435a1f28-6e12-4a00-b971-7000771c0486"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7fcaa4535280>]"
            ]
          },
          "execution_count": 15,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# example of one stance phase \n",
        "y = data.right_stances_df['r_stances'].iloc[3]\n",
        "import matplotlib.pyplot as plt\n",
        "plt.plot(y)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0_Wh3g6eut06"
      },
      "source": [
        "## FWHM\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "tzUwiIe49ZBp"
      },
      "outputs": [],
      "source": [
        "def fwhm(y_values_temp, x_values):\n",
        "    y_values, temp_l, temp_r = [], [], []\n",
        "    ## print(y_values_temp, x_values)\n",
        "    # To make 'y_values_temp', a numpy array, into a python list\n",
        "    for x in range(0,len(y_values_temp)):\n",
        "        y_values.append(y_values_temp[x])\n",
        "    peak_height = max(y_values)\n",
        "    half_peak_height = max(y_values)/2\n",
        "    \n",
        "    # Splitting the y_values data into before and after x_value at peak height\n",
        "    y_l_temp = y_values[0:y_values.index(peak_height)]\n",
        "    y_r_temp = y_values[y_values.index(peak_height):len(y_values)]\n",
        "    \n",
        "    # Find 1st closest value to half_peak_height in y_l and y_r\n",
        "    y_l = [abs(x-half_peak_height) for x in y_l_temp] # the distances\n",
        "    y_l_min = min(y_l)\n",
        "    point_left=y_l_temp[y_l.index(y_l_min)]  \n",
        "    \n",
        "    y_r = [abs(x-half_peak_height) for x in y_r_temp] # the distances\n",
        "    point_right=y_r_temp[y_r.index(min(y_r))]  \n",
        "    \n",
        "    # y_l = nsmallest(1, y_l_temp, key=lambda x: abs(x-half_peak_height))\n",
        "    # y_r = nsmallest(1, y_r_temp, key=lambda x: abs(x-half_peak_height))\n",
        "    \n",
        "    # Gets x_value pairs for y_l and y_r\n",
        "    temp_l.append(x_values[y_l_temp.index(point_left)])\n",
        "    temp_r.append(x_values[y_r_temp.index(point_right)])\n",
        "    fwhm_n = temp_l[0] - temp_r[0]\n",
        "    return abs(fwhm_n)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "BDcSn5CwlWEH"
      },
      "outputs": [],
      "source": [
        "# data.left_stances_df.drop(index=data.left_stances_df.index[0], axis=0, inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "wfhLUTe0nxw0"
      },
      "outputs": [],
      "source": [
        "#data.left_stances_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "MwGB0wAvonbm"
      },
      "outputs": [],
      "source": [
        "### SHOULD be applied for all dataframes after `explode`\n",
        "data.right_stances_df = data.right_stances_df.reset_index(drop=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "14ZC1TxJK9je"
      },
      "outputs": [],
      "source": [
        "data.right_stances_df['max_peak'] = data.right_stances_df['r_stances'].apply(lambda x: max(x))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "QQ7qyuDIMYqX"
      },
      "outputs": [],
      "source": [
        "data.right_stances_df['FWHM'] = data.right_stances_df['r_stances'].apply(lambda x: fwhm(x, np.arange(0, len(x)*0.01,0.01)))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 774
        },
        "id": "o4i5L4wQuWtC",
        "outputId": "9b8fb3dd-1fa2-444f-d7d9-7a6047337f68"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       ID  level  y                                          r_stances  \\\n",
              "0  GaPt03    3.0  1  [20.79, 27.72, 32.45, 37.07, 63.69, 110.33, 13...   \n",
              "1  GaPt03    3.0  1  [52.8, 69.19, 57.64, 51.7, 60.94, 92.84, 140.1...   \n",
              "2  GaPt03    3.0  1  [23.1, 32.45, 39.38, 46.31, 116.82, 155.43, 17...   \n",
              "3  GaPt03    3.0  1  [23.1, 41.69, 55.55, 119.13, 154.55, 200.42, 2...   \n",
              "4  GaPt03    3.0  1  [25.41, 37.07, 55.55, 106.37, 152.02, 182.05, ...   \n",
              "\n",
              "                                      r_strides_time  r_swings_time  \\\n",
              "0  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.39   \n",
              "1  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.21   \n",
              "2  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.41   \n",
              "3  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.43   \n",
              "4  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.41   \n",
              "\n",
              "                                      r_stances_time  \\\n",
              "0  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "1  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "2  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "3  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "4  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "\n",
              "                                        r_peaks_heel  \\\n",
              "0  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "1  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "2  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "3  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "4  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "\n",
              "                                         r_peaks_toe  max_peak  FWHM  \n",
              "0  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    731.61  0.48  \n",
              "1  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    774.73  0.20  \n",
              "2  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    717.75  0.05  \n",
              "3  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    742.50  0.10  \n",
              "4  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    779.79  0.17  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-fc46a381-1262-4f49-898f-db617e11ebf4\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>level</th>\n",
              "      <th>y</th>\n",
              "      <th>r_stances</th>\n",
              "      <th>r_strides_time</th>\n",
              "      <th>r_swings_time</th>\n",
              "      <th>r_stances_time</th>\n",
              "      <th>r_peaks_heel</th>\n",
              "      <th>r_peaks_toe</th>\n",
              "      <th>max_peak</th>\n",
              "      <th>FWHM</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[20.79, 27.72, 32.45, 37.07, 63.69, 110.33, 13...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.39</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>731.61</td>\n",
              "      <td>0.48</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[52.8, 69.19, 57.64, 51.7, 60.94, 92.84, 140.1...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.21</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>774.73</td>\n",
              "      <td>0.20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[23.1, 32.45, 39.38, 46.31, 116.82, 155.43, 17...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.41</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>717.75</td>\n",
              "      <td>0.05</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[23.1, 41.69, 55.55, 119.13, 154.55, 200.42, 2...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.43</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>742.50</td>\n",
              "      <td>0.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[25.41, 37.07, 55.55, 106.37, 152.02, 182.05, ...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.41</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>779.79</td>\n",
              "      <td>0.17</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fc46a381-1262-4f49-898f-db617e11ebf4')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-fc46a381-1262-4f49-898f-db617e11ebf4 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-fc46a381-1262-4f49-898f-db617e11ebf4');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ],
      "source": [
        "data.right_stances_df.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zx3AE4Ab0bRL"
      },
      "source": [
        "## Padding"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "-8UGcqn314MB"
      },
      "outputs": [],
      "source": [
        "max_length = max([len(row) for row in data.right_stances_df['r_stances']])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "3Tn0jp1d1_4e"
      },
      "outputs": [],
      "source": [
        "padded_data = np.zeros((len(data.right_stances_df['r_stances']), max_length))\n",
        "for i, row in enumerate(data.right_stances_df['r_stances']):\n",
        "    padded_data[i, :len(row)] = row"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "OlsSCtbp90I6"
      },
      "outputs": [],
      "source": [
        "padded_data=np.reshape(padded_data,(len(data.right_stances_df['r_stances']), 1,-1))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "2sJpjrRP-JGb"
      },
      "outputs": [],
      "source": [
        "padded_data=padded_data.tolist()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "1yN3RLYq7b0g"
      },
      "outputs": [],
      "source": [
        "padded_data_df = pd.DataFrame(padded_data, columns =[\"padded_stances\"])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 852
        },
        "id": "afxvzMR2-pjl",
        "outputId": "ea1094cd-5442-4ca9-dd8b-57ee2e1de5d0"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                          padded_stances\n",
              "0      [20.79, 27.72, 32.45, 37.07, 63.69, 110.33, 13...\n",
              "1      [52.8, 69.19, 57.64, 51.7, 60.94, 92.84, 140.1...\n",
              "2      [23.1, 32.45, 39.38, 46.31, 116.82, 155.43, 17...\n",
              "3      [23.1, 41.69, 55.55, 119.13, 154.55, 200.42, 2...\n",
              "4      [25.41, 37.07, 55.55, 106.37, 152.02, 182.05, ...\n",
              "...                                                  ...\n",
              "18382  [78.65, 302.28, 390.28, 478.83, 535.81, 577.39...\n",
              "18383  [27.39, 213.51, 264.44, 391.15999999999997, 46...\n",
              "18384  [47.96, 264.44, 309.65000000000003, 429.11, 50...\n",
              "18385  [61.93, 278.3, 327.58, 446.82, 507.53999999999...\n",
              "18386  [29.7, 211.09, 285.34, 393.03, 465.3, 524.8100...\n",
              "\n",
              "[18387 rows x 1 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c3aeeffe-33b1-4185-aa13-634e313245b4\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>padded_stances</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>[20.79, 27.72, 32.45, 37.07, 63.69, 110.33, 13...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>[52.8, 69.19, 57.64, 51.7, 60.94, 92.84, 140.1...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>[23.1, 32.45, 39.38, 46.31, 116.82, 155.43, 17...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>[23.1, 41.69, 55.55, 119.13, 154.55, 200.42, 2...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>[25.41, 37.07, 55.55, 106.37, 152.02, 182.05, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18382</th>\n",
              "      <td>[78.65, 302.28, 390.28, 478.83, 535.81, 577.39...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18383</th>\n",
              "      <td>[27.39, 213.51, 264.44, 391.15999999999997, 46...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18384</th>\n",
              "      <td>[47.96, 264.44, 309.65000000000003, 429.11, 50...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18385</th>\n",
              "      <td>[61.93, 278.3, 327.58, 446.82, 507.53999999999...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18386</th>\n",
              "      <td>[29.7, 211.09, 285.34, 393.03, 465.3, 524.8100...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>18387 rows × 1 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c3aeeffe-33b1-4185-aa13-634e313245b4')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-c3aeeffe-33b1-4185-aa13-634e313245b4 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-c3aeeffe-33b1-4185-aa13-634e313245b4');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ],
      "source": [
        "padded_data_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "BPwb9J0v-v4v"
      },
      "outputs": [],
      "source": [
        "data.right_stances_df['padded_stances']=padded_data_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 774
        },
        "id": "KK2Ihg-q_AQv",
        "outputId": "e031ddd2-2553-4f1d-c1ec-7dcebdd82019"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       ID  level  y                                          r_stances  \\\n",
              "0  GaPt03    3.0  1  [20.79, 27.72, 32.45, 37.07, 63.69, 110.33, 13...   \n",
              "1  GaPt03    3.0  1  [52.8, 69.19, 57.64, 51.7, 60.94, 92.84, 140.1...   \n",
              "2  GaPt03    3.0  1  [23.1, 32.45, 39.38, 46.31, 116.82, 155.43, 17...   \n",
              "3  GaPt03    3.0  1  [23.1, 41.69, 55.55, 119.13, 154.55, 200.42, 2...   \n",
              "4  GaPt03    3.0  1  [25.41, 37.07, 55.55, 106.37, 152.02, 182.05, ...   \n",
              "\n",
              "                                      r_strides_time  r_swings_time  \\\n",
