opportunity.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Practical session 1\n",
    ">  - María Castro\n",
    ">  - Jorge Martín\n",
    ">  - Miguel Angel Rojas\n",
    "\n",
    "\n",
    "## Import library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.stats import mode\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import classification_report, confusion_matrix, accuracy_score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#separate train and test train_test_split function' \n",
    "def train_test_split_local(X, y):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)\n",
    "    train_test_split(y, shuffle=False)\n",
    "    return X_train, X_test, y_train, y_test\n",
    "\n",
    "def read(file_name, fheader, fuser, ftrial):\n",
    "    fsadl1 = pd.read_csv(file_name, sep=' ', header=None)\n",
    "    fdata = fsadl1.iloc[:, :243]\n",
    "    fdata.columns = fheader\n",
    "    fdata=fdata[fdata.columns[np.r_[0:45,50:58,63:71,76:84,89:97,102:133]]]\n",
    "    flabels = fsadl1.iloc[:,243]\n",
    "    ## Preprocessing data\n",
    "    #find and remove rows with all nulls\n",
    "    fidx=fdata.index[fdata.isnull().all(1)] #1 is the axis for rows\n",
    "    #select data not in idx, that is data that is not all null\n",
    "    fdata = fdata[~fdata.index.isin(fidx)] \n",
    "    #same for labels\n",
    "    flabels = flabels[~flabels.index.isin(fidx)]\n",
    "\n",
    "    #see how many there are of each label\n",
    "    #what does it mean ?\n",
    "    flabels.value_counts()\n",
    "\n",
    "\n",
    "    #fill missing values\n",
    "    fdata = fdata.fillna(method='ffill',axis=1)\n",
    "\n",
    "\n",
    "    fdata['user'] = fuser\n",
    "    fdata['trial'] = ftrial\n",
    "    return fdata, flabels\n",
    "    \n",
    "def windowing(fdata, window_number, window_text, porcentage, flabels, frol):\n",
    "    ffiltered_data = fdata[columns].rolling(frol).median()\n",
    "    ffiltered_data['MILLISEC'] = fdata.MILLISEC\n",
    "\n",
    "\n",
    "    # Windowing and Feature Extraction\n",
    "    ffiltered_data['time']=pd.to_datetime(fdata.MILLISEC,unit='ms')\n",
    "    ffiltered_data.index=ffiltered_data.time\n",
    "    #calculate mean over a 1 second window\n",
    "    keep = ffiltered_data.time.dt.microsecond/window_number %porcentage\n",
    "    keep = keep - keep.shift() < 0\n",
    "\n",
    "    means = ffiltered_data[columns].rolling(window_text).mean()[keep]\n",
    "    means.columns = [str(col) + '_mean' for col in means.columns]\n",
    "    variances = ffiltered_data[columns].rolling(window_text).var()[keep]\n",
    "    variances.columns = [str(col) + '_var' for col in variances.columns]\n",
    "\n",
    "    #talk about apply function\n",
    "    flabels.index = ffiltered_data.time\n",
    "    mode_labels = flabels.rolling(window_text).apply(lambda x:mode(x)[0])[keep]\n",
    "\n",
    "    #all features\n",
    "    fall_features = pd.concat([means, variances],axis=1)\n",
    "    fall_features['label'] = mode_labels\n",
    "    fall_features['user'] = user\n",
    "    fall_features['trial'] = trial\n",
    "    \n",
    "    return fall_features\n",
    "\n",
    "def plot_confusion_matrix(cm, names, title='MATRIZ DE CONFUSIÓN', cmap=plt.cm.Blues):\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(names))\n",
    "    plt.xticks(tick_marks, names, rotation=45)\n",
    "    plt.yticks(tick_marks, names)\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('Clase real')\n",
    "    plt.xlabel('Clase predicha')\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Execises definitions and clasifiers\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def excercise_1 (fall_data, estimators):\n",
    "    # Exercise 1\n",
    "    # Random train test split\n",
    "\n",
    "    # Labels are the values we want to predict\n",
    "    labels = np.array(fall_data['label'])\n",
    "\n",
    "    # Remove the labels from the features\n",
    "    # axis 1 refers to the columns\n",
    "    features = fall_data.drop('label', axis = 1)\n",
    "    features = features.drop('user', axis = 1)\n",
    "    features = features.drop('trial', axis = 1)\n",
    "\n",
    "\n",
    "    # Saving feature names for later use\n",
    "    feature_list = list(features.columns)\n",
    "\n",
    "    # Convert to numpy array\n",
    "    features = np.array(features)\n",
    "\n",
    "\n",
    "    X_train, X_test, y_train, y_test = train_test_split_local(features, labels)\n",
    "\n",
    "    print('X_train:', X_train.shape)\n",
    "    print('X_test:', X_test.shape)\n",
    "    print('y_train:', y_train.shape)\n",
    "    print('y_test:', y_test.shape)\n",
    "\n",
    "\n",
    "    classifier = RandomForestClassifier(n_estimators=estimators, random_state=0)  \n",
    "    classifier.fit(X_train, y_train)  \n",
    "    y_pred = classifier.predict(X_test)\n",
    "    \n",
    "    labels = [0,1,2,4,5]\n",
    "\n",
    "    print(\"\\n _______________________________________________________\")  \n",
    "    print(\"confusion matrix: \\n\")  \n",
    "    print(confusion_matrix(y_test,y_pred))  \n",
    "    print(\"\\n _______________________________________________________\")  \n",
    "    print(\"classification report: \\n\")  \n",
    "    print(classification_report(y_test,y_pred))  \n",
    "    print(\"_______________________________________________________\")  \n",
    "    print(\"accuracy score: \" + str(accuracy_score(y_test, y_pred)))  \n",
    "\n",
    "    \n",
    "    plt.figure(figsize=(15,8))\n",
    "    plot_confusion_matrix(confusion_matrix(y_test,y_pred), labels)\n",
    "\n",
    "def excercise_2 (fall_data, estimators):\n",
    "    # Excercise 2\n",
    "    # user 1 2 3 train test 4\n",
    "\n",
    "    # Labels are the values we want to predict\n",
    "    # Remove the labels from the features\n",
    "    # axis 1 refers to the columns\n",
    "\n",
    "    user123 = fall_data[fall_data['user'].isin([1, 2, 3])]\n",
    "    user4 = fall_data[fall_data['user'].isin([4])]\n",
    "\n",
    "\n",
    "    y_train_2 = np.array(user123['label'])\n",
    "    y_test_2 = np.array(user4['label'])\n",
    "\n",
    "    user123 = user123.drop('label', axis = 1)\n",
    "    user123 = user123.drop('user', axis = 1)\n",
    "    user123 = user123.drop('trial', axis = 1)\n",
    "\n",
    "    user4 = user4.drop('label', axis = 1)\n",
    "    user4 = user4.drop('user', axis = 1)\n",
    "    user4 = user4.drop('trial', axis = 1)\n",
    "\n",
    "    X_train_2 = np.array(user123)\n",
    "    X_test_2 = np.array(user4)\n",
    "    print('X_train_2:', X_train_2.shape)\n",
    "    print('X_test_2:', X_test_2.shape)\n",
    "    print('y_train_2:', y_train_2.shape)\n",
    "    print('y_test_2:', y_test_2.shape)\n",
    "\n",
    "\n",
    "    classifier2 = RandomForestClassifier(n_estimators=estimators, random_state=0)  \n",
    "    classifier2.fit(X_train_2, y_train_2)  \n",
    "    y_pred_2 = classifier2.predict(X_test_2)\n",
    "\n",
    "    print(\"\\n _______________________________________________________\")  \n",
    "    print(\"confusion matrix: \\n\")  \n",
    "    print(confusion_matrix(y_test_2,y_pred_2))  \n",
    "    print(\"\\n _______________________________________________________\")  \n",
    "    print(\"classification report: \\n\")  \n",
    "    print(classification_report(y_test_2,y_pred_2))  \n",
    "    print(\"_______________________________________________________\")  \n",
    "    print(\"accuracy score: \" +str(accuracy_score(y_test_2, y_pred_2)))\n",
    "\n",
    "    labels = [0,1,2,4,5]\n",
    "    plt.figure(figsize=(15,8))\n",
    "    plot_confusion_matrix(confusion_matrix(y_test_2,y_pred_2), labels)\n",
    "\n",
    "def excercise_3 (fall_data, estimators):  \n",
    "    # Excercise 3\n",
    "    #  1,2,3 and drill session as training data and trials 4 and 5 as test data.\n",
    "\n",
    "    # Labels are the values we want to predict\n",
    "    # Remove the labels from the features\n",
    "    # axis 1 refers to the columns\n",
    "\n",
    "    trial1236 = fall_data[fall_data['trial'].isin([1, 2, 3, 6])]\n",
    "    trial45 = fall_data[fall_data['trial'].isin([4, 5])]\n",
    "\n",
    "    y_train_3 = np.array(trial1236['label'])\n",
    "    y_test_3 = np.array(trial45['label'])\n",
    "\n",
    "    trial1236 = trial1236.drop('label', axis = 1)\n",
    "    trial1236 = trial1236.drop('user', axis = 1)\n",
    "    trial1236 = trial1236.drop('trial', axis = 1)\n",
    "\n",
    "    trial45 = trial45.drop('label', axis = 1)\n",
    "    trial45 = trial45.drop('user', axis = 1)\n",
    "    trial45 = trial45.drop('trial', axis = 1)\n",
    "\n",
    "    X_train_3 = np.array(trial1236)\n",
    "    X_test_3 = np.array(trial45)\n",
    "\n",
    "    print('X_train_3:', X_train_3.shape)\n",
    "    print('X_test_3:', X_test_3.shape)\n",
    "    print('y_train_3:', y_train_3.shape)\n",
    "    print('y_test_3:', y_test_3.shape)\n",
    "\n",
    "\n",
    "    classifier = RandomForestClassifier(n_estimators=estimators, random_state=0)  \n",
    "    classifier.fit(X_train_3, y_train_3)  \n",
    "    y_pred_3 = classifier.predict(X_test_3)\n",
    "\n",
    "\n",
    "    print(\"\\n _______________________________________________________\")  \n",
    "    print(\"confusion matrix: \\n\")  \n",
    "    print(confusion_matrix(y_test_3,y_pred_3))  \n",
    "    print(\"\\n _______________________________________________________\")  \n",
    "    print(\"classification report: \\n\")  \n",
    "    print(classification_report(y_test_3,y_pred_3))  \n",
    "    print(\"_______________________________________________________\")  \n",
    "    print(\"accuracy score: \" + str(accuracy_score(y_test_3, y_pred_3)))\n",
    "\n",
    "    labels = [0,1,2,4,5]\n",
    "    plt.figure(figsize=(15,8))\n",
    "    plot_confusion_matrix(confusion_matrix(y_test_3,y_pred_3), labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "path = 'dataset/' #enter thepath for the dataset folder\n",
    "header_path = 'header.csv' #enter the path for the header file\n",
    "header=pd.read_csv(header_path,names=['column',''])['column'].values\n",
    "users = range(1,2)\n",
    "trials = range(1,2)\n",
    "all_data_1S = pd.DataFrame()\n",
    "all_data_2S = pd.DataFrame()\n",
    "all_data_5S = pd.DataFrame()\n",
    "all_data_10S = pd.DataFrame()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/yure/Documents/UN/Machine Learning/ml/lib/python3.7/site-packages/ipykernel_launcher.py:53: FutureWarning: Currently, 'apply' passes the values as ndarrays to the applied function. In the future, this will change to passing it as Series objects. You need to specify 'raw=True' to keep the current behaviour, and you can pass 'raw=False' to silence this warning\n"
     ]
    }
   ],
   "source": [
    "for user in users:\n",
    "    for trial in trials:\n",
    "        if trial == 6:\n",
    "            file_name = path+'S'+str(user)+'-Drill'+'.dat'\n",