              "0  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.39   \n",
              "1  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.21   \n",
              "2  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.41   \n",
              "3  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.43   \n",
              "4  [1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...           0.41   \n",
              "\n",
              "                                      r_stances_time  \\\n",
              "0  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "1  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "2  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "3  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "4  [1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...   \n",
              "\n",
              "                                        r_peaks_heel  \\\n",
              "0  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "1  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "2  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "3  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "4  [107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...   \n",
              "\n",
              "                                         r_peaks_toe  max_peak  FWHM  \\\n",
              "0  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    731.61  0.48   \n",
              "1  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    774.73  0.20   \n",
              "2  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    717.75  0.05   \n",
              "3  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    742.50  0.10   \n",
              "4  [80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...    779.79  0.17   \n",
              "\n",
              "                                      padded_stances  \n",
              "0  [20.79, 27.72, 32.45, 37.07, 63.69, 110.33, 13...  \n",
              "1  [52.8, 69.19, 57.64, 51.7, 60.94, 92.84, 140.1...  \n",
              "2  [23.1, 32.45, 39.38, 46.31, 116.82, 155.43, 17...  \n",
              "3  [23.1, 41.69, 55.55, 119.13, 154.55, 200.42, 2...  \n",
              "4  [25.41, 37.07, 55.55, 106.37, 152.02, 182.05, ...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-db01d311-46df-49fd-bbc2-23ac813fa124\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>level</th>\n",
              "      <th>y</th>\n",
              "      <th>r_stances</th>\n",
              "      <th>r_strides_time</th>\n",
              "      <th>r_swings_time</th>\n",
              "      <th>r_stances_time</th>\n",
              "      <th>r_peaks_heel</th>\n",
              "      <th>r_peaks_toe</th>\n",
              "      <th>max_peak</th>\n",
              "      <th>FWHM</th>\n",
              "      <th>padded_stances</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[20.79, 27.72, 32.45, 37.07, 63.69, 110.33, 13...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.39</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>731.61</td>\n",
              "      <td>0.48</td>\n",
              "      <td>[20.79, 27.72, 32.45, 37.07, 63.69, 110.33, 13...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[52.8, 69.19, 57.64, 51.7, 60.94, 92.84, 140.1...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.21</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>774.73</td>\n",
              "      <td>0.20</td>\n",
              "      <td>[52.8, 69.19, 57.64, 51.7, 60.94, 92.84, 140.1...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[23.1, 32.45, 39.38, 46.31, 116.82, 155.43, 17...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.41</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>717.75</td>\n",
              "      <td>0.05</td>\n",
              "      <td>[23.1, 32.45, 39.38, 46.31, 116.82, 155.43, 17...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[23.1, 41.69, 55.55, 119.13, 154.55, 200.42, 2...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.43</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>742.50</td>\n",
              "      <td>0.10</td>\n",
              "      <td>[23.1, 41.69, 55.55, 119.13, 154.55, 200.42, 2...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>GaPt03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1</td>\n",
              "      <td>[25.41, 37.07, 55.55, 106.37, 152.02, 182.05, ...</td>\n",
              "      <td>[1.59, 1.27, 1.41, 1.49, 1.56, 1.41, 1.5, 1.32...</td>\n",
              "      <td>0.41</td>\n",
              "      <td>[1.19, 1.05, 0.99, 1.05, 1.1400000000000001, 0...</td>\n",
              "      <td>[107.8, 71.17, 111.1, 141.24, 101.2, 123.42, 1...</td>\n",
              "      <td>[80.63, 134.42, 80.63, 71.61, 98.78, 130.02, 8...</td>\n",
              "      <td>779.79</td>\n",
              "      <td>0.17</td>\n",
              "      <td>[25.41, 37.07, 55.55, 106.37, 152.02, 182.05, ...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-db01d311-46df-49fd-bbc2-23ac813fa124')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-db01d311-46df-49fd-bbc2-23ac813fa124 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-db01d311-46df-49fd-bbc2-23ac813fa124');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ],
      "source": [
        "data.right_stances_df.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5njlOxC_gRAB"
      },
      "source": [
        "## train - test data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "id": "RttYJ46jgVHI"
      },
      "outputs": [],
      "source": [
        "df1 = data.right_stances_df[[ 'padded_stances','r_swings_time','max_peak','FWHM']]\n",
        "y = data.right_stances_df[[ 'y']]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "fYtjnASYhnLg"
      },
      "outputs": [],
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(df1, y, random_state=42)"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "HQJ1mk534vhR"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "S6HJ1kKDkigv"
      },
      "source": [
        "## Scaling"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "id": "nFE6OIR5qRsk"
      },
      "outputs": [],
      "source": [
        "# create a scaler object\n",
        "scaler = StandardScaler()\n",
        "\n",
        "# fit and transform the data\n",
        "scaled_data = scaler.fit_transform(np.stack(X_train['padded_stances'].values))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ex1YrGtNku1M"
      },
      "outputs": [],
      "source": [
        "# X_scaled = scaler.transform(X_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "id": "pVjBdD2IMBXA"
      },
      "outputs": [],
      "source": [
        "scaled_data=scaled_data.tolist()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "Zb3coQsTK1Jw"
      },
      "outputs": [],
      "source": [
        "# X_train['scaled_stances']= pd.DataFrame(scaled_data, columns =[\"scaled_data\"]) # this way caused some values to become Nulll\n",
        "X_train.insert(0, \"scaled_stances\", scaled_data, True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 870
        },
        "id": "GkRmPjj3K7kH",
        "outputId": "5950f9c2-24fc-49df-b93a-53e987160c7c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                          scaled_stances  \\\n",
              "337    [-0.09027499699554237, -0.6566456091893658, -0...   \n",
              "11848  [-0.13374517285585133, 0.28410271089387606, 0....   \n",
              "7791   [0.026760091859135693, 0.4419711728509268, 0.7...   \n",
              "2344   [-0.1772153487161603, 0.48166909018515575, 1.3...   \n",
              "4004   [-0.5951978089114389, -1.041623086593401, -1.3...   \n",
              "...                                                  ...   \n",
              "11284  [1.0332618560093667, 1.967109764389215, 1.6276...   \n",
              "11964  [-0.5584153524142543, -0.3436783074148622, -0....   \n",
              "5390   [-0.24743640202896694, -0.20704361519472508, 0...   \n",
              "860    [-0.5951978089114389, -1.0222357316162194, -1....   \n",
              "15795  [-0.5985416685930011, 0.7097013130120065, 1.19...   \n",
              "\n",
              "                                          padded_stances  r_swings_time  \\\n",
              "337    [41.69, 77.88, 162.14, 243.98, 317.78999999999...           0.37   \n",
              "11848  [40.26, 189.96999999999997, 284.57, 341.55, 36...           0.33   \n",
              "7791   [45.54, 208.78, 374.11, 549.78, 667.26, 710.93...           0.34   \n",
              "2344   [38.83, 213.51, 467.06, 482.13, 533.39, 602.14...           0.38   \n",
              "4004   [25.08, 32.01, 50.27, 92.95, 222.09, 318.78, 3...           0.38   \n",
              "...                                                  ...            ...   \n",
              "11284  [78.65, 390.5, 514.8000000000001, 643.72, 692....           0.38   \n",
              "11964  [26.29, 115.17, 235.18, 274.23, 297.6599999999...           0.30   \n",
              "5390   [36.52, 131.45, 281.05, 308.98999999999995, 32...           0.42   \n",
              "860    [25.08, 34.32, 82.83, 160.05, 260.37, 404.6900...           0.37   \n",
              "15795  [24.97, 240.68, 447.15, 479.04999999999995, 53...           0.33   \n",
              "\n",
              "       max_peak  FWHM  \n",
              "337      816.64  0.36  \n",
              "11848   1042.91  0.13  \n",
              "7791     846.67  0.11  \n",
              "2344    1127.39  0.37  \n",
              "4004     990.66  0.37  \n",
              "...         ...   ...  \n",
              "11284   1147.41  0.35  \n",
              "11964    892.87  0.07  \n",
              "5390     874.61  0.41  \n",
              "860      871.42  0.43  \n",
              "15795    863.61  0.38  \n",
              "\n",
              "[13790 rows x 5 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-9f35722f-8d4a-4d0d-ac6f-62f4a4aa06bb\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>scaled_stances</th>\n",
              "      <th>padded_stances</th>\n",
              "      <th>r_swings_time</th>\n",
              "      <th>max_peak</th>\n",
              "      <th>FWHM</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>337</th>\n",
              "      <td>[-0.09027499699554237, -0.6566456091893658, -0...</td>\n",
              "      <td>[41.69, 77.88, 162.14, 243.98, 317.78999999999...</td>\n",
              "      <td>0.37</td>\n",
              "      <td>816.64</td>\n",
              "      <td>0.36</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11848</th>\n",
              "      <td>[-0.13374517285585133, 0.28410271089387606, 0....</td>\n",
              "      <td>[40.26, 189.96999999999997, 284.57, 341.55, 36...</td>\n",
              "      <td>0.33</td>\n",
              "      <td>1042.91</td>\n",
              "      <td>0.13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7791</th>\n",
              "      <td>[0.026760091859135693, 0.4419711728509268, 0.7...</td>\n",
              "      <td>[45.54, 208.78, 374.11, 549.78, 667.26, 710.93...</td>\n",
              "      <td>0.34</td>\n",
              "      <td>846.67</td>\n",
              "      <td>0.11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2344</th>\n",
              "      <td>[-0.1772153487161603, 0.48166909018515575, 1.3...</td>\n",
              "      <td>[38.83, 213.51, 467.06, 482.13, 533.39, 602.14...</td>\n",
              "      <td>0.38</td>\n",
              "      <td>1127.39</td>\n",
              "      <td>0.37</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4004</th>\n",
              "      <td>[-0.5951978089114389, -1.041623086593401, -1.3...</td>\n",
              "      <td>[25.08, 32.01, 50.27, 92.95, 222.09, 318.78, 3...</td>\n",
              "      <td>0.38</td>\n",
              "      <td>990.66</td>\n",
              "      <td>0.37</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11284</th>\n",
              "      <td>[1.0332618560093667, 1.967109764389215, 1.6276...</td>\n",
              "      <td>[78.65, 390.5, 514.8000000000001, 643.72, 692....</td>\n",
              "      <td>0.38</td>\n",
              "      <td>1147.41</td>\n",
              "      <td>0.35</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11964</th>\n",
              "      <td>[-0.5584153524142543, -0.3436783074148622, -0....</td>\n",
              "      <td>[26.29, 115.17, 235.18, 274.23, 297.6599999999...</td>\n",
              "      <td>0.30</td>\n",
              "      <td>892.87</td>\n",
              "      <td>0.07</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5390</th>\n",
              "      <td>[-0.24743640202896694, -0.20704361519472508, 0...</td>\n",
              "      <td>[36.52, 131.45, 281.05, 308.98999999999995, 32...</td>\n",
              "      <td>0.42</td>\n",
              "      <td>874.61</td>\n",
              "      <td>0.41</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>860</th>\n",
              "      <td>[-0.5951978089114389, -1.0222357316162194, -1....</td>\n",
              "      <td>[25.08, 34.32, 82.83, 160.05, 260.37, 404.6900...</td>\n",
              "      <td>0.37</td>\n",
              "      <td>871.42</td>\n",
              "      <td>0.43</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15795</th>\n",
              "      <td>[-0.5985416685930011, 0.7097013130120065, 1.19...</td>\n",
              "      <td>[24.97, 240.68, 447.15, 479.04999999999995, 53...</td>\n",
              "      <td>0.33</td>\n",
              "      <td>863.61</td>\n",
              "      <td>0.38</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>13790 rows × 5 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9f35722f-8d4a-4d0d-ac6f-62f4a4aa06bb')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-9f35722f-8d4a-4d0d-ac6f-62f4a4aa06bb button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-9f35722f-8d4a-4d0d-ac6f-62f4a4aa06bb');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 506
        }
      ],
      "source": [
        "X_train"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "B_kWzzUSYpp4",
        "outputId": "bdc86ba5-f52e-476c-a85a-e04a988354ee"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "list"
            ]
          },
          "metadata": {},
          "execution_count": 507
        }
      ],
      "source": [
        "type(X_train['scaled_stances'].iloc[0])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "evtbYEBeOb7q"
      },
      "source": [
        "## The data\n",
        "yalla\n",
        "We're gonna use the features (r_swings_time \tmax_peak \tFWHM \tscaled_stances)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "lCllgMsTOjvQ",
        "outputId": "3408fdf9-452f-431b-b2fe-a06ce3a10e1a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                          scaled_stances  \\\n",
              "337    [-0.09027499699554237, -0.6566456091893658, -0...   \n",
              "11848  [-0.13374517285585133, 0.28410271089387606, 0....   \n",
              "7791   [0.026760091859135693, 0.4419711728509268, 0.7...   \n",
              "2344   [-0.1772153487161603, 0.48166909018515575, 1.3...   \n",
              "4004   [-0.5951978089114389, -1.041623086593401, -1.3...   \n",
              "\n",
              "                                          padded_stances  r_swings_time  \\\n",
              "337    [41.69, 77.88, 162.14, 243.98, 317.78999999999...           0.37   \n",
              "11848  [40.26, 189.96999999999997, 284.57, 341.55, 36...           0.33   \n",
              "7791   [45.54, 208.78, 374.11, 549.78, 667.26, 710.93...           0.34   \n",
              "2344   [38.83, 213.51, 467.06, 482.13, 533.39, 602.14...           0.38   \n",
              "4004   [25.08, 32.01, 50.27, 92.95, 222.09, 318.78, 3...           0.38   \n",
              "\n",
              "       max_peak  FWHM  \n",
              "337      816.64  0.36  \n",
              "11848   1042.91  0.13  \n",
              "7791     846.67  0.11  \n",
              "2344    1127.39  0.37  \n",
              "4004     990.66  0.37  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-31801879-5152-45b1-a785-3de930ea0f90\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>scaled_stances</th>\n",
              "      <th>padded_stances</th>\n",
              "      <th>r_swings_time</th>\n",
              "      <th>max_peak</th>\n",
              "      <th>FWHM</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>337</th>\n",
              "      <td>[-0.09027499699554237, -0.6566456091893658, -0...</td>\n",
              "      <td>[41.69, 77.88, 162.14, 243.98, 317.78999999999...</td>\n",
              "      <td>0.37</td>\n",
              "      <td>816.64</td>\n",
              "      <td>0.36</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11848</th>\n",
              "      <td>[-0.13374517285585133, 0.28410271089387606, 0....</td>\n",
              "      <td>[40.26, 189.96999999999997, 284.57, 341.55, 36...</td>\n",
              "      <td>0.33</td>\n",
              "      <td>1042.91</td>\n",
              "      <td>0.13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7791</th>\n",
              "      <td>[0.026760091859135693, 0.4419711728509268, 0.7...</td>\n",
              "      <td>[45.54, 208.78, 374.11, 549.78, 667.26, 710.93...</td>\n",
              "      <td>0.34</td>\n",
              "      <td>846.67</td>\n",
              "      <td>0.11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2344</th>\n",
              "      <td>[-0.1772153487161603, 0.48166909018515575, 1.3...</td>\n",
              "      <td>[38.83, 213.51, 467.06, 482.13, 533.39, 602.14...</td>\n",
              "      <td>0.38</td>\n",
              "      <td>1127.39</td>\n",
              "      <td>0.37</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4004</th>\n",
              "      <td>[-0.5951978089114389, -1.041623086593401, -1.3...</td>\n",
              "      <td>[25.08, 32.01, 50.27, 92.95, 222.09, 318.78, 3...</td>\n",
              "      <td>0.38</td>\n",
              "      <td>990.66</td>\n",
              "      <td>0.37</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-31801879-5152-45b1-a785-3de930ea0f90')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-31801879-5152-45b1-a785-3de930ea0f90 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-31801879-5152-45b1-a785-3de930ea0f90');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 509
        }
      ],
      "source": [
        "X_train.head()"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "c =X_train[[\"scaled_stances\",\"r_swings_time\",\"FWHM\",\"max_peak\"]]"
      ],
      "metadata": {
        "id": "YzzZK_6Hvrx_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "c.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "j5j1zN2svrs8",
        "outputId": "580fa7f4-dca8-43ed-ecbb-73243f61d4bd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                          scaled_stances  r_swings_time  FWHM  \\\n",
              "337    [-0.09027499699554237, -0.6566456091893658, -0...           0.37  0.36   \n",
              "11848  [-0.13374517285585133, 0.28410271089387606, 0....           0.33  0.13   \n",
              "7791   [0.026760091859135693, 0.4419711728509268, 0.7...           0.34  0.11   \n",
              "2344   [-0.1772153487161603, 0.48166909018515575, 1.3...           0.38  0.37   \n",
              "4004   [-0.5951978089114389, -1.041623086593401, -1.3...           0.38  0.37   \n",
              "\n",
              "       max_peak  \n",
              "337      816.64  \n",
              "11848   1042.91  \n",
              "7791     846.67  \n",
              "2344    1127.39  \n",
              "4004     990.66  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2a58ffe2-37c8-40fa-b4eb-c888895e322a\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>scaled_stances</th>\n",
              "      <th>r_swings_time</th>\n",
              "      <th>FWHM</th>\n",
              "      <th>max_peak</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>337</th>\n",
              "      <td>[-0.09027499699554237, -0.6566456091893658, -0...</td>\n",
              "      <td>0.37</td>\n",
              "      <td>0.36</td>\n",
              "      <td>816.64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11848</th>\n",
              "      <td>[-0.13374517285585133, 0.28410271089387606, 0....</td>\n",
              "      <td>0.33</td>\n",
              "      <td>0.13</td>\n",
              "      <td>1042.91</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7791</th>\n",
              "      <td>[0.026760091859135693, 0.4419711728509268, 0.7...</td>\n",
              "      <td>0.34</td>\n",
              "      <td>0.11</td>\n",
              "      <td>846.67</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2344</th>\n",
              "      <td>[-0.1772153487161603, 0.48166909018515575, 1.3...</td>\n",
              "      <td>0.38</td>\n",
              "      <td>0.37</td>\n",
              "      <td>1127.39</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4004</th>\n",
              "      <td>[-0.5951978089114389, -1.041623086593401, -1.3...</td>\n",
              "      <td>0.38</td>\n",
              "      <td>0.37</td>\n",
              "      <td>990.66</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2a58ffe2-37c8-40fa-b4eb-c888895e322a')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-2a58ffe2-37c8-40fa-b4eb-c888895e322a button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-2a58ffe2-37c8-40fa-b4eb-c888895e322a');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 512
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "glZHhvGjyTjw"
      },
      "source": [
        "### Reformat the data to be fitted to a tensor then CNN"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### The first format \n",
        "is as follows a numpy array of (13790, 1) shape, the length of each row is 750, the first 747 elements are the scaled stances, then three float numbers : swing time, max Peak, FWHM value"
      ],
      "metadata": {
        "id": "yiOwaZzBNxwy"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0EjC4Bi8ZzrO"
      },
      "outputs": [],
      "source": [
        "# data= [ np.asarray(x['scaled_stances'].tolist()) , np.asarray(x[[\"r_swings_time\" ]]),  np.asarray(x[[\"max_peak\"]]) \t, np.asarray(x[[\"FWHM\" ]]) ]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-PMBIOtpuU9E"
      },
      "outputs": [],
      "source": [
        "# u = np.asarray(x['scaled_stances'].tolist()).reshape(13790, 1,747)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "HTQ-Bi81kfY-"
      },
      "outputs": [],
      "source": [
        "data = np.append( np.asarray(X_train['scaled_stances'].tolist()) , np.asarray(X_train[[\"r_swings_time\" ]]),  axis=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "id": "EzGs3vhUunv6"
      },
      "outputs": [],
      "source": [
        "data = np.append( data , np.asarray(X_train[[\"max_peak\" ]]),  axis=1)\n",
        "data = np.append( data , np.asarray(X_train[[\"FWHM\" ]]),  axis=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-vrON8Cwuq_6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d5187d7f-fc71-477a-c0e1-a43409a71d7d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([-9.02749970e-02, -6.56645609e-01, -6.52911900e-01, -6.51835049e-01,\n",
              "       -6.80455819e-01, -7.09849904e-01, -7.31343737e-01, -7.58305251e-01,\n",
              "       -6.99138121e-01, -6.70089052e-01, -6.56100096e-01, -5.58430922e-01,\n",
              "       -4.24262579e-01, -3.97973652e-01, -3.72499450e-01, -3.59456529e-01,\n",
              "       -3.66312024e-01, -3.30327851e-01, -3.11711450e-01, -2.69219187e-01,\n",
              "       -2.85570010e-01, -2.57676917e-01, -3.12587831e-01, -3.44655496e-01,\n",
              "       -3.69652015e-01, -3.99892110e-01, -4.11848471e-01, -4.18129816e-01,\n",
              "       -4.50862014e-01, -4.85263057e-01, -4.76090410e-01, -5.06129677e-01,\n",
              "       -3.70910396e-01, -4.60304469e-01, -5.03264489e-01, -5.76371671e-01,\n",
              "       -6.19112793e-01, -6.77439031e-01, -7.11423413e-01, -9.42336071e-01,\n",
              "       -9.61340118e-01, -9.44113006e-01, -9.58073881e-01, -1.07219121e+00,\n",
              "       -9.57724554e-01, -8.51393802e-01, -8.46041730e-01, -6.87435980e-01,\n",
              "       -5.20480129e-01, -3.69789355e-01, -2.68160742e-01, -1.52371438e-01,\n",
              "       -5.81269098e-02,  3.67997402e-02,  9.57552877e-02,  9.00221676e-02,\n",
              "        1.40563799e-01,  1.08116673e-01,  6.86620503e-02,  9.77363422e-03,\n",
              "       -1.08050216e-01, -1.99922762e-01, -2.72937183e-01, -4.31177065e-01,\n",
              "       -5.94197328e-01, -7.64617119e-01, -7.47947647e-01, -6.88826000e-01,\n",
              "       -7.47918804e-01, -6.93078733e-01, -6.43377309e-01, -5.98604875e-01,\n",
              "       -5.57012643e-01, -5.19651877e-01, -4.84691003e-01, -4.52818368e-01,\n",
              "       -4.23131119e-01, -3.95214701e-01, -3.69715981e-01, -3.46785565e-01,\n",
              "       -3.25812613e-01, -3.06454011e-01, -2.88969470e-01, -2.73816845e-01,\n",
              "       -2.60441890e-01, -2.48431464e-01, -2.38338115e-01, -2.29766559e-01,\n",
              "       -2.22518743e-01, -2.15814158e-01, -2.09852312e-01, -2.04193437e-01,\n",
              "       -1.99037247e-01, -1.94188199e-01, -1.89319305e-01, -1.84499476e-01,\n",
              "       -1.79936041e-01, -1.75527574e-01, -1.71245568e-01, -1.66733509e-01,\n",
              "       -1.62420807e-01, -1.58262819e-01, -1.54230736e-01, -1.50058250e-01,\n",
              "       -1.45725825e-01, -1.41564821e-01, -1.37917045e-01, -1.34103903e-01,\n",
              "       -1.30685995e-01, -1.27174528e-01, -1.24107576e-01, -1.21095322e-01,\n",
              "       -1.17827742e-01, -1.14629507e-01, -1.11734497e-01, -1.08894584e-01,\n",
              "       -1.06331810e-01, -1.03834945e-01, -1.01596655e-01, -9.93525004e-02,\n",
              "       -9.72494819e-02, -9.53255226e-02, -9.33707967e-02, -9.16712130e-02,\n",
              "       -9.01794531e-02, -8.87511400e-02, -8.73586177e-02, -8.60645003e-02,\n",
              "       -8.47022613e-02, -8.33736690e-02, -8.21015075e-02, -8.10235298e-02,\n",