    "        else:\n",
    "            file_name = path+'S'+str(user)+'-ADL'+str(trial)+'.dat'\n",
    "        data, labels = read(file_name, header, user, trial)\n",
    "        columns = data.columns[~data.columns.isin(['user', 'trial','MILLISEC'])]\n",
    "        #we use a window of 11 elements\n",
    "        # Filtering using median filter\n",
    "        \n",
    "        all_features_1S = windowing(data, 1000, '1S', 500, labels, 11)\n",
    "        all_data_1S = pd.concat([all_data_1S, all_features_1S])\n",
    "        \n",
    "        all_features_2S = windowing(data, 2000, '2S', 1000, labels, 11)\n",
    "        all_data_2S = pd.concat([all_data_2S, all_features_2S])\n",
    "        \n",
    "        all_features_5S = windowing(data, 5000, '5S', 2500, labels, 11)\n",
    "        all_data_5S = pd.concat([all_data_5S, all_features_5S])\n",
    "        \n",
    "        all_features_10S = windowing(data, 10000, '10S', 5000, labels, 11)\n",
    "        all_data_10S = pd.concat([all_data_10S, all_features_10S])\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory data analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Accelerometer_RKN^_accX_mean</th>\n",
       "      <th>Accelerometer_RKN^_accY_mean</th>\n",
       "      <th>Accelerometer_RKN^_accZ_mean</th>\n",
       "      <th>Accelerometer_HIP_accX_mean</th>\n",
       "      <th>Accelerometer_HIP_accY_mean</th>\n",
       "      <th>Accelerometer_HIP_accZ_mean</th>\n",
       "      <th>Accelerometer_LUA^_accX_mean</th>\n",
       "      <th>Accelerometer_LUA^_accY_mean</th>\n",
       "      <th>Accelerometer_LUA^_accZ_mean</th>\n",
       "      <th>Accelerometer_RUA__accX_mean</th>\n",
       "      <th>...</th>\n",
       "      <th>InertialMeasurementUnit_R-SHOE_Body_Az_var</th>\n",
       "      <th>InertialMeasurementUnit_R-SHOE_AngVelBody_FrameX_var</th>\n",
       "      <th>InertialMeasurementUnit_R-SHOE_AngVelBody_FrameY_var</th>\n",
       "      <th>InertialMeasurementUnit_R-SHOE_AngVelBody_FrameZ_var</th>\n",
       "      <th>InertialMeasurementUnit_R-SHOE_AngVelNav_FrameX_var</th>\n",
       "      <th>InertialMeasurementUnit_R-SHOE_AngVelNav_FrameY_var</th>\n",
       "      <th>InertialMeasurementUnit_R-SHOE_AngVelNav_FrameZ_var</th>\n",
       "      <th>label</th>\n",
       "      <th>user</th>\n",
       "      <th>trial</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1970-01-01 00:00:00.500</th>\n",
       "      <td>105.000000</td>\n",
       "      <td>961.666667</td>\n",
       "      <td>-357.333333</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1023.166667</td>\n",
       "      <td>183.500000</td>\n",
       "      <td>74.666667</td>\n",
       "      <td>1008.666667</td>\n",
       "      <td>101.833333</td>\n",
       "      <td>245.333333</td>\n",
       "      <td>...</td>\n",
       "      <td>1.366667</td>\n",
       "      <td>21.766667</td>\n",
       "      <td>14.266667</td>\n",
       "      <td>9.600000</td>\n",
       "      <td>14.266667</td>\n",
       "      <td>21.766667</td>\n",
       "      <td>9.600000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1970-01-01 00:00:01.000</th>\n",
       "      <td>47.761905</td>\n",
       "      <td>975.000000</td>\n",
       "      <td>-317.857143</td>\n",
       "      <td>10.619048</td>\n",
       "      <td>1005.666667</td>\n",
       "      <td>155.476190</td>\n",
       "      <td>47.619048</td>\n",
       "      <td>1004.619048</td>\n",
       "      <td>86.142857</td>\n",
       "      <td>239.428571</td>\n",
       "      <td>...</td>\n",
       "      <td>17.747619</td>\n",
       "      <td>5822.990476</td>\n",
       "      <td>14.228571</td>\n",
       "      <td>309.290476</td>\n",
       "      <td>14.228571</td>\n",
       "      <td>5822.990476</td>\n",
       "      <td>309.290476</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1970-01-01 00:00:01.500</th>\n",
       "      <td>-31.733333</td>\n",
       "      <td>989.700000</td>\n",
       "      <td>-274.700000</td>\n",
       "      <td>20.933333</td>\n",
       "      <td>1007.933333</td>\n",
       "      <td>133.000000</td>\n",
       "      <td>11.966667</td>\n",
       "      <td>1006.100000</td>\n",
       "      <td>94.166667</td>\n",
       "      <td>244.866667</td>\n",
       "      <td>...</td>\n",
       "      <td>2257.843678</td>\n",
       "      <td>18485.903448</td>\n",
       "      <td>38647.288506</td>\n",
       "      <td>2476.855172</td>\n",
       "      <td>38647.288506</td>\n",
       "      <td>18485.903448</td>\n",
       "      <td>2476.855172</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1970-01-01 00:00:02.000</th>\n",
       "      <td>-83.633333</td>\n",
       "      <td>992.066667</td>\n",
       "      <td>-268.833333</td>\n",
       "      <td>-12.933333</td>\n",
       "      <td>1018.166667</td>\n",
       "      <td>89.500000</td>\n",
       "      <td>-26.333333</td>\n",
       "      <td>1001.300000</td>\n",
       "      <td>175.666667</td>\n",
       "      <td>234.466667</td>\n",
       "      <td>...</td>\n",
       "      <td>2173.636782</td>\n",
       "      <td>16930.562069</td>\n",
       "      <td>43889.972414</td>\n",
       "      <td>1953.374713</td>\n",
       "      <td>43889.972414</td>\n",
       "      <td>16930.562069</td>\n",
       "      <td>1953.374713</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1970-01-01 00:00:02.500</th>\n",
       "      <td>-15.100000</td>\n",
       "      <td>991.933333</td>\n",
       "      <td>-279.900000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>1018.566667</td>\n",
       "      <td>71.933333</td>\n",
       "      <td>-64.300000</td>\n",
       "      <td>977.400000</td>\n",
       "      <td>344.233333</td>\n",
       "      <td>195.366667</td>\n",
       "      <td>...</td>\n",
       "      <td>129.909195</td>\n",
       "      <td>605.075862</td>\n",
       "      <td>3338.764368</td>\n",
       "      <td>157.567816</td>\n",
       "      <td>3338.764368</td>\n",
       "      <td>605.075862</td>\n",
       "      <td>157.567816</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 217 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         Accelerometer_RKN^_accX_mean  \\\n",
       "time                                                    \n",
       "1970-01-01 00:00:00.500                    105.000000   \n",
       "1970-01-01 00:00:01.000                     47.761905   \n",
       "1970-01-01 00:00:01.500                    -31.733333   \n",
       "1970-01-01 00:00:02.000                    -83.633333   \n",
       "1970-01-01 00:00:02.500                    -15.100000   \n",
       "\n",
       "                         Accelerometer_RKN^_accY_mean  \\\n",
       "time                                                    \n",
       "1970-01-01 00:00:00.500                    961.666667   \n",
       "1970-01-01 00:00:01.000                    975.000000   \n",
       "1970-01-01 00:00:01.500                    989.700000   \n",
       "1970-01-01 00:00:02.000                    992.066667   \n",
       "1970-01-01 00:00:02.500                    991.933333   \n",
       "\n",
       "                         Accelerometer_RKN^_accZ_mean  \\\n",
       "time                                                    \n",
       "1970-01-01 00:00:00.500                   -357.333333   \n",
       "1970-01-01 00:00:01.000                   -317.857143   \n",
       "1970-01-01 00:00:01.500                   -274.700000   \n",
       "1970-01-01 00:00:02.000                   -268.833333   \n",
       "1970-01-01 00:00:02.500                   -279.900000   \n",
       "\n",
       "                         Accelerometer_HIP_accX_mean  \\\n",
       "time                                                   \n",
       "1970-01-01 00:00:00.500                     0.500000   \n",
       "1970-01-01 00:00:01.000                    10.619048   \n",
       "1970-01-01 00:00:01.500                    20.933333   \n",
       "1970-01-01 00:00:02.000                   -12.933333   \n",
       "1970-01-01 00:00:02.500                    26.000000   \n",
       "\n",
       "                         Accelerometer_HIP_accY_mean  \\\n",
       "time                                                   \n",
       "1970-01-01 00:00:00.500                  1023.166667   \n",
       "1970-01-01 00:00:01.000                  1005.666667   \n",
       "1970-01-01 00:00:01.500                  1007.933333   \n",
       "1970-01-01 00:00:02.000                  1018.166667   \n",
       "1970-01-01 00:00:02.500                  1018.566667   \n",
       "\n",
       "                         Accelerometer_HIP_accZ_mean  \\\n",
       "time                                                   \n",
       "1970-01-01 00:00:00.500                   183.500000   \n",
       "1970-01-01 00:00:01.000                   155.476190   \n",
       "1970-01-01 00:00:01.500                   133.000000   \n",
       "1970-01-01 00:00:02.000                    89.500000   \n",
       "1970-01-01 00:00:02.500                    71.933333   \n",
       "\n",
       "                         Accelerometer_LUA^_accX_mean  \\\n",
       "time                                                    \n",
       "1970-01-01 00:00:00.500                     74.666667   \n",
       "1970-01-01 00:00:01.000                     47.619048   \n",
       "1970-01-01 00:00:01.500                     11.966667   \n",
       "1970-01-01 00:00:02.000                    -26.333333   \n",
       "1970-01-01 00:00:02.500                    -64.300000   \n",
       "\n",
       "                         Accelerometer_LUA^_accY_mean  \\\n",
       "time                                                    \n",
       "1970-01-01 00:00:00.500                   1008.666667   \n",
       "1970-01-01 00:00:01.000                   1004.619048   \n",
       "1970-01-01 00:00:01.500                   1006.100000   \n",
       "1970-01-01 00:00:02.000                   1001.300000   \n",
       "1970-01-01 00:00:02.500                    977.400000   \n",
       "\n",
       "                         Accelerometer_LUA^_accZ_mean  \\\n",
       "time                                                    \n",
       "1970-01-01 00:00:00.500                    101.833333   \n",
       "1970-01-01 00:00:01.000                     86.142857   \n",
       "1970-01-01 00:00:01.500                     94.166667   \n",
       "1970-01-01 00:00:02.000                    175.666667   \n",
       "1970-01-01 00:00:02.500                    344.233333   \n",
       "\n",
       "                         Accelerometer_RUA__accX_mean  ...  \\\n",
       "time                                                   ...   \n",
       "1970-01-01 00:00:00.500                    245.333333  ...   \n",
       "1970-01-01 00:00:01.000                    239.428571  ...   \n",
       "1970-01-01 00:00:01.500                    244.866667  ...   \n",
       "1970-01-01 00:00:02.000                    234.466667  ...   \n",
       "1970-01-01 00:00:02.500                    195.366667  ...   \n",
       "\n",
       "                         InertialMeasurementUnit_R-SHOE_Body_Az_var  \\\n",
       "time                                                                  \n",
       "1970-01-01 00:00:00.500                                    1.366667   \n",
       "1970-01-01 00:00:01.000                                   17.747619   \n",
       "1970-01-01 00:00:01.500                                 2257.843678   \n",
       "1970-01-01 00:00:02.000                                 2173.636782   \n",
       "1970-01-01 00:00:02.500                                  129.909195   \n",
       "\n",
       "                         InertialMeasurementUnit_R-SHOE_AngVelBody_FrameX_var  \\\n",
       "time                                                                            \n",
       "1970-01-01 00:00:00.500                                          21.766667      \n",
       "1970-01-01 00:00:01.000                                        5822.990476      \n",
       "1970-01-01 00:00:01.500                                       18485.903448      \n",
       "1970-01-01 00:00:02.000                                       16930.562069      \n",
       "1970-01-01 00:00:02.500                                         605.075862      \n",
       "\n",
       "                         InertialMeasurementUnit_R-SHOE_AngVelBody_FrameY_var  \\\n",
       "time                                                                            \n",
       "1970-01-01 00:00:00.500                                          14.266667      \n",
       "1970-01-01 00:00:01.000                                          14.228571      \n",
       "1970-01-01 00:00:01.500                                       38647.288506      \n",
       "1970-01-01 00:00:02.000                                       43889.972414      \n",
       "1970-01-01 00:00:02.500                                        3338.764368      \n",
       "\n",
       "                         InertialMeasurementUnit_R-SHOE_AngVelBody_FrameZ_var  \\\n",
       "time                                                                            \n",
       "1970-01-01 00:00:00.500                                           9.600000      \n",
       "1970-01-01 00:00:01.000                                         309.290476      \n",
       "1970-01-01 00:00:01.500                                        2476.855172      \n",
       "1970-01-01 00:00:02.000                                        1953.374713      \n",
       "1970-01-01 00:00:02.500                                         157.567816      \n",
       "\n",
       "                         InertialMeasurementUnit_R-SHOE_AngVelNav_FrameX_var  \\\n",
       "time                                                                           \n",
       "1970-01-01 00:00:00.500                                          14.266667     \n",
       "1970-01-01 00:00:01.000                                          14.228571     \n",
       "1970-01-01 00:00:01.500                                       38647.288506     \n",
       "1970-01-01 00:00:02.000                                       43889.972414     \n",
       "1970-01-01 00:00:02.500                                        3338.764368     \n",
       "\n",
       "                         InertialMeasurementUnit_R-SHOE_AngVelNav_FrameY_var  \\\n",
       "time                                                                           \n",
       "1970-01-01 00:00:00.500                                          21.766667     \n",
       "1970-01-01 00:00:01.000                                        5822.990476     \n",
       "1970-01-01 00:00:01.500                                       18485.903448     \n",
       "1970-01-01 00:00:02.000                                       16930.562069     \n",
       "1970-01-01 00:00:02.500                                         605.075862     \n",
       "\n",
       "                         InertialMeasurementUnit_R-SHOE_AngVelNav_FrameZ_var  \\\n",
       "time                                                                           \n",
       "1970-01-01 00:00:00.500                                           9.600000     \n",
       "1970-01-01 00:00:01.000                                         309.290476     \n",
       "1970-01-01 00:00:01.500                                        2476.855172     \n",
       "1970-01-01 00:00:02.000                                        1953.374713     \n",
       "1970-01-01 00:00:02.500                                         157.567816     \n",
       "\n",
       "                         label  user  trial  \n",
       "time                                         \n",
       "1970-01-01 00:00:00.500    0.0     1      1  \n",
       "1970-01-01 00:00:01.000    0.0     1      1  \n",
       "1970-01-01 00:00:01.500    0.0     1      1  \n",
       "1970-01-01 00:00:02.000    0.0     1      1  \n",
       "1970-01-01 00:00:02.500    0.0     1      1  \n",
       "\n",
       "[5 rows x 217 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data_1S.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MILLISEC                                            float64\n",
       "Accelerometer_RKN^_accX                             float64\n",
       "Accelerometer_RKN^_accY                             float64\n",
       "Accelerometer_RKN^_accZ                             float64\n",
       "Accelerometer_HIP_accX                              float64\n",
       "Accelerometer_HIP_accY                              float64\n",
       "Accelerometer_HIP_accZ                              float64\n",
       "Accelerometer_LUA^_accX                             float64\n",
       "Accelerometer_LUA^_accY                             float64\n",
       "Accelerometer_LUA^_accZ                             float64\n",
       "Accelerometer_RUA__accX                             float64\n",
       "Accelerometer_RUA__accY                             float64\n",
       "Accelerometer_RUA__accZ                             float64\n",
       "Accelerometer_LH_accX                               float64\n",
       "Accelerometer_LH_accY                               float64\n",
       "Accelerometer_LH_accZ                               float64\n",
       "Accelerometer_BACK_accX                             float64\n",
       "Accelerometer_BACK_accY                             float64\n",
       "Accelerometer_BACK_accZ                             float64\n",
       "Accelerometer_RKN__accX                             float64\n",
       "Accelerometer_RKN__accY                             float64\n",
       "Accelerometer_RKN__accZ                             float64\n",
       "Accelerometer_RWR_accX                              float64\n",
       "Accelerometer_RWR_accY                              float64\n",
       "Accelerometer_RWR_accZ                              float64\n",
       "Accelerometer_RUA^_accX                             float64\n",
       "Accelerometer_RUA^_accY                             float64\n",
       "Accelerometer_RUA^_accZ                             float64\n",
       "Accelerometer_LUA__accX                             float64\n",
       "Accelerometer_LUA__accY                             float64\n",
       "                                                     ...   \n",
       "InertialMeasurementUnit_L-SHOE_Nav_Ax               float64\n",
       "InertialMeasurementUnit_L-SHOE_Nav_Ay               float64\n",
       "InertialMeasurementUnit_L-SHOE_Nav_Az               float64\n",
       "InertialMeasurementUnit_L-SHOE_Body_Ax              float64\n",
       "InertialMeasurementUnit_L-SHOE_Body_Ay              float64\n",
       "InertialMeasurementUnit_L-SHOE_Body_Az              float64\n",
       "InertialMeasurementUnit_L-SHOE_AngVelBody_FrameX    float64\n",
       "InertialMeasurementUnit_L-SHOE_AngVelBody_FrameY    float64\n",
       "InertialMeasurementUnit_L-SHOE_AngVelBody_FrameZ    float64\n",
       "InertialMeasurementUnit_L-SHOE_AngVelNav_FrameX     float64\n",
       "InertialMeasurementUnit_L-SHOE_AngVelNav_FrameY     float64\n",
       "InertialMeasurementUnit_L-SHOE_AngVelNav_FrameZ     float64\n",
       "InertialMeasurementUnit_L-SHOE_Compass              float64\n",
       "InertialMeasurementUnit_R-SHOE_EuX                  float64\n",
       "InertialMeasurementUnit_R-SHOE_EuY                  float64\n",
       "InertialMeasurementUnit_R-SHOE_EuZ                  float64\n",
       "InertialMeasurementUnit_R-SHOE_Nav_Ax               float64\n",
       "InertialMeasurementUnit_R-SHOE_Nav_Ay               float64\n",
       "InertialMeasurementUnit_R-SHOE_Nav_Az               float64\n",
       "InertialMeasurementUnit_R-SHOE_Body_Ax              float64\n",
       "InertialMeasurementUnit_R-SHOE_Body_Ay              float64\n",
       "InertialMeasurementUnit_R-SHOE_Body_Az              float64\n",
       "InertialMeasurementUnit_R-SHOE_AngVelBody_FrameX    float64\n",
       "InertialMeasurementUnit_R-SHOE_AngVelBody_FrameY    float64\n",
       "InertialMeasurementUnit_R-SHOE_AngVelBody_FrameZ    float64\n",
       "InertialMeasurementUnit_R-SHOE_AngVelNav_FrameX     float64\n",
       "InertialMeasurementUnit_R-SHOE_AngVelNav_FrameY     float64\n",
       "InertialMeasurementUnit_R-SHOE_AngVelNav_FrameZ     float64\n",
       "user                                                  int64\n",
       "trial                                                 int64\n",
       "Length: 110, dtype: object"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1072e25f8>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Correlaton matrix to identify principal sensor \n",
    "sns.heatmap(data.corr())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "features_analisys = all_data_1S.filter(['Accelerometer_RKN^_accY_mean',\n",
    "'Accelerometer_HIP_accY_mean',\n",
    "'Accelerometer_BACK_accY_mean',\n",
    "'Accelerometer_RKN__accY_mean',\n",
    "'InertialMeasurementUnit_BACK_accY_mean',\n",
    "'InertialMeasurementUnit_BACK_gyroY_mean',\n",
    "'InertialMeasurementUnit_BACK_magneticY_mean',\n",
    "'Accelerometer_RKN^_accY_var',\n",
    "'Accelerometer_HIP_accY_var',\n",
    "'Accelerometer_BACK_accY_var',\n",
    "'Accelerometer_RKN__accY_var',\n",
    "'InertialMeasurementUnit_BACK_accY_var',\n",
    "'InertialMeasurementUnit_BACK_gyroY_var',\n",
    "'InertialMeasurementUnit_BACK_magneticY_var'\n",
    "], axis=1)\n",
    "corr = features_analisys.corr()\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "cax = ax.matshow(corr,cmap='coolwarm', vmin=-1, vmax=1)\n",
    "fig.colorbar(cax)\n",
    "ticks = np.arange(0,len(features_analisys.columns),1)\n",
    "ax.set_xticks(ticks)\n",
    "plt.xticks(rotation=90)\n",
    "ax.set_yticks(ticks)\n",
    "ax.set_xticklabels(features_analisys.columns)\n",
    "ax.set_yticklabels(features_analisys.columns)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "features_analisys = all_data_1S.filter(['Accelerometer_RKN^_accY_mean',\n",