              "       -7.98514517e-02, -7.86526198e-02, -7.74580922e-02, -7.65495747e-02,\n",
              "       -7.56052285e-02, -7.48012340e-02, -7.41140204e-02, -7.34531423e-02,\n",
              "       -7.27512929e-02, -7.19984498e-02, -7.11737050e-02, -7.03475580e-02,\n",
              "       -6.93107177e-02, -6.82139677e-02, -6.72055729e-02, -6.62081377e-02,\n",
              "       -6.53809879e-02, -6.45041170e-02, -6.36134870e-02, -6.27696494e-02,\n",
              "       -6.18433268e-02, -6.11805562e-02, -6.02123755e-02, -5.94286448e-02,\n",
              "       -5.87787205e-02, -5.80191249e-02, -5.73185303e-02, -5.66710295e-02,\n",
              "       -5.59751754e-02, -5.54032111e-02, -5.48198424e-02, -5.42698372e-02,\n",
              "       -5.37436650e-02, -5.33397027e-02, -5.27959251e-02, -5.23855522e-02,\n",
              "       -5.21531930e-02, -5.21123569e-02, -5.18977454e-02, -5.16077887e-02,\n",
              "       -5.14893012e-02, -5.14556402e-02, -5.13523793e-02, -5.13587703e-02,\n",
              "       -5.12259079e-02, -5.10991003e-02, -5.10064364e-02, -5.08203070e-02,\n",
              "       -5.06172661e-02, -5.02547297e-02, -4.99613701e-02, -4.97049974e-02,\n",
              "       -4.94635649e-02, -4.92310736e-02, -4.91087213e-02, -4.89973591e-02,\n",
              "       -4.88490177e-02, -4.86533635e-02, -4.85010219e-02, -4.83558878e-02,\n",
              "       -4.81024402e-02, -4.78073919e-02, -4.73881161e-02, -4.69681520e-02,\n",
              "       -4.64892131e-02, -4.60463376e-02, -4.55762627e-02, -4.50285718e-02,\n",
              "       -4.45909859e-02, -4.42181913e-02, -4.39935332e-02, -4.36868725e-02,\n",
              "       -4.34789130e-02, -4.32714862e-02, -4.30880429e-02, -4.28446407e-02,\n",
              "       -4.26669163e-02, -4.25077732e-02, -4.22664161e-02, -4.19685169e-02,\n",
              "       -4.14837755e-02, -4.10530035e-02, -4.04298833e-02, -3.97387093e-02,\n",
              "       -3.91322518e-02, -3.83798946e-02, -3.77253008e-02, -3.71896396e-02,\n",
              "       -3.64761743e-02, -3.57140675e-02, -3.50214524e-02, -3.43893913e-02,\n",
              "       -3.38307188e-02, -3.31944745e-02, -3.28047743e-02, -3.24989991e-02,\n",
              "       -3.20665011e-02, -3.17827049e-02, -3.14903545e-02, -3.11396662e-02,\n",
              "       -3.06367306e-02, -3.01884977e-02, -2.96024949e-02, -2.92876040e-02,\n",
              "       -2.89393647e-02, -2.84840367e-02, -2.79908069e-02, -2.75952688e-02,\n",
              "       -2.72630278e-02, -2.69532738e-02, -2.67798168e-02, -2.64801558e-02,\n",
              "       -2.60419381e-02, -2.59667960e-02, -2.58728836e-02, -2.56488100e-02,\n",
              "       -2.52885104e-02, -2.47777956e-02, -2.42989427e-02, -2.40059459e-02,\n",
              "       -2.35553704e-02, -2.30934509e-02, -2.26128078e-02, -2.21115456e-02,\n",
              "       -2.15988507e-02, -2.14494326e-02, -2.11912236e-02, -2.09161572e-02,\n",
              "       -2.04692781e-02, -2.02003305e-02, -1.98741301e-02, -1.95436567e-02,\n",
              "       -1.90213509e-02, -1.87318234e-02, -1.83058787e-02, -1.76587771e-02,\n",
              "       -1.74170098e-02, -1.74350758e-02, -1.74540046e-02, -1.74601920e-02,\n",
              "       -1.74799717e-02, -1.74572737e-02, -1.74528594e-02, -1.74847403e-02,\n",
              "       -1.74448001e-02, -1.74222643e-02, -1.74612498e-02, -1.73765113e-02,\n",
              "       -1.75520244e-02, -1.75611378e-02, -1.76561926e-02, -1.76255586e-02,\n",
              "       -1.77213682e-02, -1.77279973e-02, -1.76600921e-02, -1.73531745e-02,\n",
              "       -1.71156103e-02, -1.68297721e-02, -1.64594988e-02, -1.58711588e-02,\n",
              "       -1.51384850e-02, -1.48347697e-02, -1.46002788e-02, -1.45989964e-02,\n",
              "       -1.46022934e-02, -1.45969819e-02, -1.45971796e-02, -1.45995244e-02,\n",
              "       -1.45978225e-02, -1.45969460e-02, -1.45893391e-02, -1.45875522e-02,\n",
              "       -1.45822159e-02, -1.45853434e-02, -1.45651574e-02, -1.45586747e-02,\n",
              "       -1.44924418e-02, -1.44777466e-02, -1.44384350e-02, -1.44131343e-02,\n",
              "       -1.43721300e-02, -1.41999661e-02, -1.41513839e-02, -1.40145837e-02,\n",
              "       -1.40359251e-02, -1.40271065e-02, -1.40515620e-02, -1.40522038e-02,\n",
              "       -1.40724619e-02, -1.41021517e-02, -1.41430838e-02, -1.41540404e-02,\n",
              "       -1.41569607e-02, -1.41681731e-02, -1.42635616e-02, -1.42870762e-02,\n",
              "       -1.42914830e-02, -1.43167281e-02, -1.43167281e-02, -1.42983401e-02,\n",
              "       -1.43224067e-02, -1.44037545e-02, -1.44090746e-02, -1.44253821e-02,\n",
              "       -1.44210342e-02, -1.44289281e-02, -1.44319123e-02, -1.44472104e-02,\n",
              "       -1.44677190e-02, -1.44739806e-02, -1.44745744e-02, -1.44786911e-02,\n",
              "       -1.44748973e-02, -1.44686815e-02, -1.44803418e-02, -1.44744611e-02,\n",
              "       -1.44820602e-02, -1.44980859e-02, -1.44979095e-02, -1.44889403e-02,\n",
              "       -1.44887866e-02, -1.45000602e-02, -1.44754478e-02, -1.44734325e-02,\n",
              "       -1.44692501e-02, -1.44656121e-02, -1.44663110e-02, -1.44561805e-02,\n",
              "       -1.44574701e-02, -1.44544050e-02, -1.44418100e-02, -1.44482715e-02,\n",
              "       -1.44293604e-02, -1.44317233e-02, -1.44146178e-02, -1.44195574e-02,\n",
              "       -1.44111096e-02, -1.44031131e-02, -1.44128411e-02, -1.44018747e-02,\n",
              "       -1.43712010e-02, -1.43868243e-02, -1.43836312e-02, -1.43797419e-02,\n",
              "       -1.43553499e-02, -1.43452730e-02, -1.43459723e-02, -1.43170243e-02,\n",
              "       -1.43114214e-02, -1.43006180e-02, -1.43065064e-02, -1.42958037e-02,\n",
              "       -1.42853451e-02, -1.42875309e-02, -1.42808114e-02, -1.42876501e-02,\n",
              "       -1.42794957e-02, -1.42859054e-02, -1.42763482e-02, -1.42737995e-02,\n",
              "       -1.42614707e-02, -1.42684550e-02, -1.42471438e-02, -1.42605873e-02,\n",
              "       -1.42380227e-02, -1.42212328e-02, -1.42335758e-02, -1.42215718e-02,\n",
              "       -1.42113227e-02, -1.41991213e-02, -1.41822529e-02, -1.41934000e-02,\n",
              "       -1.41776702e-02, -1.41624857e-02, -1.41249710e-02, -1.40926290e-02,\n",
              "       -1.40471346e-02, -1.39930501e-02, -1.39295487e-02, -1.38722099e-02,\n",
              "       -1.37184163e-02, -1.37049521e-02, -1.35826477e-02, -1.35364608e-02,\n",
              "       -1.33937149e-02, -1.31429154e-02, -1.30149430e-02, -1.28758571e-02,\n",
              "       -1.27315574e-02, -1.23988975e-02, -1.22684395e-02, -1.21958881e-02,\n",
              "       -1.21223146e-02, -1.20354050e-02, -1.17628177e-02, -1.14447599e-02,\n",
              "       -1.12940994e-02, -1.08136835e-02, -1.04504877e-02, -1.04005643e-02,\n",
              "       -1.04202547e-02, -1.03839158e-02, -1.04090272e-02, -1.03713832e-02,\n",
              "       -1.03952365e-02, -1.03991152e-02, -1.03964284e-02, -1.03907792e-02,\n",
              "       -1.03616442e-02, -1.03718256e-02, -1.04000006e-02, -1.04041314e-02,\n",
              "       -1.03953000e-02, -1.03950069e-02, -1.04029495e-02, -1.03982354e-02,\n",
              "       -1.04278231e-02, -1.04020639e-02, -1.04059068e-02, -1.04385893e-02,\n",
              "       -1.04136356e-02, -1.04467366e-02, -1.04849057e-02, -1.04750751e-02,\n",
              "       -1.04787512e-02, -1.04824397e-02, -1.04769116e-02, -1.05014903e-02,\n",
              "       -1.05138941e-02, -1.05039600e-02, -1.05367543e-02, -1.04856472e-02,\n",
              "       -1.05087470e-02, -1.05236712e-02, -1.05099902e-02, -1.05037883e-02,\n",
              "       -1.05103012e-02, -1.05118575e-02, -1.05199857e-02, -1.05177915e-02,\n",
              "       -1.05182544e-02, -1.05151237e-02, -1.05104442e-02, -1.04929085e-02,\n",
              "       -1.04960261e-02, -1.04845352e-02, -1.04854624e-02, -1.04527046e-02,\n",
              "       -1.05033071e-02, -1.04625163e-02, -1.04661551e-02, -1.04691969e-02,\n",
              "       -1.05120641e-02, -1.05095329e-02, -1.05120641e-02, -1.05021122e-02,\n",
              "       -1.05046283e-02, -1.05065191e-02, -1.05021122e-02, -1.04917935e-02,\n",
              "       -1.04756384e-02, -1.04616946e-02, -1.05072794e-02, -1.04730101e-02,\n",
              "       -1.04345057e-02, -1.04209340e-02, -1.03900898e-02, -1.03463279e-02,\n",
              "       -1.03193511e-02, -1.02788987e-02, -1.02695072e-02, -1.02487547e-02,\n",
              "       -1.02299639e-02, -1.01475081e-02, -1.01217219e-02, -1.00435739e-02,\n",
              "       -1.00151425e-02, -9.96975655e-03, -9.89401270e-03, -9.84984517e-03,\n",
              "       -9.80830152e-03, -9.79807061e-03, -9.78454449e-03, -9.80597453e-03,\n",
              "       -9.80136826e-03, -9.78305140e-03, -9.80433928e-03, -9.86644884e-03,\n",
              "       -9.88524623e-03, -9.96964093e-03, -1.00266866e-02, -1.01299647e-02,\n",
              "       -1.01938211e-02, -1.02525867e-02, -1.03856510e-02, -1.04928415e-02,\n",
              "       -1.07555211e-02, -1.09407047e-02, -1.11442806e-02, -1.13597816e-02,\n",
              "       -1.15567836e-02, -1.17299376e-02, -1.17820257e-02, -1.18519828e-02,\n",
              "       -1.19595151e-02, -1.20087622e-02, -1.13303649e-02, -1.08206270e-02,\n",
              "       -1.05794127e-02, -1.02570176e-02, -9.67693146e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,\n",
              "       -8.51596003e-03, -8.51596003e-03, -8.51596003e-03,  3.70000000e-01,\n",
              "        8.16640000e+02,  3.60000000e-01])"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ],
      "source": [
        "data[0]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pi5HcLQTHsg-",
        "outputId": "366b15d5-5b5f-45a8-a70e-af5459313673"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(13790, 750), dtype=float64, numpy=\n",
              "array([[-9.02749970e-02, -6.56645609e-01, -6.52911900e-01, ...,\n",
              "         3.70000000e-01,  8.16640000e+02,  3.60000000e-01],\n",
              "       [-1.33745173e-01,  2.84102711e-01,  1.38825966e-01, ...,\n",
              "         3.30000000e-01,  1.04291000e+03,  1.30000000e-01],\n",
              "       [ 2.67600919e-02,  4.41971173e-01,  7.17868753e-01, ...,\n",
              "         3.40000000e-01,  8.46670000e+02,  1.10000000e-01],\n",
              "       ...,\n",
              "       [-2.47436402e-01, -2.07043615e-01,  1.16062613e-01, ...,\n",
              "         4.20000000e-01,  8.74610000e+02,  4.10000000e-01],\n",
              "       [-5.95197809e-01, -1.02223573e+00, -1.16579869e+00, ...,\n",
              "         3.70000000e-01,  8.71420000e+02,  4.30000000e-01],\n",
              "       [-5.98541669e-01,  7.09701313e-01,  1.19020832e+00, ...,\n",
              "         3.30000000e-01,  8.63610000e+02,  3.80000000e-01]])>"
            ]
          },
          "execution_count": 27,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "tf.constant((data))"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "type(data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "blLobAyjYIaV",
        "outputId": "7c25745c-76c2-401b-e38f-86170fb67d56"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "numpy.ndarray"
            ]
          },
          "metadata": {},
          "execution_count": 77
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### The sexond format\n",
        "\n",
        "is as follows a numpy array of ((13790, 4)) shape, the first elements is a numpy array of length 747 represents the scaled stances, then three float numbers : swing time, max Peak, FWHM value"
      ],
      "metadata": {
        "id": "xI7iDgCZPsZt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_train['scaled_stances'] = X_train['scaled_stances'].apply(np.array)"
      ],
      "metadata": {
        "id": "ZvKh4mmqNfR1"
      },
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x = X_train[['scaled_stances', 'r_swings_time','max_peak','FWHM']]"
      ],
      "metadata": {
        "id": "goDGDcl7M89x"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "itemized_train_data = x.to_numpy()"
      ],
      "metadata": {
        "id": "fWR4Y_V4Mk9G"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "(itemized_train_data).shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4-d1b_XmMrJf",
        "outputId": "d354625d-69d9-4398-8c14-f47fc4313e35"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(13790, 4)"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "type(itemized_train_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dBPLvWdHYNO_",
        "outputId": "b63bdfa1-3a1b-4814-afc9-9a2bd682dbf8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "numpy.ndarray"
            ]
          },
          "metadata": {},
          "execution_count": 78
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "itemized_train_data[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H_4jxc0uYW_O",