    "'Accelerometer_HIP_accY_mean',\n",
    "'Accelerometer_BACK_accY_mean',\n",
    "'InertialMeasurementUnit_BACK_accY_mean',\n",
    "'Accelerometer_RKN^_accY_var',\n",
    "'Accelerometer_HIP_accY_var',\n",
    "'Accelerometer_BACK_accY_var',\n",
    "'InertialMeasurementUnit_BACK_accY_var'\n",
    "], axis=1)\n",
    "corr = features_analisys.corr()\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "cax = ax.matshow(corr,cmap='coolwarm', vmin=-1, vmax=1)\n",
    "fig.colorbar(cax)\n",
    "ticks = np.arange(0,len(features_analisys.columns),1)\n",
    "ax.set_xticks(ticks)\n",
    "plt.xticks(rotation=90)\n",
    "ax.set_yticks(ticks)\n",
    "ax.set_xticklabels(features_analisys.columns)\n",
    "ax.set_yticklabels(features_analisys.columns)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_data_1S_filtered = all_data_1S.filter(['Accelerometer_RKN^_accY_mean',\n",
    "'Accelerometer_HIP_accY_mean',\n",
    "'Accelerometer_BACK_accY_mean',\n",
    "'InertialMeasurementUnit_BACK_accY_mean',\n",
    "'Accelerometer_RKN^_accY_var',\n",
    "'Accelerometer_HIP_accY_var',\n",
    "'Accelerometer_BACK_accY_var',\n",
    "'InertialMeasurementUnit_BACK_accY_var',\n",
    "'label',\n",
    "'user',\n",
    "'trial'\n",
    "], axis=1)\n",
    "\n",
    "all_data_2S_filtered = all_data_2S.filter(['Accelerometer_RKN^_accY_mean',\n",
    "'Accelerometer_HIP_accY_mean',\n",
    "'Accelerometer_BACK_accY_mean',\n",
    "'InertialMeasurementUnit_BACK_accY_mean',\n",
    "'Accelerometer_RKN^_accY_var',\n",
    "'Accelerometer_HIP_accY_var',\n",
    "'Accelerometer_BACK_accY_var',\n",
    "'InertialMeasurementUnit_BACK_accY_var',\n",
    "'label',\n",
    "'user',\n",
    "'trial'\n",
    "], axis=1)\n",
    "\n",
    "all_data_5S_filtered = all_data_5S.filter(['Accelerometer_RKN^_accY_mean',\n",
    "'Accelerometer_HIP_accY_mean',\n",
    "'Accelerometer_BACK_accY_mean',\n",
    "'InertialMeasurementUnit_BACK_accY_mean',\n",
    "'Accelerometer_RKN^_accY_var',\n",
    "'Accelerometer_HIP_accY_var',\n",
    "'Accelerometer_BACK_accY_var',\n",
    "'InertialMeasurementUnit_BACK_accY_var',\n",
    "'label',\n",
    "'user',\n",
    "'trial'\n",
    "], axis=1)\n",
    "\n",
    "all_data_10S_filtered = all_data_10S.filter(['Accelerometer_RKN^_accY_mean',\n",
    "'Accelerometer_HIP_accY_mean',\n",
    "'Accelerometer_BACK_accY_mean',\n",
    "'InertialMeasurementUnit_BACK_accY_mean',\n",
    "'Accelerometer_RKN^_accY_var',\n",
    "'Accelerometer_HIP_accY_var',\n",
    "'Accelerometer_BACK_accY_var',\n",
    "'InertialMeasurementUnit_BACK_accY_var',\n",
    "'label',\n",
    "'user',\n",
    "'trial'\n",
    "], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " _______________________________________________________\n",
      "                         Accelerometer_RKN^_accY_mean  \\\n",
      "time                                                    \n",
      "1970-01-01 00:00:00.500                    961.666667   \n",
      "1970-01-01 00:00:01.000                    975.000000   \n",
      "1970-01-01 00:00:01.500                    989.700000   \n",
      "1970-01-01 00:00:02.000                    992.066667   \n",
      "1970-01-01 00:00:02.500                    991.933333   \n",
      "\n",
      "                         Accelerometer_HIP_accY_mean  \\\n",
      "time                                                   \n",
      "1970-01-01 00:00:00.500                  1023.166667   \n",
      "1970-01-01 00:00:01.000                  1005.666667   \n",
      "1970-01-01 00:00:01.500                  1007.933333   \n",
      "1970-01-01 00:00:02.000                  1018.166667   \n",
      "1970-01-01 00:00:02.500                  1018.566667   \n",
      "\n",
      "                         Accelerometer_BACK_accY_mean  \\\n",
      "time                                                    \n",
      "1970-01-01 00:00:00.500                    961.666667   \n",
      "1970-01-01 00:00:01.000                    959.809524   \n",
      "1970-01-01 00:00:01.500                    969.600000   \n",
      "1970-01-01 00:00:02.000                    966.100000   \n",
      "1970-01-01 00:00:02.500                    968.833333   \n",
      "\n",
      "                         InertialMeasurementUnit_BACK_accY_mean  \\\n",
      "time                                                              \n",
      "1970-01-01 00:00:00.500                             -177.000000   \n",
      "1970-01-01 00:00:01.000                             -151.857143   \n",
      "1970-01-01 00:00:01.500                             -156.200000   \n",
      "1970-01-01 00:00:02.000                             -212.133333   \n",
      "1970-01-01 00:00:02.500                             -206.866667   \n",
      "\n",
      "                         Accelerometer_RKN^_accY_var  \\\n",
      "time                                                   \n",
      "1970-01-01 00:00:00.500                     5.466667   \n",
      "1970-01-01 00:00:01.000                   390.200000   \n",
      "1970-01-01 00:00:01.500                   890.768966   \n",
      "1970-01-01 00:00:02.000                   646.685057   \n",
      "1970-01-01 00:00:02.500                   159.719540   \n",
      "\n",
      "                         Accelerometer_HIP_accY_var  \\\n",
      "time                                                  \n",
      "1970-01-01 00:00:00.500                    0.966667   \n",
      "1970-01-01 00:00:01.000                  223.233333   \n",
      "1970-01-01 00:00:01.500                  192.685057   \n",
      "1970-01-01 00:00:02.000                  110.143678   \n",
      "1970-01-01 00:00:02.500                  185.219540   \n",
      "\n",
      "                         Accelerometer_BACK_accY_var  \\\n",
      "time                                                   \n",
      "1970-01-01 00:00:00.500                    15.466667   \n",
      "1970-01-01 00:00:01.000                    40.861905   \n",
      "1970-01-01 00:00:01.500                   172.317241   \n",
      "1970-01-01 00:00:02.000                   246.782759   \n",
      "1970-01-01 00:00:02.500                   738.005747   \n",
      "\n",
      "                         InertialMeasurementUnit_BACK_accY_var  label  user  \\\n",
      "time                                                                          \n",
      "1970-01-01 00:00:00.500                             363.200000    0.0     1   \n",
      "1970-01-01 00:00:01.000                             364.928571    0.0     1   \n",
      "1970-01-01 00:00:01.500                            1127.475862    0.0     1   \n",
      "1970-01-01 00:00:02.000                            3587.085057    0.0     1   \n",
      "1970-01-01 00:00:02.500                            3193.498851    0.0     1   \n",
      "\n",
      "                         trial  \n",
      "time                            \n",
      "1970-01-01 00:00:00.500      1  \n",
      "1970-01-01 00:00:01.000      1  \n",
      "1970-01-01 00:00:01.500      1  \n",
      "1970-01-01 00:00:02.000      1  \n",
      "1970-01-01 00:00:02.500      1  \n",
      "\n",
      " _______________________________________________________\n",
      "                     Accelerometer_RKN^_accY_mean  \\\n",
      "time                                                \n",
      "1970-01-01 00:00:01                    975.000000   \n",
      "1970-01-01 00:00:02                    985.039216   \n",
      "1970-01-01 00:00:03                    995.183333   \n",
      "1970-01-01 00:00:04                    989.433333   \n",
      "1970-01-01 00:00:05                    978.966667   \n",
      "\n",
      "                     Accelerometer_HIP_accY_mean  \\\n",
      "time                                               \n",
      "1970-01-01 00:00:01                  1005.666667   \n",
      "1970-01-01 00:00:02                  1013.019608   \n",
      "1970-01-01 00:00:03                  1013.683333   \n",
      "1970-01-01 00:00:04                  1006.233333   \n",
      "1970-01-01 00:00:05                   991.150000   \n",
      "\n",
      "                     Accelerometer_BACK_accY_mean  \\\n",
      "time                                                \n",
      "1970-01-01 00:00:01                    959.809524   \n",
      "1970-01-01 00:00:02                    963.509804   \n",
      "1970-01-01 00:00:03                    979.900000   \n",
      "1970-01-01 00:00:04                    991.316667   \n",
      "1970-01-01 00:00:05                   1005.833333   \n",
      "\n",
      "                     InertialMeasurementUnit_BACK_accY_mean  \\\n",
      "time                                                          \n",
      "1970-01-01 00:00:01                             -151.857143   \n",
      "1970-01-01 00:00:02                             -187.313725   \n",
      "1970-01-01 00:00:03                             -189.983333   \n",
      "1970-01-01 00:00:04                             -175.733333   \n",
      "1970-01-01 00:00:05                             -171.750000   \n",
      "\n",
      "                     Accelerometer_RKN^_accY_var  Accelerometer_HIP_accY_var  \\\n",
      "time                                                                           \n",
      "1970-01-01 00:00:01                   390.200000                  223.233333   \n",
      "1970-01-01 00:00:02                   603.118431                  191.779608   \n",
      "1970-01-01 00:00:03                   383.067514                  189.067514   \n",
      "1970-01-01 00:00:04                   143.503955                  612.893785   \n",
      "1970-01-01 00:00:05                    15.219209                  643.485593   \n",
      "\n",
      "                     Accelerometer_BACK_accY_var  \\\n",
      "time                                               \n",
      "1970-01-01 00:00:01                    40.861905   \n",
      "1970-01-01 00:00:02                   169.254902   \n",
      "1970-01-01 00:00:03                   592.972881   \n",
      "1970-01-01 00:00:04                   359.067514   \n",
      "1970-01-01 00:00:05                   378.446328   \n",
      "\n",
      "                     InertialMeasurementUnit_BACK_accY_var  label  user  trial  \n",
      "time                                                                            \n",
      "1970-01-01 00:00:01                             364.928571    0.0     1      1  \n",
      "1970-01-01 00:00:02                            3124.099608    0.0     1      1  \n",
      "1970-01-01 00:00:03                            2356.965819    0.0     1      1  \n",
      "1970-01-01 00:00:04                             291.859887    0.0     1      1  \n",
      "1970-01-01 00:00:05                             306.495763    0.0     1      1  \n",
      "\n",
      " _______________________________________________________\n",
      "                     Accelerometer_RKN^_accY_mean  \\\n",