        "outputId": "dc4dbbe7-0824-4811-b056-2ec6af7cdc23"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([array([-0.090275  , -0.65664561, -0.6529119 , -0.65183505, -0.68045582,\n",
              "              -0.7098499 , -0.73134374, -0.75830525, -0.69913812, -0.67008905,\n",
              "              -0.6561001 , -0.55843092, -0.42426258, -0.39797365, -0.37249945,\n",
              "              -0.35945653, -0.36631202, -0.33032785, -0.31171145, -0.26921919,\n",
              "              -0.28557001, -0.25767692, -0.31258783, -0.3446555 , -0.36965201,\n",
              "              -0.39989211, -0.41184847, -0.41812982, -0.45086201, -0.48526306,\n",
              "              -0.47609041, -0.50612968, -0.3709104 , -0.46030447, -0.50326449,\n",
              "              -0.57637167, -0.61911279, -0.67743903, -0.71142341, -0.94233607,\n",
              "              -0.96134012, -0.94411301, -0.95807388, -1.07219121, -0.95772455,\n",
              "              -0.8513938 , -0.84604173, -0.68743598, -0.52048013, -0.36978935,\n",
              "              -0.26816074, -0.15237144, -0.05812691,  0.03679974,  0.09575529,\n",
              "               0.09002217,  0.1405638 ,  0.10811667,  0.06866205,  0.00977363,\n",
              "              -0.10805022, -0.19992276, -0.27293718, -0.43117707, -0.59419733,\n",
              "              -0.76461712, -0.74794765, -0.688826  , -0.7479188 , -0.69307873,\n",
              "              -0.64337731, -0.59860487, -0.55701264, -0.51965188, -0.484691  ,\n",
              "              -0.45281837, -0.42313112, -0.3952147 , -0.36971598, -0.34678557,\n",
              "              -0.32581261, -0.30645401, -0.28896947, -0.27381685, -0.26044189,\n",
              "              -0.24843146, -0.23833812, -0.22976656, -0.22251874, -0.21581416,\n",
              "              -0.20985231, -0.20419344, -0.19903725, -0.1941882 , -0.18931931,\n",
              "              -0.18449948, -0.17993604, -0.17552757, -0.17124557, -0.16673351,\n",
              "              -0.16242081, -0.15826282, -0.15423074, -0.15005825, -0.14572583,\n",
              "              -0.14156482, -0.13791705, -0.1341039 , -0.13068599, -0.12717453,\n",
              "              -0.12410758, -0.12109532, -0.11782774, -0.11462951, -0.1117345 ,\n",
              "              -0.10889458, -0.10633181, -0.10383494, -0.10159666, -0.0993525 ,\n",
              "              -0.09724948, -0.09532552, -0.0933708 , -0.09167121, -0.09017945,\n",
              "              -0.08875114, -0.08735862, -0.0860645 , -0.08470226, -0.08337367,\n",
              "              -0.08210151, -0.08102353, -0.07985145, -0.07865262, -0.07745809,\n",
              "              -0.07654957, -0.07560523, -0.07480123, -0.07411402, -0.07345314,\n",
              "              -0.07275129, -0.07199845, -0.0711737 , -0.07034756, -0.06931072,\n",
              "              -0.06821397, -0.06720557, -0.06620814, -0.06538099, -0.06450412,\n",
              "              -0.06361349, -0.06276965, -0.06184333, -0.06118056, -0.06021238,\n",
              "              -0.05942864, -0.05877872, -0.05801912, -0.05731853, -0.05667103,\n",
              "              -0.05597518, -0.05540321, -0.05481984, -0.05426984, -0.05374366,\n",
              "              -0.0533397 , -0.05279593, -0.05238555, -0.05215319, -0.05211236,\n",
              "              -0.05189775, -0.05160779, -0.0514893 , -0.05145564, -0.05135238,\n",
              "              -0.05135877, -0.05122591, -0.0510991 , -0.05100644, -0.05082031,\n",
              "              -0.05061727, -0.05025473, -0.04996137, -0.049705  , -0.04946356,\n",
              "              -0.04923107, -0.04910872, -0.04899736, -0.04884902, -0.04865336,\n",
              "              -0.04850102, -0.04835589, -0.04810244, -0.04780739, -0.04738812,\n",
              "              -0.04696815, -0.04648921, -0.04604634, -0.04557626, -0.04502857,\n",
              "              -0.04459099, -0.04421819, -0.04399353, -0.04368687, -0.04347891,\n",
              "              -0.04327149, -0.04308804, -0.04284464, -0.04266692, -0.04250777,\n",
              "              -0.04226642, -0.04196852, -0.04148378, -0.041053  , -0.04042988,\n",
              "              -0.03973871, -0.03913225, -0.03837989, -0.0377253 , -0.03718964,\n",
              "              -0.03647617, -0.03571407, -0.03502145, -0.03438939, -0.03383072,\n",
              "              -0.03319447, -0.03280477, -0.032499  , -0.0320665 , -0.0317827 ,\n",
              "              -0.03149035, -0.03113967, -0.03063673, -0.0301885 , -0.02960249,\n",
              "              -0.0292876 , -0.02893936, -0.02848404, -0.02799081, -0.02759527,\n",
              "              -0.02726303, -0.02695327, -0.02677982, -0.02648016, -0.02604194,\n",
              "              -0.0259668 , -0.02587288, -0.02564881, -0.02528851, -0.0247778 ,\n",
              "              -0.02429894, -0.02400595, -0.02355537, -0.02309345, -0.02261281,\n",
              "              -0.02211155, -0.02159885, -0.02144943, -0.02119122, -0.02091616,\n",
              "              -0.02046928, -0.02020033, -0.01987413, -0.01954366, -0.01902135,\n",
              "              -0.01873182, -0.01830588, -0.01765878, -0.01741701, -0.01743508,\n",
              "              -0.017454  , -0.01746019, -0.01747997, -0.01745727, -0.01745286,\n",
              "              -0.01748474, -0.0174448 , -0.01742226, -0.01746125, -0.01737651,\n",
              "              -0.01755202, -0.01756114, -0.01765619, -0.01762556, -0.01772137,\n",
              "              -0.017728  , -0.01766009, -0.01735317, -0.01711561, -0.01682977,\n",
              "              -0.0164595 , -0.01587116, -0.01513848, -0.01483477, -0.01460028,\n",
              "              -0.014599  , -0.01460229, -0.01459698, -0.01459718, -0.01459952,\n",
              "              -0.01459782, -0.01459695, -0.01458934, -0.01458755, -0.01458222,\n",
              "              -0.01458534, -0.01456516, -0.01455867, -0.01449244, -0.01447775,\n",
              "              -0.01443843, -0.01441313, -0.01437213, -0.01419997, -0.01415138,\n",
              "              -0.01401458, -0.01403593, -0.01402711, -0.01405156, -0.0140522 ,\n",
              "              -0.01407246, -0.01410215, -0.01414308, -0.01415404, -0.01415696,\n",
              "              -0.01416817, -0.01426356, -0.01428708, -0.01429148, -0.01431673,\n",
              "              -0.01431673, -0.01429834, -0.01432241, -0.01440375, -0.01440907,\n",
              "              -0.01442538, -0.01442103, -0.01442893, -0.01443191, -0.01444721,\n",
              "              -0.01446772, -0.01447398, -0.01447457, -0.01447869, -0.0144749 ,\n",
              "              -0.01446868, -0.01448034, -0.01447446, -0.01448206, -0.01449809,\n",
              "              -0.01449791, -0.01448894, -0.01448879, -0.01450006, -0.01447545,\n",
              "              -0.01447343, -0.01446925, -0.01446561, -0.01446631, -0.01445618,\n",
              "              -0.01445747, -0.01445441, -0.01444181, -0.01444827, -0.01442936,\n",
              "              -0.01443172, -0.01441462, -0.01441956, -0.01441111, -0.01440311,\n",
              "              -0.01441284, -0.01440187, -0.0143712 , -0.01438682, -0.01438363,\n",
              "              -0.01437974, -0.01435535, -0.01434527, -0.01434597, -0.01431702,\n",
              "              -0.01431142, -0.01430062, -0.01430651, -0.0142958 , -0.01428535,\n",
              "              -0.01428753, -0.01428081, -0.01428765, -0.0142795 , -0.01428591,\n",
              "              -0.01427635, -0.0142738 , -0.01426147, -0.01426845, -0.01424714,\n",
              "              -0.01426059, -0.01423802, -0.01422123, -0.01423358, -0.01422157,\n",
              "              -0.01421132, -0.01419912, -0.01418225, -0.0141934 , -0.01417767,\n",
              "              -0.01416249, -0.01412497, -0.01409263, -0.01404713, -0.01399305,\n",
              "              -0.01392955, -0.01387221, -0.01371842, -0.01370495, -0.01358265,\n",
              "              -0.01353646, -0.01339371, -0.01314292, -0.01301494, -0.01287586,\n",
              "              -0.01273156, -0.0123989 , -0.01226844, -0.01219589, -0.01212231,\n",
              "              -0.01203541, -0.01176282, -0.01144476, -0.0112941 , -0.01081368,\n",
              "              -0.01045049, -0.01040056, -0.01042025, -0.01038392, -0.01040903,\n",
              "              -0.01037138, -0.01039524, -0.01039912, -0.01039643, -0.01039078,\n",
              "              -0.01036164, -0.01037183, -0.0104    , -0.01040413, -0.0103953 ,\n",
              "              -0.01039501, -0.01040295, -0.01039824, -0.01042782, -0.01040206,\n",
              "              -0.01040591, -0.01043859, -0.01041364, -0.01044674, -0.01048491,\n",
              "              -0.01047508, -0.01047875, -0.01048244, -0.01047691, -0.01050149,\n",
              "              -0.01051389, -0.01050396, -0.01053675, -0.01048565, -0.01050875,\n",
              "              -0.01052367, -0.01050999, -0.01050379, -0.0105103 , -0.01051186,\n",
              "              -0.01051999, -0.01051779, -0.01051825, -0.01051512, -0.01051044,\n",
              "              -0.01049291, -0.01049603, -0.01048454, -0.01048546, -0.0104527 ,\n",
              "              -0.01050331, -0.01046252, -0.01046616, -0.0104692 , -0.01051206,\n",
              "              -0.01050953, -0.01051206, -0.01050211, -0.01050463, -0.01050652,\n",
              "              -0.01050211, -0.01049179, -0.01047564, -0.01046169, -0.01050728,\n",
              "              -0.01047301, -0.01043451, -0.01042093, -0.01039009, -0.01034633,\n",
              "              -0.01031935, -0.0102789 , -0.01026951, -0.01024875, -0.01022996,\n",
              "              -0.01014751, -0.01012172, -0.01004357, -0.01001514, -0.00996976,\n",
              "              -0.00989401, -0.00984985, -0.0098083 , -0.00979807, -0.00978454,\n",
              "              -0.00980597, -0.00980137, -0.00978305, -0.00980434, -0.00986645,\n",
              "              -0.00988525, -0.00996964, -0.01002669, -0.01012996, -0.01019382,\n",
              "              -0.01025259, -0.01038565, -0.01049284, -0.01075552, -0.0109407 ,\n",
              "              -0.01114428, -0.01135978, -0.01155678, -0.01172994, -0.01178203,\n",
              "              -0.01185198, -0.01195952, -0.01200876, -0.01133036, -0.01082063,\n",
              "              -0.01057941, -0.01025702, -0.00967693, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596, -0.00851596, -0.00851596, -0.00851596,\n",
              "              -0.00851596, -0.00851596])                                      ,\n",
              "       0.37, 816.64, 0.36], dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tf.constant((itemized_train_data))"
      ],
      "metadata": {
        "id": "AxXwfWdQYB45"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "t8-y8Sg9zCEK"
      },
      "source": [
        "### Applying the process on the test data as well"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "id": "ROEiRZCczGCY"
      },
      "outputs": [],
      "source": [
        "scaled_test_data = scaler.transform(np.stack(X_test['padded_stances'].values))\n",
        "scaled_test_data=scaled_test_data.tolist()\n",
        "X_test.insert(0, \"scaled_stances\", scaled_test_data, True)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### First format"
      ],
      "metadata": {
        "id": "-vE3OMWoScJW"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "id": "OGCKXsNZ0tBx"
      },
      "outputs": [],
      "source": [
        "test_data = np.append( np.asarray(X_test['scaled_stances'].tolist()) , np.asarray(X_test[[\"r_swings_time\" ]]),  axis=1)\n",
        "test_data = np.append( test_data , np.asarray(X_test[[\"max_peak\" ]]),  axis=1)\n",
        "test_data = np.append( test_data , np.asarray(X_test[[\"FWHM\" ]]),  axis=1)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(X_test['scaled_stances'][0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4kh4o-rEXsYF",
        "outputId": "fe8bdfb7-2824-42b8-aa43-49a8deda474c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "747"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Second format"
      ],
      "metadata": {
        "id": "7IF5TGIDSeir"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_test['scaled_stances'] = X_test['scaled_stances'].apply(np.array)"
      ],
      "metadata": {
        "id": "mKMpUwPQSlYN"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x_test = X_test[['scaled_stances', 'r_swings_time','max_peak','FWHM']]"
      ],
      "metadata": {
        "id": "FLlBZd9wStrN"
      },
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "itemized_test_data = x_test.to_numpy()"
      ],
      "metadata": {
        "id": "DFm8ZMqUSvYd"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "(itemized_test_data).shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D36H4WwxSwCG",
        "outputId": "59f2895d-ce35-4407-cfa0-b16c04e3018c"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(4597, 4)"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iH7p432Juk6I"