      "time                                                \n",
      "1970-01-01 00:00:01                    975.000000   \n",
      "1970-01-01 00:00:02                    985.039216   \n",
      "1970-01-01 00:00:03                    989.950617   \n",
      "1970-01-01 00:00:04                    987.414414   \n",
      "1970-01-01 00:00:05                    985.276596   \n",
      "\n",
      "                     Accelerometer_HIP_accY_mean  \\\n",
      "time                                               \n",
      "1970-01-01 00:00:01                  1005.666667   \n",
      "1970-01-01 00:00:02                  1013.019608   \n",
      "1970-01-01 00:00:03                  1011.604938   \n",
      "1970-01-01 00:00:04                  1009.351351   \n",
      "1970-01-01 00:00:05                  1002.900709   \n",
      "\n",
      "                     Accelerometer_BACK_accY_mean  \\\n",
      "time                                                \n",
      "1970-01-01 00:00:01                    959.809524   \n",
      "1970-01-01 00:00:02                    963.509804   \n",
      "1970-01-01 00:00:03                    974.691358   \n",
      "1970-01-01 00:00:04                    978.540541   \n",
      "1970-01-01 00:00:05                    987.943262   \n",
      "\n",
      "                     InertialMeasurementUnit_BACK_accY_mean  \\\n",
      "time                                                          \n",
      "1970-01-01 00:00:01                             -151.857143   \n",
      "1970-01-01 00:00:02                             -187.313725   \n",
      "1970-01-01 00:00:03                             -180.098765   \n",
      "1970-01-01 00:00:04                             -181.054054   \n",
      "1970-01-01 00:00:05                             -176.546099   \n",
      "\n",
      "                     Accelerometer_RKN^_accY_var  Accelerometer_HIP_accY_var  \\\n",
      "time                                                                           \n",
      "1970-01-01 00:00:01                   390.200000                  223.233333   \n",
      "1970-01-01 00:00:02                   603.118431                  191.779608   \n",
      "1970-01-01 00:00:03                   459.272531                  207.741975   \n",
      "1970-01-01 00:00:04                   355.953972                  427.448157   \n",
      "1970-01-01 00:00:05                   298.558663                  492.904357   \n",
      "\n",
      "                     Accelerometer_BACK_accY_var  \\\n",
      "time                                               \n",
      "1970-01-01 00:00:01                    40.861905   \n",
      "1970-01-01 00:00:02                   169.254902   \n",
      "1970-01-01 00:00:03                   526.016049   \n",
      "1970-01-01 00:00:04                   463.305160   \n",
      "1970-01-01 00:00:05                   698.839615   \n",
      "\n",
      "                     InertialMeasurementUnit_BACK_accY_var  label  user  trial  \n",
      "time                                                                            \n",
      "1970-01-01 00:00:01                             364.928571    0.0     1      1  \n",
      "1970-01-01 00:00:02                            3124.099608    0.0     1      1  \n",
      "1970-01-01 00:00:03                            2112.140123    0.0     1      1  \n",
      "1970-01-01 00:00:04                            1610.197052    0.0     1      1  \n",
      "1970-01-01 00:00:05                            1353.263931    0.0     1      1  \n",
      "\n",
      " _______________________________________________________\n",
      "                     Accelerometer_RKN^_accY_mean  \\\n",
      "time                                                \n",
      "1970-01-01 00:00:01                    975.000000   \n",
      "1970-01-01 00:00:02                    985.039216   \n",
      "1970-01-01 00:00:03                    989.950617   \n",
      "1970-01-01 00:00:04                    987.414414   \n",
      "1970-01-01 00:00:05                    985.276596   \n",
      "\n",
      "                     Accelerometer_HIP_accY_mean  \\\n",
      "time                                               \n",
      "1970-01-01 00:00:01                  1005.666667   \n",
      "1970-01-01 00:00:02                  1013.019608   \n",
      "1970-01-01 00:00:03                  1011.604938   \n",
      "1970-01-01 00:00:04                  1009.351351   \n",
      "1970-01-01 00:00:05                  1002.900709   \n",
      "\n",
      "                     Accelerometer_BACK_accY_mean  \\\n",
      "time                                                \n",
      "1970-01-01 00:00:01                    959.809524   \n",
      "1970-01-01 00:00:02                    963.509804   \n",
      "1970-01-01 00:00:03                    974.691358   \n",
      "1970-01-01 00:00:04                    978.540541   \n",
      "1970-01-01 00:00:05                    987.943262   \n",
      "\n",
      "                     InertialMeasurementUnit_BACK_accY_mean  \\\n",
      "time                                                          \n",
      "1970-01-01 00:00:01                             -151.857143   \n",
      "1970-01-01 00:00:02                             -187.313725   \n",
      "1970-01-01 00:00:03                             -180.098765   \n",
      "1970-01-01 00:00:04                             -181.054054   \n",
      "1970-01-01 00:00:05                             -176.546099   \n",
      "\n",
      "                     Accelerometer_RKN^_accY_var  Accelerometer_HIP_accY_var  \\\n",
      "time                                                                           \n",
      "1970-01-01 00:00:01                   390.200000                  223.233333   \n",
      "1970-01-01 00:00:02                   603.118431                  191.779608   \n",
      "1970-01-01 00:00:03                   459.272531                  207.741975   \n",
      "1970-01-01 00:00:04                   355.953972                  427.448157   \n",
      "1970-01-01 00:00:05                   298.558663                  492.904357   \n",
      "\n",
      "                     Accelerometer_BACK_accY_var  \\\n",
      "time                                               \n",
      "1970-01-01 00:00:01                    40.861905   \n",
      "1970-01-01 00:00:02                   169.254902   \n",
      "1970-01-01 00:00:03                   526.016049   \n",
      "1970-01-01 00:00:04                   463.305160   \n",
      "1970-01-01 00:00:05                   698.839615   \n",
      "\n",
      "                     InertialMeasurementUnit_BACK_accY_var  label  user  trial  \n",
      "time                                                                            \n",
      "1970-01-01 00:00:01                             364.928571    0.0     1      1  \n",
      "1970-01-01 00:00:02                            3124.099608    0.0     1      1  \n",
      "1970-01-01 00:00:03                            2112.140123    0.0     1      1  \n",
      "1970-01-01 00:00:04                            1610.197052    0.0     1      1  \n",
      "1970-01-01 00:00:05                            1353.263931    0.0     1      1  \n"
     ]
    }
   ],
   "source": [
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_1S_filtered.head())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_2S_filtered.head())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_5S_filtered.head())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_10S_filtered.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " _______________________________________________________\n",
      "1.0    23172\n",
      "2.0    13218\n",
      "0.0    10747\n",
      "4.0     9119\n",
      "5.0     1692\n",
      "Name: label, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "1.0    11558\n",
      "2.0     6633\n",
      "0.0     5368\n",
      "4.0     4565\n",
      "5.0      844\n",
      "Name: label, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "1.0    11796\n",
      "2.0     6378\n",
      "0.0     5326\n",
      "4.0     4607\n",
      "5.0      861\n",
      "Name: label, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "1.0    12483\n",
      "2.0     5500\n",
      "0.0     5354\n",
      "4.0     4703\n",
      "5.0      928\n",
      "Name: label, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_1S_filtered.label.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_2S_filtered.label.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_5S_filtered.label.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_10S_filtered.label.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " _______________________________________________________\n",
      "6    14981\n",
      "1    11513\n",
      "3     7956\n",
      "5     7930\n",
      "2     7863\n",
      "4     7705\n",
      "Name: trial, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "6    7490\n",
      "1    5755\n",
      "3    3977\n",
      "5    3964\n",
      "2    3931\n",
      "4    3851\n",
      "Name: trial, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "6    7490\n",
      "1    5755\n",
      "3    3977\n",
      "5    3964\n",
      "2    3931\n",
      "4    3851\n",
      "Name: trial, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "6    7490\n",
      "1    5755\n",
      "3    3977\n",
      "5    3964\n",
      "2    3931\n",
      "4    3851\n",
      "Name: trial, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_1S_filtered.trial.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_2S_filtered.trial.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_5S_filtered.trial.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_10S_filtered.trial.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " _______________________________________________________\n",
      "1    15641\n",
      "2    15010\n",
      "3    14455\n",
      "4    12842\n",
      "Name: user, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "1    7820\n",
      "2    7503\n",
      "3    7226\n",
      "4    6419\n",
      "Name: user, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "1    7820\n",
      "2    7503\n",
      "3    7226\n",
      "4    6419\n",
      "Name: user, dtype: int64\n",
      "\n",
      " _______________________________________________________\n",
      "1    7820\n",
      "2    7503\n",
      "3    7226\n",
      "4    6419\n",
      "Name: user, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_1S_filtered.user.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_2S_filtered.user.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_5S_filtered.user.value_counts())\n",
    "print(\"\\n _______________________________________________________\")  \n",
    "print(all_data_10S_filtered.user.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11decf2e8>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#separate by class, see feature mean\n",