      },
      "source": [
        "## SVC Model\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Applying the SVC model on the first format of the data"
      ],
      "metadata": {
        "id": "TriseUw9XZRO"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D3SMYLucuoJy",
        "outputId": "bbc7d541-7a7a-4f7d-e755-f9bb8826ae94"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/sklearn/utils/validation.py:1143: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ]
        }
      ],
      "source": [
        "svc = SVC(kernel='poly').fit(data, y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jTyvdcjjxfHu",
        "outputId": "1e98e20b-7fef-47e4-f959-a51be4ef9de1"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6756580378507723"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ],
      "source": [
        "svc.score(test_data, (y_test))"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## LSTM"
      ],
      "metadata": {
        "id": "-fVPbbjL4ZC4"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Applying LSTM model on the first format of the data\n"
      ],
      "metadata": {
        "id": "fQWhjzRB6Opg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# define the model architecture\n",
        "model = Sequential()\n",
        "model.add(LSTM(64, input_shape=(750, 1))) # 64 is the number of LSTM units\n",
        "model.add(Dense(32, activation='relu'))\n",
        "model.add(Dense(1, activation='sigmoid'))\n",
        "\n",
        "# compile the model\n",
        "model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "MVqGORqe4bwr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# train the model\n",
        "model.fit(data, y_train, epochs=10, batch_size=32)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Av-UiQCRq1Ei",
        "outputId": "6dab518f-1514-4569-eb89-cc9627ab03e2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "431/431 [==============================] - 138s 313ms/step - loss: 0.6211 - accuracy: 0.6891\n",
            "Epoch 2/10\n",
            "431/431 [==============================] - 152s 352ms/step - loss: 0.6196 - accuracy: 0.6901\n",
            "Epoch 3/10\n",
            "431/431 [==============================] - 134s 310ms/step - loss: 0.6174 - accuracy: 0.6900\n",
            "Epoch 4/10\n",
            "431/431 [==============================] - 132s 306ms/step - loss: 0.6115 - accuracy: 0.6899\n",
            "Epoch 5/10\n",
            "431/431 [==============================] - 134s 311ms/step - loss: 0.6079 - accuracy: 0.6806\n",
            "Epoch 6/10\n",
            "431/431 [==============================] - 132s 307ms/step - loss: 0.6442 - accuracy: 0.6901\n",
            "Epoch 7/10\n",
            "431/431 [==============================] - 132s 307ms/step - loss: 0.6211 - accuracy: 0.6901\n",
            "Epoch 8/10\n",
            "431/431 [==============================] - 133s 309ms/step - loss: 0.6190 - accuracy: 0.6901\n",
            "Epoch 9/10\n",
            "431/431 [==============================] - 132s 307ms/step - loss: 0.6186 - accuracy: 0.6901\n",
            "Epoch 10/10\n",
            "431/431 [==============================] - 132s 307ms/step - loss: 0.6179 - accuracy: 0.6901\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f34c3766b20>"
            ]
          },
          "metadata": {},
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# evaluate the model on the test set\n",
        "loss, accuracy = model.evaluate(test_data, y_test)\n",
        "print(\"Test accuracy: \", accuracy)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xg2a38fZq-W9",
        "outputId": "5b6f4faf-1216-471d-8a27-226f0245b1ad"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "144/144 [==============================] - 14s 97ms/step - loss: 0.6284 - accuracy: 0.6761\n",
            "Test accuracy:  0.6760931015014648\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# evaluate the model on the test set\n",
        "loss, accuracy = model.evaluate(test_data, y_test)\n",
        "print(\"Test accuracy: \", accuracy)"
      ],
      "metadata": {
        "id": "DlS4LeSig5zC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Random Forest\n"
      ],
      "metadata": {
        "id": "5mlKVa62dAr7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Random Forest with 100 estimators"
      ],
      "metadata": {
        "id": "Y52AEp-Rr9fa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Train a Random Forest classifier on the training set\n",
        "rf_classifier = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "rf_classifier.fit(data, y_train.values.ravel())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 74
        },
        "id": "kq4udiavdDoW",
        "outputId": "c0777100-8b19-4121-f602-9a17f0432051"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(random_state=42)"
            ],
            "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>"
            ]
          },
          "metadata": {},
          "execution_count": 75
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf_classifier.predict(test_data)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mLJwmO1pdSmb",
        "outputId": "633c833b-66a6-47bd-d16d-f950e571855b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.9417011094191864\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# save the model\n",
        "with open(\"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Models/rf_classifier\", 'wb') as f:\n",
        "    pickle.dump(rf_classifier, f)"
      ],
      "metadata": {
        "id": "twDQXlTQiWFz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load the model                                                                                                                                                                                                       \n",
        "\n",
        "with open(\"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Models/rf_classifier\", 'rb') as f:\n",
        "    rf = pickle.load(f)"
      ],
      "metadata": {
        "id": "HP7eeIeCiZbC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Checking everything is ok after saving and loading the model\n",
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf.predict(test_data)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9rz3ngVGjSmG",
        "outputId": "55ad8f85-b22a-4d82-cd2f-ca8b507baee2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.9417011094191864\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Random forest with 200 estimators"
      ],
      "metadata": {
        "id": "lG-NZkN_sBPB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Train a Random Forest classifier on the training set\n",
        "rf_classifier_200 = RandomForestClassifier(n_estimators=200, random_state=42,  n_jobs=2)\n",
        "rf_classifier_200.fit(data, y_train.values.ravel())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 74
        },
        "id": "tm6AoHSEsF06",
        "outputId": "146a1494-a424-4235-9a07-0dfaa28e3527"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(n_estimators=200, n_jobs=2)"
            ],
            "text/html": [
              "<style>#sk-container-id-3 {color: black;background-color: white;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 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-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 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-3 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-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 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-3 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-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 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-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 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-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 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-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(n_estimators=200, n_jobs=2)</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-3\" type=\"checkbox\" checked><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(n_estimators=200, n_jobs=2)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf_classifier_200.predict(test_data)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "icXE-BhNsj9B",
        "outputId": "76ebe340-7e45-417c-ae10-63926d77d63d"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.9417011094191864\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Apply hybrid model!"
      ],
      "metadata": {
        "id": "7I3UrLTO32MD"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The approach I am trying to perform here is to use a deep learning model such as a Convolutional Neural Network (CNN) or a Recurrent Neural Network (RNN) to classify the sequences in the first column. \n",
        "\n",
        "The other 3 numeric columns can be concatenated with the output of the CNN (or RNN) and fed into a fully connected neural network for classification. "
      ],
      "metadata": {
        "id": "FzSUDzT65opU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_test_seq = np.reshape( np.asarray(scaled_test_data), (np.asarray(scaled_test_data).shape[0], np.asarray(scaled_test_data).shape[1], 1))\n"
      ],
      "metadata": {
        "id": "DkSm4NUpm4MR"
      },
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Reshape the sequences to be 3-dimensional (samples, timesteps, features)\n",
        "X_train_seq = np.reshape((np.asarray(scaled_data)), ( (np.asarray(scaled_data)).shape[0], (np.asarray(scaled_data)).shape[1], 1))\n",
        "X_test_seq = np.reshape( np.asarray(scaled_test_data), (np.asarray(scaled_test_data).shape[0], np.asarray(scaled_test_data).shape[1], 1))\n",
        "\n",
        "# Define the input shape for the CNN\n",
        "input_shape = (X_train_seq.shape[1], 1)\n",
        "\n",
        "\n",
        "# Define the CNN model\n",
        "cnn_input = Input(shape=input_shape)\n",
        "conv1 = Conv1D(filters=32, kernel_size=3, activation='relu')(cnn_input)\n",
        "pool1 = MaxPooling1D(pool_size=2)(conv1)\n",
        "conv2 = Conv1D(filters=64, kernel_size=3, activation='relu')(pool1)\n",
        "pool2 = MaxPooling1D(pool_size=2)(conv2)\n",
        "flatten = Flatten()(pool2)\n",
        "\n",
        "cnn_output = Dense(128, activation='relu')(flatten)\n",
        "\n",
        "cnn_model = Model(inputs=cnn_input, outputs=cnn_output)\n",
        "\n",
        "# Define the fully connected model\n",
        "\n",
        "fc_input = Input(shape=(itemized_train_data.shape[1]-1,))\n",
        "\n",
        "concat = Concatenate()([cnn_output, fc_input])\n",
        "dense1 = Dense(64, activation='relu')(concat)\n",
        "dropout = Dropout(0.5)(dense1)\n",
        "fc_output = Dense(1, activation='sigmoid')(dropout)\n",
        "fc_model = Model(inputs=[cnn_input, fc_input], outputs=fc_output)\n",
        "\n",
        "# Compile the model\n",
        "optimizer = Adam(learning_rate=0.001)\n",
        "fc_model.compile(optimizer=optimizer, loss='binary_crossentropy', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "owAqJFzN34zF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x = np.asarray(itemized_train_data[:,1:]).astype('float32')"
      ],
      "metadata": {
        "id": "eVn5z2lGdzv1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train the model\n",
        "early_stopping = EarlyStopping(monitor='val_loss', patience=5)\n",
        "history = fc_model.fit([X_train_seq, x], y_train, validation_split=0.2, epochs=50, batch_size=32, callbacks=[early_stopping])"
      ],
      "metadata": {
        "id": "YoVNlQzVZyrl",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "769eaa18-cf3a-47a5-d355-500e8e497cc2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/50\n",
            "345/345 [==============================] - 16s 36ms/step - loss: 4.4989 - accuracy: 0.6666 - val_loss: 0.6245 - val_accuracy: 0.6925\n",
            "Epoch 2/50\n",
            "345/345 [==============================] - 12s 34ms/step - loss: 0.9359 - accuracy: 0.6967 - val_loss: 0.5896 - val_accuracy: 0.6933\n",
            "Epoch 3/50\n",
            "345/345 [==============================] - 12s 36ms/step - loss: 0.6579 - accuracy: 0.6953 - val_loss: 0.6285 - val_accuracy: 0.6875\n",
            "Epoch 4/50\n",
            "345/345 [==============================] - 11s 33ms/step - loss: 0.5937 - accuracy: 0.6963 - val_loss: 0.5629 - val_accuracy: 0.6911\n",
            "Epoch 5/50\n",
            "345/345 [==============================] - 12s 36ms/step - loss: 0.6408 - accuracy: 0.6933 - val_loss: 0.5784 - val_accuracy: 0.6846\n",
            "Epoch 6/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.6113 - accuracy: 0.6916 - val_loss: 0.6196 - val_accuracy: 0.6827\n",
            "Epoch 7/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.5853 - accuracy: 0.6918 - val_loss: 0.6023 - val_accuracy: 0.6827\n",
            "Epoch 8/50\n",
            "345/345 [==============================] - 12s 36ms/step - loss: 0.5801 - accuracy: 0.6918 - val_loss: 0.6015 - val_accuracy: 0.6827\n",