    "all_data_1S_0 = all_data_1S_filtered[all_data_1S_filtered.label==0]\n",
    "all_data_1S_1 = all_data_1S_filtered[all_data_1S_filtered.label==1]\n",
    "all_data_1S_2 = all_data_1S_filtered[all_data_1S_filtered.label==2]\n",
    "all_data_1S_4 = all_data_1S_filtered[all_data_1S_filtered.label==4]\n",
    "all_data_1S_5 = all_data_1S_filtered[all_data_1S_filtered.label==5]\n",
    "\n",
    "draw_col = 10\n",
    "sns.distplot(all_data_1S_0.iloc[:,draw_col], hist=False, kde=True, color='red')\n",
    "sns.distplot(all_data_1S_1.iloc[:,draw_col], hist=False, kde=True, color='green')\n",
    "sns.distplot(all_data_1S_2.iloc[:,draw_col], hist=False, kde=True, color='yellow')\n",
    "sns.distplot(all_data_1S_4.iloc[:,draw_col], hist=False, kde=True, color='blue')\n",
    "sns.distplot(all_data_1S_5.iloc[:,draw_col], hist=False, kde=True, color='black')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11e0de358>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#separate by class, see feature mean\n",
    "all_data_2S_0 = all_data_2S_filtered[all_data_2S_filtered.label==0]\n",
    "all_data_2S_1 = all_data_2S_filtered[all_data_2S_filtered.label==1]\n",
    "all_data_2S_2 = all_data_2S_filtered[all_data_2S_filtered.label==2]\n",
    "all_data_2S_4 = all_data_2S_filtered[all_data_2S_filtered.label==4]\n",
    "all_data_2S_5 = all_data_2S_filtered[all_data_2S_filtered.label==5]\n",
    "\n",
    "draw_col = 10\n",
    "sns.distplot(all_data_2S_0.iloc[:,draw_col], hist=False, kde=True, color='red')\n",
    "sns.distplot(all_data_2S_1.iloc[:,draw_col], hist=False, kde=True, color='green')\n",
    "sns.distplot(all_data_2S_2.iloc[:,draw_col], hist=False, kde=True, color='yellow')\n",
    "sns.distplot(all_data_2S_4.iloc[:,draw_col], hist=False, kde=True, color='blue')\n",
    "sns.distplot(all_data_2S_5.iloc[:,draw_col], hist=False, kde=True, color='black')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11dd07278>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#separate by class, see feature mean\n",
    "all_data_5S_0 = all_data_5S_filtered[all_data_5S_filtered.label==0]\n",
    "all_data_5S_1 = all_data_5S_filtered[all_data_5S_filtered.label==1]\n",
    "all_data_5S_2 = all_data_5S_filtered[all_data_5S_filtered.label==2]\n",
    "all_data_5S_4 = all_data_5S_filtered[all_data_5S_filtered.label==4]\n",
    "all_data_5S_5 = all_data_5S_filtered[all_data_5S_filtered.label==5]\n",
    "\n",
    "draw_col = 10\n",
    "sns.distplot(all_data_5S_0.iloc[:,draw_col], hist=False, kde=True, color='red')\n",
    "sns.distplot(all_data_5S_1.iloc[:,draw_col], hist=False, kde=True, color='green')\n",
    "sns.distplot(all_data_5S_2.iloc[:,draw_col], hist=False, kde=True, color='yellow')\n",
    "sns.distplot(all_data_5S_4.iloc[:,draw_col], hist=False, kde=True, color='blue')\n",
    "sns.distplot(all_data_5S_5.iloc[:,draw_col], hist=False, kde=True, color='black')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11a4d6780>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#separate by class, see feature mean\n",
    "all_data_10S_0 = all_data_10S_filtered[all_data_10S_filtered.label==0]\n",
    "all_data_10S_1 = all_data_10S_filtered[all_data_10S_filtered.label==1]\n",
    "all_data_10S_2 = all_data_10S_filtered[all_data_10S_filtered.label==2]\n",
    "all_data_10S_4 = all_data_10S_filtered[all_data_10S_filtered.label==4]\n",
    "all_data_10S_5 = all_data_10S_filtered[all_data_10S_filtered.label==5]\n",
    "\n",
    "draw_col = 10\n",
    "sns.distplot(all_data_10S_0.iloc[:,draw_col], hist=False, kde=True, color='red')\n",
    "sns.distplot(all_data_10S_1.iloc[:,draw_col], hist=False, kde=True, color='green')\n",
    "sns.distplot(all_data_10S_2.iloc[:,draw_col], hist=False, kde=True, color='yellow')\n",
    "sns.distplot(all_data_10S_4.iloc[:,draw_col], hist=False, kde=True, color='blue')\n",
    "sns.distplot(all_data_10S_5.iloc[:,draw_col], hist=False, kde=True, color='black')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#############################\n",
      "X_train: (38825, 8)\n",
      "X_test: (19123, 8)\n",
      "y_train: (38825,)\n",
      "y_test: (19123,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[2604  535  340   51   11]\n",
      " [ 261 6668  628   54    1]\n",
      " [ 261  989 3173   18    0]\n",
      " [  46   87   24 2788    8]\n",
      " [   8    2    1    8  557]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.82      0.74      0.77      3541\n",
      "         1.0       0.81      0.88      0.84      7612\n",
      "         2.0       0.76      0.71      0.74      4441\n",
      "         4.0       0.96      0.94      0.95      2953\n",
      "         5.0       0.97      0.97      0.97       576\n",
      "\n",
      "   micro avg       0.83      0.83      0.83     19123\n",
      "   macro avg       0.86      0.85      0.85     19123\n",
      "weighted avg       0.83      0.83      0.82     19123\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.8257072635046803\n",
      "#############################\n",
      "X_train: (19408, 8)\n",
      "X_test: (9560, 8)\n",
      "y_train: (19408,)\n",
      "y_test: (9560,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[1311  262  151   26    5]\n",
      " [ 123 3308  323   35    0]\n",
      " [  89  498 1591    4    0]\n",
      " [  32   30   22 1449    6]\n",
      " [  12    0    0    3  280]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.84      0.75      0.79      1755\n",
      "         1.0       0.81      0.87      0.84      3789\n",
      "         2.0       0.76      0.73      0.75      2182\n",
      "         4.0       0.96      0.94      0.95      1539\n",
      "         5.0       0.96      0.95      0.96       295\n",
      "\n",
      "   micro avg       0.83      0.83      0.83      9560\n",
      "   macro avg       0.86      0.85      0.86      9560\n",
      "weighted avg       0.83      0.83      0.83      9560\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.830439330543933\n",
      "#############################\n",
      "X_train: (19408, 8)\n",
      "X_test: (9560, 8)\n",
      "y_train: (19408,)\n",
      "y_test: (9560,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[1507  122   83   18   10]\n",
      " [  57 3410  356   25    1]\n",
      " [  48  481 1581   20    0]\n",
      " [  21   32   21 1467    6]\n",
      " [   1    1    2    3  287]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.92      0.87      0.89      1740\n",
      "         1.0       0.84      0.89      0.86      3849\n",
      "         2.0       0.77      0.74      0.76      2130\n",
      "         4.0       0.96      0.95      0.95      1547\n",
      "         5.0       0.94      0.98      0.96       294\n",
      "\n",
      "   micro avg       0.86      0.86      0.86      9560\n",
      "   macro avg       0.89      0.88      0.89      9560\n",
      "weighted avg       0.86      0.86      0.86      9560\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.8631799163179916\n",
      "#############################\n",
      "X_train: (19408, 8)\n",
      "X_test: (9560, 8)\n",
      "y_train: (19408,)\n",
      "y_test: (9560,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[1621   72   55   13   11]\n",
      " [  41 3819  179   24    0]\n",
      " [  37  337 1429   17    1]\n",
      " [  10   57   23 1492    6]\n",
      " [   1    3    4    1  307]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.95      0.91      0.93      1772\n",
      "         1.0       0.89      0.94      0.91      4063\n",
      "         2.0       0.85      0.78      0.81      1821\n",
      "         4.0       0.96      0.94      0.95      1588\n",
      "         5.0       0.94      0.97      0.96       316\n",
      "\n",
      "   micro avg       0.91      0.91      0.91      9560\n",
      "   macro avg       0.92      0.91      0.91      9560\n",
      "weighted avg       0.91      0.91      0.91      9560\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.9066945606694561\n",
      "#############################\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"#############################\")\n",
    "excercise_1(all_data_1S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_1(all_data_2S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_1(all_data_5S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_1(all_data_10S_filtered, 50)\n",
    "print(\"#############################\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#############################\n",
      "X_train_2: (45106, 8)\n",
      "X_test_2: (12842, 8)\n",
      "y_train_2: (45106,)\n",
      "y_test_2: (12842,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[1168 1479  358   31    0]\n",
      " [ 208 4060  803   48    0]\n",
      " [ 340  857 1555    1    0]\n",
      " [  89  171   37 1284    0]\n",
      " [  50   17    0  175  111]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.63      0.38      0.48      3036\n",
      "         1.0       0.62      0.79      0.69      5119\n",
      "         2.0       0.56      0.56      0.56      2753\n",
      "         4.0       0.83      0.81      0.82      1581\n",
      "         5.0       1.00      0.31      0.48       353\n",
      "\n",
      "   micro avg       0.64      0.64      0.64     12842\n",
      "   macro avg       0.73      0.57      0.61     12842\n",
      "weighted avg       0.65      0.64      0.63     12842\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.6368166952188132\n",
      "#############################\n",
      "X_train_2: (22549, 8)\n",
      "X_test_2: (6419, 8)\n",
      "y_train_2: (22549,)\n",
      "y_test_2: (6419,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[ 630  675  192   18    0]\n",
      " [ 106 1951  462   23    0]\n",
      " [ 129  466  794    4    0]\n",
      " [  48   88   16  641    0]\n",
      " [  36    3    1   90   46]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.66      0.42      0.51      1515\n",
      "         1.0       0.61      0.77      0.68      2542\n",
      "         2.0       0.54      0.57      0.56      1393\n",
      "         4.0       0.83      0.81      0.82       793\n",
      "         5.0       1.00      0.26      0.41       176\n",
      "\n",
      "   micro avg       0.63      0.63      0.63      6419\n",
      "   macro avg       0.73      0.56      0.60      6419\n",
      "weighted avg       0.65      0.63      0.62      6419\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.6328088487303318\n",
      "#############################\n",
      "X_train_2: (22549, 8)\n",
      "X_test_2: (6419, 8)\n",
      "y_train_2: (22549,)\n",
      "y_test_2: (6419,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[ 674  568  257   21    0]\n",
      " [  90 2094  511   21    0]\n",
      " [  57  395  747    2    0]\n",
      " [  61  102   15  617    0]\n",
      " [  58    7    4   38   80]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.72      0.44      0.55      1520\n",
      "         1.0       0.66      0.77      0.71      2716\n",