            "Epoch 9/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.6066 - accuracy: 0.6923 - val_loss: 0.5544 - val_accuracy: 0.6846\n",
            "Epoch 10/50\n",
            "345/345 [==============================] - 12s 34ms/step - loss: 0.5740 - accuracy: 0.6925 - val_loss: 0.5729 - val_accuracy: 0.6831\n",
            "Epoch 11/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.5800 - accuracy: 0.6918 - val_loss: 0.5718 - val_accuracy: 0.6831\n",
            "Epoch 12/50\n",
            "345/345 [==============================] - 13s 37ms/step - loss: 0.5710 - accuracy: 0.6921 - val_loss: 0.5495 - val_accuracy: 0.6831\n",
            "Epoch 13/50\n",
            "345/345 [==============================] - 12s 36ms/step - loss: 0.5686 - accuracy: 0.6920 - val_loss: 0.5403 - val_accuracy: 0.6846\n",
            "Epoch 14/50\n",
            "345/345 [==============================] - 14s 40ms/step - loss: 0.5579 - accuracy: 0.6923 - val_loss: 0.5336 - val_accuracy: 0.6831\n",
            "Epoch 15/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.5509 - accuracy: 0.6922 - val_loss: 0.5265 - val_accuracy: 0.6831\n",
            "Epoch 16/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.5453 - accuracy: 0.6919 - val_loss: 0.5431 - val_accuracy: 0.6827\n",
            "Epoch 17/50\n",
            "345/345 [==============================] - 11s 33ms/step - loss: 0.5436 - accuracy: 0.6918 - val_loss: 0.5197 - val_accuracy: 0.6827\n",
            "Epoch 18/50\n",
            "345/345 [==============================] - 12s 33ms/step - loss: 0.5478 - accuracy: 0.6920 - val_loss: 0.5172 - val_accuracy: 0.6831\n",
            "Epoch 19/50\n",
            "345/345 [==============================] - 11s 33ms/step - loss: 0.5354 - accuracy: 0.6918 - val_loss: 0.6554 - val_accuracy: 0.6827\n",
            "Epoch 20/50\n",
            "345/345 [==============================] - 13s 38ms/step - loss: 0.5272 - accuracy: 0.6918 - val_loss: 0.6633 - val_accuracy: 0.6827\n",
            "Epoch 21/50\n",
            "345/345 [==============================] - 17s 49ms/step - loss: 0.5248 - accuracy: 0.6918 - val_loss: 0.6576 - val_accuracy: 0.6827\n",
            "Epoch 22/50\n",
            "345/345 [==============================] - 12s 34ms/step - loss: 0.5749 - accuracy: 0.6918 - val_loss: 0.5142 - val_accuracy: 0.6827\n",
            "Epoch 23/50\n",
            "345/345 [==============================] - 15s 43ms/step - loss: 0.5243 - accuracy: 0.6916 - val_loss: 0.5058 - val_accuracy: 0.6827\n",
            "Epoch 24/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.5150 - accuracy: 0.6918 - val_loss: 0.5011 - val_accuracy: 0.6827\n",
            "Epoch 25/50\n",
            "345/345 [==============================] - 17s 49ms/step - loss: 0.5080 - accuracy: 0.6918 - val_loss: 0.5065 - val_accuracy: 0.6827\n",
            "Epoch 26/50\n",
            "345/345 [==============================] - 13s 37ms/step - loss: 0.4947 - accuracy: 0.6919 - val_loss: 0.4980 - val_accuracy: 0.6831\n",
            "Epoch 27/50\n",
            "345/345 [==============================] - 16s 46ms/step - loss: 0.4870 - accuracy: 0.6920 - val_loss: 0.4908 - val_accuracy: 0.6827\n",
            "Epoch 28/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.4778 - accuracy: 0.7377 - val_loss: 0.4786 - val_accuracy: 0.7770\n",
            "Epoch 29/50\n",
            "345/345 [==============================] - 12s 36ms/step - loss: 0.4657 - accuracy: 0.7598 - val_loss: 0.4873 - val_accuracy: 0.7991\n",
            "Epoch 30/50\n",
            "345/345 [==============================] - 13s 36ms/step - loss: 0.4630 - accuracy: 0.7688 - val_loss: 0.4676 - val_accuracy: 0.8165\n",
            "Epoch 31/50\n",
            "345/345 [==============================] - 17s 48ms/step - loss: 0.4575 - accuracy: 0.7777 - val_loss: 0.4660 - val_accuracy: 0.8281\n",
            "Epoch 32/50\n",
            "345/345 [==============================] - 14s 41ms/step - loss: 0.4454 - accuracy: 0.7817 - val_loss: 0.4732 - val_accuracy: 0.8220\n",
            "Epoch 33/50\n",
            "345/345 [==============================] - 13s 36ms/step - loss: 0.4303 - accuracy: 0.7931 - val_loss: 0.4413 - val_accuracy: 0.8426\n",
            "Epoch 34/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.4368 - accuracy: 0.7970 - val_loss: 0.4518 - val_accuracy: 0.8383\n",
            "Epoch 35/50\n",
            "345/345 [==============================] - 16s 45ms/step - loss: 0.4335 - accuracy: 0.8005 - val_loss: 0.4375 - val_accuracy: 0.8437\n",
            "Epoch 36/50\n",
            "345/345 [==============================] - 12s 36ms/step - loss: 0.4207 - accuracy: 0.8085 - val_loss: 0.4182 - val_accuracy: 0.8557\n",
            "Epoch 37/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.4172 - accuracy: 0.8123 - val_loss: 0.4348 - val_accuracy: 0.8390\n",
            "Epoch 38/50\n",
            "345/345 [==============================] - 13s 38ms/step - loss: 0.4101 - accuracy: 0.8144 - val_loss: 0.3875 - val_accuracy: 0.8626\n",
            "Epoch 39/50\n",
            "345/345 [==============================] - 13s 36ms/step - loss: 0.4119 - accuracy: 0.8114 - val_loss: 0.4048 - val_accuracy: 0.8586\n",
            "Epoch 40/50\n",
            "345/345 [==============================] - 12s 36ms/step - loss: 0.3977 - accuracy: 0.8241 - val_loss: 0.3897 - val_accuracy: 0.8698\n",
            "Epoch 41/50\n",
            "345/345 [==============================] - 13s 38ms/step - loss: 0.3886 - accuracy: 0.8299 - val_loss: 0.4119 - val_accuracy: 0.8586\n",
            "Epoch 42/50\n",
            "345/345 [==============================] - 12s 36ms/step - loss: 0.3960 - accuracy: 0.8248 - val_loss: 0.3838 - val_accuracy: 0.8814\n",
            "Epoch 43/50\n",
            "345/345 [==============================] - 12s 34ms/step - loss: 0.3837 - accuracy: 0.8379 - val_loss: 0.4014 - val_accuracy: 0.8655\n",
            "Epoch 44/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.3798 - accuracy: 0.8413 - val_loss: 0.3744 - val_accuracy: 0.8909\n",
            "Epoch 45/50\n",
            "345/345 [==============================] - 13s 38ms/step - loss: 0.3759 - accuracy: 0.8421 - val_loss: 0.3562 - val_accuracy: 0.8883\n",
            "Epoch 46/50\n",
            "345/345 [==============================] - 13s 38ms/step - loss: 0.3606 - accuracy: 0.8490 - val_loss: 0.3649 - val_accuracy: 0.8854\n",
            "Epoch 47/50\n",
            "345/345 [==============================] - 12s 35ms/step - loss: 0.3622 - accuracy: 0.8451 - val_loss: 0.3434 - val_accuracy: 0.8916\n",
            "Epoch 48/50\n",
            "345/345 [==============================] - 12s 34ms/step - loss: 0.3526 - accuracy: 0.8568 - val_loss: 0.3496 - val_accuracy: 0.8930\n",
            "Epoch 49/50\n",
            "345/345 [==============================] - 11s 33ms/step - loss: 0.3419 - accuracy: 0.8626 - val_loss: 0.3339 - val_accuracy: 0.9028\n",
            "Epoch 50/50\n",
            "345/345 [==============================] - 11s 33ms/step - loss: 0.3252 - accuracy: 0.8740 - val_loss: 0.3211 - val_accuracy: 0.9014\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the model on the testing set\n",
        "#X_test_seq = np.expand_dims(X_test.iloc[:,0].values, axis=2)\n",
        "x_test_features = np.asarray(itemized_test_data[:,1:]).astype('float32')\n",
        "accuracy = fc_model.evaluate([X_test_seq, x_test_features], y_test)[1]\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "id": "Q1IJfxmjZ0ad",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a5bf1d5e-420a-4d93-9685-383bf3e26d37"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "144/144 [==============================] - 2s 11ms/step - loss: 0.4425 - accuracy: 0.8936\n",
            "Accuracy: 0.8936262726783752\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(fc_model.summary())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "loA7UTO7e2el",
        "outputId": "c8eeb9a4-f494-4ba7-b420-8d764660c9ac"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_3\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_3 (InputLayer)           [(None, 747, 1)]     0           []                               \n",
            "                                                                                                  \n",
            " conv1d_5 (Conv1D)              (None, 745, 32)      128         ['input_3[0][0]']                \n",
            "                                                                                                  \n",
            " max_pooling1d_5 (MaxPooling1D)  (None, 372, 32)     0           ['conv1d_5[0][0]']               \n",
            "                                                                                                  \n",
            " conv1d_6 (Conv1D)              (None, 370, 64)      6208        ['max_pooling1d_5[0][0]']        \n",
            "                                                                                                  \n",
            " max_pooling1d_6 (MaxPooling1D)  (None, 185, 64)     0           ['conv1d_6[0][0]']               \n",
            "                                                                                                  \n",
            " flatten_4 (Flatten)            (None, 11840)        0           ['max_pooling1d_6[0][0]']        \n",
            "                                                                                                  \n",
            " dense_16 (Dense)               (None, 128)          1515648     ['flatten_4[0][0]']              \n",
            "                                                                                                  \n",
            " input_4 (InputLayer)           [(None, 3)]          0           []                               \n",
            "                                                                                                  \n",
            " concatenate_1 (Concatenate)    (None, 131)          0           ['dense_16[0][0]',               \n",
            "                                                                  'input_4[0][0]']                \n",
            "                                                                                                  \n",
            " dense_17 (Dense)               (None, 64)           8448        ['concatenate_1[0][0]']          \n",
            "                                                                                                  \n",
            " dropout_1 (Dropout)            (None, 64)           0           ['dense_17[0][0]']               \n",
            "                                                                                                  \n",
            " dense_18 (Dense)               (None, 1)            65          ['dropout_1[0][0]']              \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 1,530,497\n",
            "Trainable params: 1,530,497\n",
            "Non-trainable params: 0\n",
            "__________________________________________________________________________________________________\n",
            "None\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_model(fc_model, to_file=\"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Models/combined_model.png\", show_shapes=True, show_layer_names=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "ui727u5Be3l_",
        "outputId": "ae848295-0933-4b57-f782-bdc0c95f5685"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 74
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# save the model\n",
        "fc_model.save(\"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Models/combined_model\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XnsaOlate49p",
        "outputId": "6409ad79-0cde-452a-d54d-cb2aae05c70a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 2 of 2). These functions will not be directly callable after loading.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# load the model\n",
        "loaded_fc_model = keras.models.load_model(\"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Models/combined_model\")\n"
      ],
      "metadata": {
        "id": "XM4uVOw5i9co"
      },
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# check the loaded model\n",
        "accuracy = loaded_fc_model.evaluate([X_test_seq, x_test_features], y_test)[1]\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KAlYTtpEfFh-",
        "outputId": "de39fef0-6015-4519-adeb-ddbfd1048580"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "144/144 [==============================] - 2s 13ms/step - loss: 0.4425 - accuracy: 0.8936\n",
            "Accuracy: 0.8936262726783752\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Apply Foourier transformation then RF (or SVM)"
      ],
      "metadata": {
        "id": "cfMO4_2F35P-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Here I am trying to use a traditional machine learning algorithm (Random Forest or Support Vector Machines (SVM)) to classify the data. In this case, the time series sequence in the first column can be transformed into features using techniques such as Fourier Transform or Wavelet Transform."
      ],
      "metadata": {
        "id": "OKXiOjD7sYvs"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Applying Fourier transformation on scaled stances"
      ],
      "metadata": {
        "id": "zhm5bpQfxO-B"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Define a function to apply Fourier Transform to the sequence data\n",
        "def apply_fourier_transform(data):\n",
        "    transformed_data = []\n",
        "    for row in data:\n",
        "        transformed_row = np.abs(fft(row))\n",
        "        transformed_data.append(transformed_row)\n",
        "    return np.array(transformed_data)"
      ],
      "metadata": {
        "id": "w72R1_vkuYog"
      },
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Apply Fourier Transform to the sequence data in the first column\n",
        "X_train_seq = apply_fourier_transform(scaled_data)\n",
        "X_test_seq = apply_fourier_transform(scaled_test_data)\n"
      ],
      "metadata": {
        "id": "GId5Ios7sYiP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Concatenate the transformed sequence data with the other 3 numeric columns\n",
        "X_train_concat = np.concatenate((X_train_seq, itemized_train_data[:,1:]), axis=1)\n",
        "X_test_concat = np.concatenate((X_test_seq, itemized_test_data[:,1:]), axis=1)"
      ],
      "metadata": {
        "id": "SOsCyRaz3-nR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train a Random Forest classifier on the concatenated data\n",
        "rf_classifier = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "rf_classifier.fit(X_train_concat, y_train.values.ravel())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "hg4LtGasvOs4",
        "outputId": "04e98179-f9bb-430f-f2fb-0c7b8e236653"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(random_state=42)"
            ],
            "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>"
            ]
          },
          "metadata": {},
          "execution_count": 86
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf_classifier.predict(X_test_concat)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-yt7DgrTvbaO",
        "outputId": "d9a128b9-8ad0-4dcc-a30b-7db614ca1cfd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.8696976288884055\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Applying Fourier transformation on padded stances"
      ],
      "metadata": {
        "id": "g_1dBp4PwGoZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Apply Fourier Transform to the sequence data in the first column\n",
        "X_train_seq = apply_fourier_transform(X_train['padded_stances'].values)\n",
        "X_test_seq = apply_fourier_transform(X_test['padded_stances'].values)\n"
      ],
      "metadata": {
        "id": "_g2fWlG6wG74"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Concatenate the transformed sequence data with the other 3 numeric columns\n",
        "X_train_concat = np.concatenate((X_train_seq, itemized_train_data[:,1:]), axis=1)\n",
        "X_test_concat = np.concatenate((X_test_seq, itemized_test_data[:,1:]), axis=1)"
      ],
      "metadata": {
        "id": "7NYJuqJKwV20"
      },
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "RF with 100 estimators"
      ],
      "metadata": {
        "id": "JfxcD30fttSC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Train a Random Forest classifier on the concatenated data\n",
        "rf_classifier = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "rf_classifier.fit(X_train_concat, y_train.values.ravel())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "JK6YxOUTwgeB",
        "outputId": "1e222698-e849-4ab1-9467-272e12c5bc50"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(random_state=42)"
            ],
            "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>"
            ]
          },
          "metadata": {},
          "execution_count": 95
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf_classifier.predict(X_test_concat)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a-AeuM86whvM",
        "outputId": "ab019e61-c70d-494d-88c1-70c3677d697d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.9169023276049597\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# save the model\n",
        "with open(\"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Models/rf_classifier_Fourier\", 'wb') as f:\n",
        "    pickle.dump(rf_classifier, f)"
      ],
      "metadata": {
        "id": "XhZRxakHw-AY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load the model                                                                                                                                                                                                       \n",
        "with open(\"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Models/rf_classifier_Fourier\", 'rb') as f:\n",
        "    rf = pickle.load(f)"
      ],
      "metadata": {
        "id": "8O3JuRNtw_AY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### RF with 200 estimators"
      ],
      "metadata": {
        "id": "iKaR9vpRt0hz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Train a Random Forest classifier on the concatenated data\n",
        "rf_classifier = RandomForestClassifier(n_estimators=200,  n_jobs=2, random_state=42)\n",
        "rf_classifier.fit(X_train_concat, y_train.values.ravel())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 74
        },
        "id": "4rC9I3zpt2mi",
        "outputId": "36cb8e3b-1648-4c33-aa52-99dab87687f7"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(n_estimators=200, n_jobs=2, random_state=42)"
            ],
            "text/html": [
              "<style>#sk-container-id-4 {color: black;background-color: white;}#sk-container-id-4 pre{padding: 0;}#sk-container-id-4 div.sk-toggleable {background-color: white;}#sk-container-id-4 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-4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 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-4 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-4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 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-4 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-4 div.sk-item {position: relative;z-index: 1;}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-4 div.sk-item::before, #sk-container-id-4 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-4 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-4 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-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-4 div.sk-label-container {text-align: center;}#sk-container-id-4 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-4 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-4\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(n_estimators=200, n_jobs=2, 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-4\" type=\"checkbox\" checked><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(n_estimators=200, n_jobs=2, random_state=42)</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf_classifier.predict(X_test_concat)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G-Yzof-Dt_TM",
        "outputId": "741563d5-eea8-4879-cbc4-e71fa3f87907"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.9190776593430499\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Apply RF on the transformed stances by Fourier without the additional three numeric features"
      ],
      "metadata": {
        "id": "Ah-yETBL4o1a"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Apply Fourier Transform to the sequence data in the first column\n",
        "X_train_seq = apply_fourier_transform(X_train['padded_stances'].values)\n",
        "X_test_seq = apply_fourier_transform(X_test['padded_stances'].values)\n"
      ],
      "metadata": {
        "id": "97c23qYf4pTj"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train a Random Forest classifier on the concatenated data\n",
        "rf_classifier = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "rf_classifier.fit(X_train_seq, y_train.values.ravel())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "ELsDMOoB4sIZ",
        "outputId": "0f2b63cb-a3e1-4eb4-9631-47a848ea1e1b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(random_state=42)"
            ],
            "text/html": [
              "<style>#sk-container-id-3 {color: black;background-color: white;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 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-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 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-3 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-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 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-3 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-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 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-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 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-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 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-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" 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-3\" type=\"checkbox\" checked><label for=\"sk-estimator-id-3\" 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>"
            ]
          },
          "metadata": {},
          "execution_count": 157
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf_classifier.predict(X_test_seq)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xsa-CW_F4tsT",
        "outputId": "ce619d9b-19ff-4d5b-ae05-e355235206f7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.9110289319121166\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# save the model\n",
        "with open(\"/content/drive/MyDrive/ITMO-Master's/Thesis/3rd_semester/Models/rf_classifier_Fourier_only_stances\", 'wb') as f:\n",
        "    pickle.dump(rf_classifier, f)"
      ],
      "metadata": {
        "id": "NU4qLx1U4_wc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Dealing with total sequence data instead of the stances"
      ],
      "metadata": {
        "id": "SDErnaIzOEd1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Right stances only"
      ],
      "metadata": {
        "id": "3rJc2FtgUEym"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#x_signal_train = (data.signals_data_df[\"Total_Force_Right\"].values)\n",
        "y_signal = (data.signals_data_df[\"y\"].values)"
      ],
      "metadata": {
        "id": "AODtWxT_ORut"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "## padding\n",
        "max_length = max([len(row) for row in data.signals_data_df[\"Total_Force_Right\"]])\n",
        "padded_signal_data = np.zeros((len(data.signals_data_df[\"Total_Force_Right\"]), max_length))\n",
        "for i, row in enumerate(data.signals_data_df[\"Total_Force_Right\"]):\n",
        "    padded_signal_data[i, :len(row)] = row"
      ],
      "metadata": {
        "id": "TWn3OAXeSm3I"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(padded_signal_data, y_signal, random_state=42)"
      ],
      "metadata": {
        "id": "0PMZ1qcqP25C"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Apply Fourier Transform to the sequence data in the first column\n",
        "X_train_seq = apply_fourier_transform(X_train)\n",
        "X_test_seq = apply_fourier_transform(X_test)\n"
      ],
      "metadata": {
        "id": "MNRqVui9O7yv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6VHEglKAjzGK",
        "outputId": "93421917-b9cf-4a42-fd79-f39ee6bfd0cd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(228, 22366)"
            ]
          },
          "metadata": {},
          "execution_count": 383
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Train a Random Forest classifier on the concatenated data\n",
        "rf_classifier = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "rf_classifier.fit(X_train_seq, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "rS_YQhgpPTLp",
        "outputId": "37fcfbbe-227a-4b5e-b59f-a4b62aedcf8d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(random_state=42)"
            ],
            "text/html": [
              "<style>#sk-container-id-4 {color: black;background-color: white;}#sk-container-id-4 pre{padding: 0;}#sk-container-id-4 div.sk-toggleable {background-color: white;}#sk-container-id-4 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-4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 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-4 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-4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 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-4 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-4 div.sk-item {position: relative;z-index: 1;}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-4 div.sk-item::before, #sk-container-id-4 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-4 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-4 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-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-4 div.sk-label-container {text-align: center;}#sk-container-id-4 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-4 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-4\" 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-4\" type=\"checkbox\" checked><label for=\"sk-estimator-id-4\" 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>"
            ]
          },
          "metadata": {},
          "execution_count": 190
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf_classifier.predict(X_test_seq)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "id": "mUMqeMPWPWpu",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2434aa03-a6d2-422a-90af-7405ab0db861"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.75\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Right and left stances"
      ],
      "metadata": {
        "id": "QiqhVkY9UIVv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "right = data.signals_data_df[\"Total_Force_Right\"].to_numpy().reshape(data.signals_data_df[\"Total_Force_Right\"].shape[0],1)\n",
        "left = data.signals_data_df[\"Total_Force_Left\"].to_numpy().reshape(data.signals_data_df[\"Total_Force_Left\"].shape[0],1)\n",
        "y_signal = (data.signals_data_df[\"y\"].values)"
      ],
      "metadata": {
        "id": "yk117wg2Wb5t"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "both_signals =[]\n",
        "for i in range(right.shape[0]):\n",
        "  both_signals.append(np.append(right[i][0],left[i][0]))"
      ],
      "metadata": {
        "id": "KXhRCjyBiIFu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### padding"
      ],
      "metadata": {
        "id": "svZuBDx8lP_i"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "max_length = max([len(row) for row in both_signals])\n",
        "padded_data = np.zeros((len(both_signals), max_length))\n",
        "for i, row in enumerate(both_signals):\n",
        "    padded_data[i, :len(row)] = row"
      ],
      "metadata": {
        "id": "ESLYiijklZ19"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(padded_data, y_signal, random_state=42)"
      ],
      "metadata": {
        "id": "Sgwur_lYh56s"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Apply Fourier Transform to the sequence data in the first column\n",
        "X_train_seq = apply_fourier_transform(X_train)\n",
        "X_test_seq = apply_fourier_transform(X_test)"
      ],
      "metadata": {
        "id": "sjeqVreGi-cK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train a Random Forest classifier on the concatenated data\n",
        "rf_classifier = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "rf_classifier.fit(X_train_seq, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 75
        },
        "id": "2QJB6XJii_Na",
        "outputId": "acf7a594-92b6-4374-f6eb-b398d33ab65e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(random_state=42)"
            ],
            "text/html": [
              "<style>#sk-container-id-6 {color: black;background-color: white;}#sk-container-id-6 pre{padding: 0;}#sk-container-id-6 div.sk-toggleable {background-color: white;}#sk-container-id-6 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-6 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-6 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-6 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-6 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-6 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-6 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-6 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-6 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-6 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-6 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-6 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-6 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-6 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-6 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-6 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-6 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-6 div.sk-item {position: relative;z-index: 1;}#sk-container-id-6 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-6 div.sk-item::before, #sk-container-id-6 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-6 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-6 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-6 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-6 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-6 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-6 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-6 div.sk-label-container {text-align: center;}#sk-container-id-6 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-6 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-6\" 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-6\" type=\"checkbox\" checked><label for=\"sk-estimator-id-6\" 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>"
            ]
          },
          "metadata": {},
          "execution_count": 478
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the performance of the classifier on the testing set\n",
        "y_pred = rf_classifier.predict(X_test_seq)\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "print(f'Accuracy: {accuracy}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hnPGx2u2l8RT",
        "outputId": "369f3682-f1b5-4eb1-e908-9147626aec37"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy: 0.7763157894736842\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### All data from all sensors"
      ],
      "metadata": {
        "id": "aVa7tEBxp4JM"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "-ovxRcw9p6d8"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [
        "xo1CVjZb-oDD",
        "yoPlShO_1mD5",
        "VLlmkpXwQPt0",
        "H_sk20fPzLuN",
        "6gvkVqJ2zQFA",
        "gmQ47Qcp6D5S",
        "0_Wh3g6eut06",
        "Zx3AE4Ab0bRL",
        "5njlOxC_gRAB",
        "evtbYEBeOb7q",
        "iH7p432Juk6I",
        "-fVPbbjL4ZC4",
        "zhm5bpQfxO-B",
        "Ah-yETBL4o1a",
        "SDErnaIzOEd1",
        "3rJc2FtgUEym"
      ],
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "009a2cc28ae84588a4287bf39386769d": {
          "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": ""
          }
        },
        "0cc9fd76f4c34854b31e0409a13d62f3": {
          "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
          }
        },
        "29ebc352951d4fa0938beba218a394b8": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_bff29d2b3d834c9ca2263759ef64a52e",
            "placeholder": "​",
            "style": "IPY_MODEL_009a2cc28ae84588a4287bf39386769d",
            "value": "100%"
          }
        },
        "2fabca6f213f46d5ac3f834242ebefa1": {
          "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
          }
        },
        "549e36afb1344e5dac211583191e8bdf": {
          "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
          }
        },
        "76d430b27faf43bdadf157f552f13ae2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_29ebc352951d4fa0938beba218a394b8",
              "IPY_MODEL_b6828ee43f97471abf45ccc60c277629",
              "IPY_MODEL_ef49c0063f0244d2aeb3a2d9cab4ef4d"
            ],
            "layout": "IPY_MODEL_549e36afb1344e5dac211583191e8bdf"
          }
        },
        "7c27d950cf8e4efebf2699e943a9e7d8": {
          "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": ""
          }
        },
        "884b0b1abc344bf2bdb4e38e6fc2fdbd": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "b6828ee43f97471abf45ccc60c277629": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2fabca6f213f46d5ac3f834242ebefa1",
            "max": 304,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_884b0b1abc344bf2bdb4e38e6fc2fdbd",
            "value": 304
          }
        },
        "bff29d2b3d834c9ca2263759ef64a52e": {
          "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
          }
        },
        "ef49c0063f0244d2aeb3a2d9cab4ef4d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0cc9fd76f4c34854b31e0409a13d62f3",
            "placeholder": "​",
            "style": "IPY_MODEL_7c27d950cf8e4efebf2699e943a9e7d8",
            "value": " 304/304 [00:52&lt;00:00,  8.92it/s]"
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}