      "         2.0       0.49      0.62      0.55      1201\n",
      "         4.0       0.88      0.78      0.83       795\n",
      "         5.0       1.00      0.43      0.60       187\n",
      "\n",
      "   micro avg       0.66      0.66      0.66      6419\n",
      "   macro avg       0.75      0.61      0.65      6419\n",
      "weighted avg       0.68      0.66      0.65      6419\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.6561769746066366\n",
      "#############################\n",
      "X_train_2: (22549, 8)\n",
      "X_test_2: (6419, 8)\n",
      "y_train_2: (22549,)\n",
      "y_test_2: (6419,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[ 753  511  246   23    0]\n",
      " [  58 2386  408   17    0]\n",
      " [  41  546  427    1    0]\n",
      " [  50  175    9  561    0]\n",
      " [  66   21    0   41   79]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.78      0.49      0.60      1533\n",
      "         1.0       0.66      0.83      0.73      2869\n",
      "         2.0       0.39      0.42      0.41      1015\n",
      "         4.0       0.87      0.71      0.78       795\n",
      "         5.0       1.00      0.38      0.55       207\n",
      "\n",
      "   micro avg       0.66      0.66      0.66      6419\n",
      "   macro avg       0.74      0.57      0.61      6419\n",
      "weighted avg       0.68      0.66      0.65      6419\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.6552422495715844\n",
      "#############################\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqAAAAJGCAYAAACEITdaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3Xu4JHV95/H3Z4a7gFwlCChEB7LoBtQJoijxyk1XNEaFJIDX0Q2umsQousnCaky8xRgfFRd1AhrF4CJKDIqIEi8rhgFHrhIGBJmR24ABRUAGv/tH1zHt8ZwzPeP0r3tOv1889ZzuX1VXfXuOMF8/9auqVBWSJElSKwtGXYAkSZImiw2oJEmSmrIBlSRJUlM2oJIkSWrKBlSSRiDJzkmuTfIbo65FklqLV8FLUntJTgEuqKpPjroWSWrNBlSSJElNeQpeGrEk1yf5WZKdpo1/J0kl2XPa+End+OP7xt6c5Cfdcm+SB/reX9FtU0nu7sZWJXlPkoV9+7ggycu711/o+3z/fivJw2b4Dk9J8vO+bVcmOSPJ70zbrr+GqeUNc/zZHJDknCT/keSOJP+W5CV967dLcnKSm5P8NMll/ev7/nxvTfKgvrGXJ7lgjrr+oxt/cZJvzFDX9Ume0b3ePcmZSVYnuTPJ5Ule3K3bs9v3Jn2ffWKSryT5cbf9PyfZd9qfZSX54LRjfmNqv5K0sbMBlcbD94Gjp94k+a/AVtM3ShLgWOCO7icAVfXXVbV1VW0NvAr41tT7qnpU3y7267b5XeBFwEtnKqaqDu/7/NbAg4ELgY9V1Q9m+Q4/7LbdBjgQ+B7w9SRPn7bdfv37rqp3zrSzJE8AvgL8K/BIYEfgvwOHd+s3A74MPBx4QlfjnwNvT/Kn03a3EHjtLHXPVNd2a9m238eBG7s6dgSOAW6Z4zt9Cfgc8FBgL+C7wDeT/GbfpncDx0z/Px+SNF/YgErj4eP0NZTAccDHZtjuycCuwGuAo7ombJ1V1Qrgm8D+A37kr4Ed6DWAa9t3VdXKqvpfwEeAd6xPjcC7gNOq6h1Vtbrb78VV9cJu/THAw4AXVNX3q+r+qvoivT+btyTZdtq+Xp9kXRrLQf0OcGpV3V1Va6rqO1X1hVm2fSe9Jv7vq+rHVXVHVf0Fveb+pL7t/gM4FThxCPVK0sjZgErj4UJg2yT/pTstfhTwjzNsdxzwz8AZ3fv/tj4HS/Jb9JrZFQNseyTwSuD5VfXTdTzUZ4DH9p/+HrC+reilmv93js2eCXyhqu6eNn4msEX3+SnLgAuA169LHQO6EPhAkqNmmp4wpftOTwQ+PcPqM+h9n35vA56fZJ8NVqkkjQkbUGl8TKWgzwSuAlb1r+wamBcAn6yq++k1Z8dO38laXJLk7m7/FwAfnGvjJI+gl8S9rKquWcdjAfwQCNCfPF7SzemcWg6d4XPb0/vv001z7HunmdZX1Rpgdbe+3/8C/keSnWfZX39d75vjuNO9APg68JfA95Msnz73tbMDs3+nm6bXW1U3Ax8C3rIOtUjSRsEGVBofHwf+AHgxM59+fx6wBjine/8J4PA5GqqZPBbYmt78z8cDsyaTSbag1+Quraoz1+EY/XYDit4p5V/UUFXb9S3nzvC5HwE/pzfdYDarZ1rfXfCzU7f+F6rqcuDzwAmz7K+/rtd0Y2uATWfYdlPg/m6/P6qqE7q5trsAy4HPdvN1B/1Ou06vt/MO4NAk+81SsyRtlGxApTFRVTfQuxjpCHqnrqc7jl7z+IMkN9M7lbspvaZ1XY5TVXUG8C16qeBsPgD8BHjjuux/mucBl8xwmnxO3an+bwHPn2OzL9NrwKc30c8H7qN3any6E4FX0GuMB/ED4GH9zWSXRD8EuGGGulcD76Z3gdEO09bdTe87vWCG47wQOH+G/d0OvBd464D1StJGwQZUGi8vA542vWFLshvwdODZ9C4c2h/Yj15Ctq6n4ae8HXhFZngST5KXdsd6UXdKe2Dp2S3JicDLgTevZ31vAF6c5M+T7Njte78kn+rWfxxYCXy6u93Rpt3p/PcBJ1XVndN32F189U/0LlQaxLeBe4ETkmzRNbtvpzen9IaupnckeXSSTZJsQ+9CrRVd8zjdCcBxSV6TZJsk2yf5K3rzVf/3LDW8h97c0f8yYM2SNPZsQKUxUlXXVtWyGVYdAyyvqi9V1c1TC71m67eTPHo9jnUZ8DV6ty6a7i/oJXj/nl+9H+iTZ9nlQ5P8hF5qehHwX4GnVNWXpm333Wn7e+8s9f0/4Gndcl2SO4BT6KYgVNV9wDPo3QLp28Bd9Jq1/1lV75rjq7+FOaYeTKvhPuBZwFPoNbvX0Us3X1j/+RSPrYCz6E0zuI7e7ZieM8v+vgEcCvwevXmfNwCPAZ402xzbqrqL3tXzO8y0XpI2Rj4JSZIkSU2ZgEqSJKkpG1BJkiQ1ZQMqSZKkpmxAJUmS1NQmoy6g37bb71APeegeoy5Ds1j4K/fV1jjZbJOFoy5Bc9h0of/+jDN/O+PrhhuuZ/Xq1fPqV7Rw24dXrbln6Mepe247t6oOG/qB1sNYNaAPeegevPv0L466DM1iu803G3UJmsPDd1ynx62rsV223XzUJWgOCxbMq/5mXjno8YtHXcIGV2vuYfN9Xjj049y7/APTH0k8NjwFL0mSpKbGKgGVJEma/wKZ7Axwsr+9JEmSmjMBlSRJainAhF/YawIqSZKkpkxAJUmSWnMOqCRJktSOCagkSVJrzgGVJEmS2jEBlSRJasr7gE72t5ckSVJzJqCSJEmtOQdUkiRJascEVJIkqaXgHNBRFyBJkqTJYgIqSZLUVJwDOuoCJEmSNFlMQCVJklpzDqgkSZLUjg2oJElSa8nwl7WWkD2SfDXJlUmuSPLabnyHJOcluab7uX03niTvS7IiyaVJHtu3r+O67a9Jctzajm0DKkmSNJnWAH9WVfsCBwLHJ9kXOAE4v6oWAed37wEOBxZ1yxLgZOg1rMCJwOOBA4ATp5rW2diASpIkNdU9C37Yy1pU1U1VdUn3+sfAVcBuwJHAad1mpwHP7V4fCXysei4EtkuyK3AocF5V3VFVPwLOAw6b69hehCRJkjQ/7ZRkWd/7U6rqlJk2TLIn8Bjg28AuVXVTt+pmYJfu9W7AjX0fW9mNzTY+KxtQSZKklkKr+4CurqrFa9soydbAmcDrququ9NVWVZWkNnRhnoKXJEmaUEk2pdd8fqKqPtMN39KdWqf7eWs3vgrYo+/ju3djs43PygZUkiSptTGYA5pe1PlR4Kqqek/fqrOBqSvZjwM+1zd+bHc1/IHAnd2p+nOBQ5Js3118dEg3NitPwUuSJE2mg4BjgMuSLO/G3gy8HTgjycuAG4AXduvOAY4AVgA/BV4CUFV3JHkrcFG33Vuq6o65DmwDKkmS1FTG4klIVfWNXjEzevoM2xdw/Cz7WgosHfTYo//2kiRJmigmoJIkSa0taHIV/NgyAZUkSVJTJqCSJEkthbGYAzpKk/3tJUmS1JwJqCRJUmttnoQ0tkxAJUmS1JQJqCRJUlPjcR/QUZrsby9JkqTmTEAlSZJacw6oJEmS1I4NqCRJkpryFLwkSVJrXoQkSZIktWMCKkmS1FLiRUijLkCSJEmTxQRUkiSpNeeASpIkSe2YgEqSJLXmHNDhSXJYkquTrEhywjCPJUmSpI3D0BLQJAuBDwDPBFYCFyU5u6quHNYxJUmSxl+cAzrEfR8ArKiq66rqZ8CngCOHeDxJkiRtBIY5B3Q34Ma+9yuBx0/fKMkSYAnAzrvuNsRyJEmSxoRzQEerqk6pqsVVtXjb7XccdTmSJEkasmEmoKuAPfre796NSZIkTa7gHNAh7vsiYFGSvZJsBhwFnD3E40mSJGkjMLQEtKrWJHk1cC6wEFhaVVcM63iSJEkbB6+CH+qN6KvqHOCcYR5DkiRJGxefhCRJktSaV8FLkiRJ7ZiASpIktTbhc0An+9tLkiSpORNQSZKk1pwDKkmSJLVjAipJktRSvA/oZH97SZIkNWcCKkmS1JpzQCVJkqR2TEAlSZIaiwmoJEmS1I4JqCRJUkPBBNQEVJIkSU2ZgEqSJLWUbplgJqCSJElqygRUkiSpqTgHdNQFSJIkabKYgEqSJDVmAipJkiQ1ZAIqSZLUmAmoJEmS1JAJqCRJUmMmoJIkSVJDJqCSJEkt+SQkE1BJkqRJlGRpkluTXN439k9JlnfL9UmWd+N7Jrmnb92H+j7zuCSXJVmR5H0ZYH6BCagkSVJDGZ8nIZ0KvB/42NRAVb1o6nWSvwXu7Nv+2qraf4b9nAy8Avg2cA5wGPCFuQ5sAipJkjSBquprwB0zretSzBcCp8+1jyS7AttW1YVVVfSa2eeu7dgmoJIkSY01SkB3SrKs7/0pVXXKgJ99MnBLVV3TN7ZXku8AdwF/UVVfB3YDVvZts7Ibm5MNqCRJ0vy0uqoWr+dnj+aX08+bgIdV1e1JHgd8Nsmj1rcwG1BJkqTGxmQO6IySbAL8HvC4qbGqug+4r3t9cZJrgb2BVcDufR/fvRubk3NAJUmS1O8ZwPeq6hen1pPsnGRh9/o3gUXAdVV1E3BXkgO7eaPHAp9b2wFsQCVJkhpLMvRlgBpOB74F7JNkZZKXdauO4lcvPjoYuLS7LdP/BV5VVVMXMP0x8BFgBXAta7kCHjwFL0mSNJGq6uhZxl88w9iZwJmzbL8MePS6HNsGVJIkqSWfhOQpeEmSJLVlAipJktTYOF8F34IJqCRJkpoyAZUkSWpojJ4FPzImoJIkSWrKBFSSJKkxE1BJkiSpIRNQSZKk1iY7ADUBlSRJUlsmoJIkSS3FOaAmoJIkSWpqrBLQbTbflN995ENGXYZm8dCDXjvqEjSHr376r0Zdguawy7abj7oESWPEBFSSJElqaKwSUEmSpElgAipJkiQ1ZAMqSZKkpjwFL0mS1FCIp+BHXYAkSZImiwmoJElSa5MdgJqASpIkqS0TUEmSpJZ8FKcJqCRJktoyAZUkSWrMBFSSJElqyARUkiSpMRNQSZIkqSETUEmSpNYmOwA1AZUkSVJbJqCSJEmNOQdUkiRJasgEVJIkqaEkJqCjLkCSJEmTxQRUkiSpMRNQSZIkqSETUEmSpMZMQCVJkqSGTEAlSZJam+wA1ARUkiRJbZmASpIkNeYcUEmSJKkhE1BJkqSWYgJqAipJkqSmTEAlSZIaCjDhAagJqCRJktoyAZUkSWoqzgEddQGSJEmaLCagkiRJjU14AGoCKkmSpLZMQCVJkhpzDqgkSZLUkAmoJElSS3EOqAmoJEmSmrIBlSRJaijAggUZ+rLWOpKlSW5Ncnnf2ElJViVZ3i1H9K17U5IVSa5Ocmjf+GHd2IokJwzyZ2ADKkmSNJlOBQ6bYfzvqmr/bjkHIMm+wFHAo7rPfDDJwiQLgQ8AhwP7Akd3287JOaCSJEmNjcMc0Kr6WpI9B9z8SOBTVXUf8P0kK4ADunUrquo6gCSf6ra9cq6dmYBKkiTNTzslWda3LBnwc69Ocml3in77bmw34Ma+bVZ2Y7ONz8kEVJIkqbFG9wFdXVWL1/EzJwNvBar7+bfASzd0YTagkiRJAqCqbpl6neTDwOe7t6uAPfo23b0bY47xWXkKXpIkqaXuPqDDXtartGTXvrfPA6aukD8bOCrJ5kn2AhYB/wZcBCxKsleSzehdqHT22o5jAipJkjSBkpwOPIXeXNGVwInAU5LsT+8U/PXAKwGq6ookZ9C7uGgNcHxVPdDt59XAucBCYGlVXbG2Yw+tAU2yFHg2cGtVPXpYx5EkSdqYhPF4FnxVHT3D8Efn2P5twNtmGD8HOGddjj3MU/CnMvO9pSRJkjTBhpaAruO9pSRJkiZExiIBHaWRX4SUZMnU/alWr75t1OVIkiRpyEbegFbVKVW1uKoW77TTzqMuR5IkaejG9Sr4VkbegEqSJGmyeBsmSZKkxpwDOiTdvaW+BeyTZGWSlw3rWJIkSdp4DPMq+JnuLSVJkjTZNoI5msPmHFBJkiQ15RxQSZKkhsblSUijZAIqSZKkpkxAJUmSGpvwANQEVJIkSW2ZgEqSJDXmHFBJkiSpIRNQSZKkxiY8ADUBlSRJUlsmoJIkSS3FOaAmoJIkSWrKBFSSJKmh3pOQRl3FaJmASpIkqSkbUEmSJDXlKXhJkqSm4kVIoy5AkiRJk8UEVJIkqbEJD0BNQCVJktSWCagkSVJjzgGVJEmSGjIBlSRJainOATUBlSRJUlMmoJIkSQ31HsU52RGoCagkSZKaMgGVJElqzARUkiRJasgEVJIkqbEJD0BNQCVJktSWCagkSVJjzgGVJEmSGjIBlSRJasknIZmASpIkqS0TUEmSpIZCnAM66gIkSZI0WUxAJUmSGpvwANQEVJIkSW2ZgEqSJDW2YMIjUBNQSZIkNWUCKkmS1NiEB6AmoJIkSWrLBFSSJKmhxGfBm4BKkiSpKRNQSZKkxhZMdgBqAipJkqS2TEAlSZIacw6oJEmSJk6SpUluTXJ539i7knwvyaVJzkqyXTe+Z5J7kizvlg/1feZxSS5LsiLJ+zJAdz1WCWgCm0z6pIgx9u2z/2bUJWgOl95656hL0Bz2X7DdqEuQNEbGJAA9FXg/8LG+sfOAN1XVmiTvAN4EvLFbd21V7T/Dfk4GXgF8GzgHOAz4wlwHNgGVJEmaQFX1NeCOaWNfqqo13dsLgd3n2keSXYFtq+rCqip6zexz13ZsG1BJkqSGAqTBP8BOSZb1LUvWsdSX8stJ5l5JvpPkX5M8uRvbDVjZt83KbmxOY3UKXpIkSRvM6qpavD4fTPI/gTXAJ7qhm4CHVdXtSR4HfDbJo9a3MBtQSZKkxsb5kpckLwaeDTy9O61OVd0H3Ne9vjjJtcDewCp++TT97t3YnDwFL0mSJACSHAa8AXhOVf20b3znJAu7178JLAKuq6qbgLuSHNhd/X4s8Lm1HccEVJIkqaVkLO4DmuR04Cn05oquBE6kd9X75sB5XY0XVtWrgIOBtyS5H/g58KqqmrqA6Y/pXVG/Jb05o3NeAQ82oJIkSROpqo6eYfijs2x7JnDmLOuWAY9el2PbgEqSJDU2BgHoSDkHVJIkSU2ZgEqSJDUUYMGER6AmoJIkSWrKBFSSJKmxCQ9ATUAlSZLUlgmoJElSY+NwH9BRMgGVJElSUyagkiRJDSXOATUBlSRJUlMmoJIkSY15H1BJkiSpIRNQSZKkxiY7/zQBlSRJUmMmoJIkSY15H1BJkiSpIRNQSZKkhgIsmOwA1ARUkiRJbZmASpIktZQ4B3TUBUiSJGmymIBKkiQ1NuEBqAmoJEmS2po1AU3yz0DNtr6qnjOUiiRJkua5SZ8DOtcp+Hc3q0KSJEkTY9YGtKr+tWUhkiRJmgxrvQgpySLgb4B9gS2mxqvqN4dYlyRJ0rzkjegHuwjpH4CTgTXAU4GPAf84zKIkSZI0fw3SgG5ZVecDqaobquok4FnDLUuSJGn+Sncz+mEu42yQ+4Del2QBcE2SVwOrgK2HW5YkSZLmq0ES0NcCWwGvAR4H/BFw3DCLkiRJms/SYBlna01Aq+oigCQ/r6qXDL8kSZIkzWdrTUCTPCHJlcD3uvf7Jfng0CuTJEmahxJYkAx9GWeDnIJ/L3AocDtAVX0XOHiYRUmSJGn+GuQiJKrqxmlXUz0wnHIkSZLmvzEPKIdukAb0xiRPBCrJpvQuSrpquGVJkiRpvhqkAX0V8PfAbvRuwfQl4PhhFiVJkjSfjft9OodtzgY0yULgmKr6w0b1SJIkaZ6b8yKkqnoA+INGtUiSJE2EZPjLOBvkFPw3krwf+Cfg7qnBqrpkaFVJkiRp3hqkAd2/+/mWvrECnrbhy5EkSZrfwvjfp3PYBnkS0lNbFCJJkqTJMMiN6NdLkj2SfDXJlUmuSPLaYR1LkiRpo9Fg/ue4B6wD3Yh+Pa0B/qyqLkmyDXBxkvOq6sohHlOSJEljbmgNaFXdBNzUvf5xkqvo3UvUBlSSJE20Sb8P6FpPwSfZKslfJvlw935Rkmevy0GS7Ak8Bvj2DOuWJFmWZNnq225bl91KkiRpIzTIHNB/AO4DntC9XwX81aAHSLI1cCbwuqq6a/r6qjqlqhZX1eKddt550N1KkiRttBY0WMbZIPU9oqreCdwPUFU/BQbKjbtnx58JfKKqPrPeVUqSJGneGGQO6M+SbEnv3p8keQS9RHRO6U1u+ChwVVW959eqUpIkaZ4IzgEdJAE9EfgisEeSTwDnA28Y4HMHAccAT0uyvFuOWP9SJUmSNB8MciP685JcAhxIr2l/bVWtHuBz32DAU/WSJEmTZMGEd0iDXAV/EHBvVf0LsB3w5iQPH3plkiRJmpcGOQV/MvDTJPsBfwpcC3xsqFVJkiTNYwsy/GWcDdKArqmqAo4EPlBVHwC2GW5ZkiRJmq8GaUB/nORNwB8B/5JkAbDpcMuSJEman3rPas/Ql7XXkaVJbk1yed/YDknOS3JN93P7bjxJ3pdkRZJLkzy27zPHddtfk+S4Qf4MBmlAX0Tvtksvq6qbgd2Bdw2yc0mSJI2tU4HDpo2dAJxfVYvo3fnohG78cGBRtyyhN0WTJDvQu2PS44EDgBOnmta5rLUBraqbq+o9VfX17v0Pqso5oJIkSetpHOaAVtXXgDumDR8JnNa9Pg14bt/4x6rnQmC7JLsChwLnVdUdVfUj4Dx+tan91e+/tg2SHJjkoiQ/SfKzJA8kuXPtX0uSJEkjtFOSZX3LkgE+s0tV3dS9vhnYpXu9G3Bj33Yru7HZxuc0yJOQ3g8cBXwaWAwcC+w9wOckSZI0g0YPQlpdVYvX98NVVUlqQxY0ZaBn1VfVCmBhVT1QVf/AANGqJEmSNjq3dKfW6X7e2o2vAvbo2273bmy28TkN0oD+NMlmwPIk70zyJwN+TpIkSdMEWJAMfVlPZwNTV7IfB3yub/zY7mr4A4E7u1P15wKHJNm+u/jokG5sToM0kscAC4FXA3fT63Kfvy7fRJIkSeMlyenAt4B9kqxM8jLg7cAzk1wDPKN7D3AOcB2wAvgw8McAVXUH8Fbgom55Szc2p0GeBX9D9/Ie4H+vw/eSJEnSDMbhVHJVHT3LqqfPsG0Bx8+yn6XA0nU59qwNaJLLgFknnlbVb6/LgSRJkiSYOwF9drMqJEmSJkijq+DH1lwN6Kb07gX1zf7BJAfRuy+UJEmStM7mmoLwXuCuGcbv6tZJkiRpHaXBFfC/xlXwTczVgO5SVZdNH+zG9hxaRZIkSZrX5joFv90c67bc0IVIkiRNijEPKIdurgR0WZJXTB9M8nLg4uGVJEmSpPlsrgT0dcBZSf6Q/2w4FwObAc8bdmGSJEnz1YIJT0BnbUCr6hbgiUmeCjy6G/6XqvpKk8okSZI0Lw3yJKSvAl9tUIskSdK8N/Us+Ek2Dk+CkiRJ0gRZawIqSZKkDWvCA1ATUEmSJLVlAipJktRSvAreBFSSJElNmYBKkiQ1FiY7AjUBlSRJUlMmoJIkSQ317gM66ipGywRUkiRJTZmASpIkNWYCKkmSJDVkAipJktRYJvxRSCagkiRJasoEVJIkqSGvgjcBlSRJUmMmoJIkSS0FJnwKqAmoJEmS2jIBlSRJamzBhEegJqCSJElqygZUkiRJTXkKXpIkqSFvw2QCKkmSpMZMQCVJkhqb8GuQTEAlSZLU1lgloNUtGk+777DlqEvQHBb9xtajLkFzuODq20ZdgubwlH12HnUJmihhAZMdgZqASpIkqamxSkAlSZLmu+AcUBNQSZIkNWUCKkmS1FK8D6gJqCRJkpoyAZUkSWpswYRPAjUBlSRJUlMmoJIkSQ15FbwJqCRJkhozAZUkSWrMOaCSJElSQyagkiRJjU14AGoCKkmSpLZMQCVJkhoKJoCT/v0lSZLUmAmoJElSS4FM+CRQE1BJkiQ1ZQIqSZLU2GTnnyagkiRJEynJPkmW9y13JXldkpOSrOobP6LvM29KsiLJ1UkOXd9jm4BKkiQ1FMbjSUhVdTWwP0CShcAq4CzgJcDfVdW7+7dPsi9wFPAo4KHAl5PsXVUPrOuxTUAlSZL0dODaqrphjm2OBD5VVfdV1feBFcAB63MwG1BJkqTG0mABdkqyrG9ZMkdJRwGn971/dZJLkyxNsn03thtwY982K7uxdWYDKkmSND+trqrFfcspM22UZDPgOcCnu6GTgUfQOz1/E/C3G7ow54BKkiQ1NgZTQPsdDlxSVbcATP0ESPJh4PPd21XAHn2f270bW2cmoJIkSZPtaPpOvyfZtW/d84DLu9dnA0cl2TzJXsAi4N/W54AmoJIkSU1lbJ6ElORBwDOBV/YNvzPJ/kAB10+tq6orkpwBXAmsAY5fnyvgwQZUkiRpYlXV3cCO08aOmWP7twFv+3WPawMqSZLUUHAO5KR/f0mSJDVmAipJktTYuMwBHRUTUEmSJDVlAipJktTYZOefJqCSJElqzARUkiSppTgH1ARUkiRJTZmASpIkNeR9QP3+kiRJaswEVJIkqTHngEqSJEkNmYBKkiQ1Ntn5pwmoJEmSGjMBlSRJamzCp4CagEqSJKktE1BJkqSGevcBnewI1ARUkiRJTZmASpIkNeYcUEmSJKkhE1BJkqSmQpwDOlxJFib5TpLPD/tYkiRJGn8tEtDXAlcB2zY4liRJ0thzDugQJdkdeBbwkWEeR5IkSRuPYSeg7wXeAGwz2wZJlgBLAPbY42FDLkeSJGm0vA/oEBPQJM8Gbq2qi+farqpOqarFVbV4x513HlY5kiRJGhPDTEAPAp6T5AhgC2DbJP9YVX80xGNKkiSNtzgHdGgJaFW9qap2r6o9gaOAr9h8SpIkyfuASpIkNTbpCWiTBrSqLgAuaHEsSZIkjTcTUEmSpMZ8EpIkSZLUkAmoJElSQwEWTHYAagIqSZKktmxAJUmS1JSn4CVJkhrzIiRJkiSpIRNQSZKkxib9RvQmoJIkSWrKBFSSJKkx54BKkiRJDZmASpIkNeSN6E1AJUmS1JgJqCRJUlNxDuioC5AkSdJkMQGVJElqKd4H1ARUkiRJTZmASpIkNTbhAagJqCRJktoyAZUkSWqodx/Qyc5ATUAlSZLUlAmoJElSY5Odf5qCTsf6AAAOxElEQVSASpIkqTEbUEmSpNbSYBmkjOT6JJclWZ5kWTe2Q5LzklzT/dy+G0+S9yVZkeTSJI9d369vAypJkjTZnlpV+1fV4u79CcD5VbUIOL97D3A4sKhblgAnr+8BbUAlSZIaS4N/fg1HAqd1r08Dnts3/rHquRDYLsmu63MAG1BJkqT5aacky/qWJTNsU8CXklzct36Xqrqpe30zsEv3ejfgxr7PruzG1plXwUuSJDXW6Dagq/tOq8/mSVW1KslDgPOSfK9/ZVVVktrQhZmASpIkTaiqWtX9vBU4CzgAuGXq1Hr389Zu81XAHn0f370bW2c2oJIkSY2Nw0XwSR6UZJup18AhwOXA2cBx3WbHAZ/rXp8NHNtdDX8gcGffqfp14il4SZKkybQLcFZ68wE2AT5ZVV9MchFwRpKXATcAL+y2Pwc4AlgB/BR4yfoe2AZUkiSptTF4FFJVXQfsN8P47cDTZxgv4PgNcWxPwUuSJKkpE1BJkqSGenM0xyACHSETUEmSJDVlAipJktRSmt0HdGyZgEqSJKkpE1BJkqTGJjwANQGVJElSWyagkiRJrU14BGoCKkmSpKZMQCVJkpqK9wEddQGSJEmaLCagkiRJjXkfUEmSJKkhE1BJkqSGwsRfBG8DqsFttbn/c5HW15MeueOoS9Ac7r3/gVGXoFn8vEZdgYbBjkKSJKm1CY9AnQMqSZKkpkxAJUmSGvM+oJIkSVJDJqCSJEmNeR9QSZIkqSETUEmSpMYmPAA1AZUkSVJbJqCSJEkt+SgkE1BJkiS1ZQIqSZLUmPcBlSRJkhoyAZUkSWooeB9QE1BJkiQ1ZQIqSZLU2IQHoCagkiRJassEVJIkqbUJj0BNQCVJktSUCagkSVJj3gdUkiRJasgEVJIkqTHvAypJkiQ1ZAIqSZLU2IQHoCagkiRJassGVJIkSU15Cl6SJKm1CT8HbwIqSZKkpkxAJUmSGgreiN4EVJIkSU2ZgEqSJLUUb0RvAipJkqSmTEAlSZIam/AA1ARUkiRJbZmASpIktTbhEagJqCRJkpoyAZUkSWoq3gd01AVIkiRpspiASpIkNeZ9QCVJkjRxkuyR5KtJrkxyRZLXduMnJVmVZHm3HNH3mTclWZHk6iSHru+xTUAlSZIaCmNzEfwa4M+q6pIk2wAXJzmvW/d3VfXu/o2T7AscBTwKeCjw5SR7V9UD63pgE1BJkqQJVFU3VdUl3esfA1cBu83xkSOBT1XVfVX1fWAFcMD6HNsGVJIkqbU0WGCnJMv6liWzlpPsCTwG+HY39OoklyZZmmT7bmw34Ma+j61k7oZ1VjagkiRJ89Pqqlrct5wy00ZJtgbOBF5XVXcBJwOPAPYHbgL+dkMX5hxQSZKkxsblPqBJNqXXfH6iqj4DUFW39K3/MPD57u0qYI++j+/eja0zE1BJkqQJlCTAR4Grquo9feO79m32PODy7vXZwFFJNk+yF7AI+Lf1ObYJqCRJUmNjch/Qg4BjgMuSLO/G3gwcnWR/oIDrgVcCVNUVSc4ArqR3Bf3x63MFPNiASpIkTaSq+gYz3xHqnDk+8zbgbb/usW1AJUmSGhuPAHR0htqAJrke+DHwALCmqhYP83iSJEkafy0S0KdW1eoGx5EkSRp/GZs5oCPjVfCSJElqatgNaAFfSnLxbHffT7Jk6g79t99225DLkSRJGgdtHoU0robdgD6pqh4LHA4cn+Tg6RtU1SlTd+jfceedh1yOJEmSRm2oDWhVrep+3gqcxXo+sF6SJGm+CL05oMNextnQGtAkD0qyzdRr4BD+8076kiRJmlDDvAp+F+Cs3lOe2AT4ZFV9cYjHkyRJ2iiMeUA5dENrQKvqOmC/Ye1fkiRJGyefhCRJktTYuM/RHDbvAypJkqSmTEAlSZIay4TPAjUBlSRJUlMmoJIkSa1NdgBqAipJkqS2TEAlSZIam/AA1ARUkiRJbZmASpIkNbQxPKt92ExAJUmS1JQJqCRJUmPeB1SSJElqyARUkiSptckOQE1AJUmS1JYJqCRJUmMTHoCagEqSJKktE1BJkqTGvA+oJEmS1JAJqCRJUlPxPqCjLkCSJEmTxQRUkiSpoeAcUBNQSZIkNWUDKkmSpKZsQCVJktSUc0AlSZIacw6oJEmS1JAJqCRJUmPeB1SSJElqyARUkiSppTgH1ARUkiRJTZmASpIkNZRumWQmoJIkSWrKBlSSJElNeQpekiSptQk/B28CKkmSpKZMQCVJkhrzRvSSJElSQyagkiRJjXkjekmSJKkhE1BJkqTGJjwANQGVJElSWyagkiRJrU14BGoCKkmSpKZsQCVJkhpLg38GqiM5LMnVSVYkOWHIX/sXbEAlSZImUJKFwAeAw4F9gaOT7Nvi2M4BlSRJaiiMzX1ADwBWVNV1AEk+BRwJXDnsA49VA7r8kotXP3jLhTeMuo4NaCdg9aiL0Kz8/Ywvfzfjzd/PeJtvv5+Hj7qADe2SSy4+d8tNs1ODQ22RZFnf+1Oq6pS+97sBN/a9Xwk8vkFd49WAVtXOo65hQ0qyrKoWj7oOzczfz/jydzPe/P2MN38/46+qDht1DaPmHFBJkqTJtArYo+/97t3Y0NmASpIkTaaLgEVJ9kqyGXAUcHaLA4/VKfh56JS1b6IR8vczvvzdjDd/P+PN348GUlVrkrwaOBdYCCytqitaHDtV1eI4kiRJEuApeEmSJDVmAypJkqSmbEAlSeskGZNbaEvaaNmAbkBJ9knyhCSbdo+30hjydzOekjwyyeIkm4+6Fv2qJE9KcgxAVZVN6HhJ8t+SvHbUdUiD8ir4DSTJ7wF/Te/+WauAZUlOraq7RluZpiTZu6r+vaoeSLKwqh4YdU3qSfJsev/+3A7cnOTEqvr3EZclIMkCYCvg//Te5kFV9aGuCV1QVT8fcYkTL8khwFuBPx91LdKgTEA3gCSbAi8CXlZVTwc+R+/Grm9Msu1IixPwiwZneZJPAkw1oSMuS0CSJwLvAo6rqqcCPwJOGG1VmlJVP6+qnwCnAR8FnpjkT6bWjbQ4Tf3783FgSVWdl+TBSR6eZKtR1ybNxQZ0w9kWWNS9Pgv4PLAp8AeeqhqtJA8CXg28DvhZkn8Em9Ax846q+k73+kRgB0/Fj5019P6P9WnAAUnek+Rv0uPfJaNzO3A/sGuSHYHPAicDpyb5ff/+0bjyPxobQFXdD7wH+L0kT+5SgW8Ay4EnjbQ4UVV3Ay8FPgm8HtiivwkdZW0C4NvAZ+AX83M3Bx5O7//U0f2lqtH7HHBzVZ0PLANeBWxbPSahI1JVVwPPAv4O+C69/849G/gi8Hxg+9FVJ83OBnTD+TrwJeCYJAdX1QNV9UngocB+oy1NVfXDqvpJVa0GXglsOdWEJnlskt8abYWTq/t3ZWqudID/AO6oqtuS/CHwV0m2HF2F6twD7JPkFfSaz7cDD0vyytGWpar6Lr2m8+1V9eFu2sRSes3nw0ZbnTQzL0LaQKrq3iSfAAp4U9fQ3AfsAtw00uL0S6rq9u4vzXcl+R69x489dcRlid5j4YCfJLkxyd8AhwAvrqp7RlzaxKuqHya5EfhL4Piq+uckTwVWjLg0AVV1JXDl1Pskzwd2xr9/NKZ8FOcGlmQz4CB6Kdu9wN/3zW3TGOkupHgj8MyqumzU9egX95fcFLiq+/n0qrpmtFVpSpI9gIdU1cXde6+CHzPdv0MvoTfd6AWtnustrSsb0CHp5rI5N2pMJdkeOAP4s6q6dNT16JcleTFwkX95jqckKf/yGEtdA/q79Obrfm/U9UizsQHVxEqyRVXdO+o69KtscCRpfrMBlSRJUlNeBS9JkqSmbEAlSZLUlA2oJEmSmrIBlSRJUlM2oJJI8htJPpXk2iQXJzknyd5J9kxy+ajrG1SSU5P8fvf6I0n2nWPbk5K8vl11kqQpPglJmnDdfQPPAk6rqqO6sf3oPcXrxlHW1tWySfeEpHVSVS8fRj2SpF+fCaikpwL3V9WHpgaq6rtV9fX+jbo09OtJLumWJ3bjuyb5WpLlSS5P8uRu/JAk3+q2/XSSracfOMkFSf6+77MHdOMnJfl4km8CH0+yMMm7klyU5NKp54+n5/1Jrk7yZeAh0/a9uHt9WFfHd5Oc31fCvt121yV5Td9nP9slwVckWfLr/xFLkvqZgEp6NHDxANvdSu+xpfcmWQScDiwG/gA4t6re1j0BbKskOwF/ATyjqu5O8kbgT4G3zLDfrapq/yQHA0u7egD2BZ5UVfd0TeCdVfU7STYHvpnkS8BjgH26bXeh9yzspf07T7Iz8GHg4Kr6fpId+lb/Fr0GfBvg6iQnV9X9wEur6o4kWwIXJTmzqm4f4M9IkjQAG1BJg9oUeH+S/YEHgL278YuApUk2BT5bVcuT/C69pvCbvTP8bAZ8a5b9ng5QVV9Lsm2S7brxs6vqnu71IcBvT83vBB4MLAIOBk6vqgeAHyb5ygz7PxD4WlV9vzvOHX3r/qWq7gPuS3IrvSZ2JfCaJM/rttmjO5YNqCRtIDagkq4Afn+tW8GfALcA+9GbvnMv/KJxPBh4FnBqkvcAPwLOq6qjB9jv9MexTb2/u28swP+oqnP7N0xyxAD7n8t9fa8fADZJ8hTgGcATquqnSS4Atvg1jyNJ6uMcUElfATbvn+uY5Len5nL2eTBwU1X9HDgGWNht+3Dglqr6MPAR4LHAhcBBSR7ZbfOgJHszsxd12zyJ3mn2O2fY5lzgv3cpK90V+g8Cvga8qJsjuiu90+nTXQgcnGSv7rM7zLDN9O/5o675/C16CaokaQMyAZUmXFVVd7r5vd1czXuB64HXTdv0g8CZSY4Fvsh/JpRPAf48yf3AT4Bjq+q2JC8GTu/mbEJvTui/z1DCvUm+Q+8U/0tnKfMjwJ7AJd1V+7cBz6V39f7T6M39/AEznObvalkCfCbJArq5rLP+gfS+26uSXAVcTa+BlSRtQKmafvZLktroTm+/vqqWjboWSVI7noKXJElSUyagkiRJasoEVJIkSU3ZgEqSJKkpG1BJkiQ1ZQMqSZKkpmxAJUmS1NT/B/LcrnI9lTHDAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"#############################\")\n",
    "excercise_2(all_data_1S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_2(all_data_2S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_2(all_data_5S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_2(all_data_10S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#############################\n",
      "X_train_3: (42313, 8)\n",
      "X_test_3: (15635, 8)\n",
      "y_train_3: (42313,)\n",
      "y_test_3: (15635,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[1677  934  838  132   16]\n",
      " [ 259 4249  455   27    0]\n",
      " [ 139  865 2496    6    0]\n",
      " [  49   76   13 2767    4]\n",
      " [  13    0    0    7  613]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.78      0.47      0.58      3597\n",
      "         1.0       0.69      0.85      0.76      4990\n",
      "         2.0       0.66      0.71      0.68      3506\n",
      "         4.0       0.94      0.95      0.95      2909\n",
      "         5.0       0.97      0.97      0.97       633\n",
      "\n",
      "   micro avg       0.75      0.75      0.75     15635\n",
      "   macro avg       0.81      0.79      0.79     15635\n",
      "weighted avg       0.76      0.75      0.75     15635\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.7548448992644707\n",
      "#############################\n",
      "X_train_3: (21153, 8)\n",
      "X_test_3: (7815, 8)\n",
      "y_train_3: (21153,)\n",
      "y_test_3: (7815,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[ 924  392  391   75   14]\n",
      " [ 132 2082  236   18    0]\n",
      " [  75  438 1257    7    0]\n",
      " [  36   44    7 1369    2]\n",
      " [   5    1    0    6  304]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.79      0.51      0.62      1796\n",
      "         1.0       0.70      0.84      0.77      2468\n",
      "         2.0       0.66      0.71      0.69      1777\n",
      "         4.0       0.93      0.94      0.93      1458\n",
      "         5.0       0.95      0.96      0.96       316\n",
      "\n",
      "   micro avg       0.76      0.76      0.76      7815\n",
      "   macro avg       0.81      0.79      0.79      7815\n",
      "weighted avg       0.77      0.76      0.75      7815\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.7595649392194498\n",
      "#############################\n",
      "X_train_3: (21153, 8)\n",
      "X_test_3: (7815, 8)\n",
      "y_train_3: (21153,)\n",
      "y_test_3: (7815,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[1216  238  275   56   11]\n",
      " [ 157 2016  253   18    0]\n",
      " [  61  642 1072   19    0]\n",
      " [  53   45    6 1351    2]\n",
      " [  10    2    0    4  308]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.81      0.68      0.74      1796\n",
      "         1.0       0.69      0.82      0.75      2444\n",
      "         2.0       0.67      0.60      0.63      1794\n",
      "         4.0       0.93      0.93      0.93      1457\n",
      "         5.0       0.96      0.95      0.96       324\n",
      "\n",
      "   micro avg       0.76      0.76      0.76      7815\n",
      "   macro avg       0.81      0.80      0.80      7815\n",
      "weighted avg       0.77      0.76      0.76      7815\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.763019833653231\n",
      "#############################\n",
      "X_train_3: (21153, 8)\n",
      "X_test_3: (7815, 8)\n",
      "y_train_3: (21153,)\n",
      "y_test_3: (7815,)\n",
      "\n",
      " _______________________________________________________\n",
      "confusion matrix: \n",
      "\n",
      "[[1352  207  190   53   14]\n",
      " [ 183 2172  176    9    0]\n",
      " [ 100  779  757   13    5]\n",
      " [  58   59   14 1321    4]\n",
      " [  10    4    0    2  333]]\n",
      "\n",
      " _______________________________________________________\n",
      "classification report: \n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.79      0.74      0.77      1816\n",
      "         1.0       0.67      0.86      0.75      2540\n",
      "         2.0       0.67      0.46      0.54      1654\n",
      "         4.0       0.94      0.91      0.93      1456\n",
      "         5.0       0.94      0.95      0.94       349\n",
      "\n",
      "   micro avg       0.76      0.76      0.76      7815\n",
      "   macro avg       0.80      0.78      0.79      7815\n",
      "weighted avg       0.76      0.76      0.75      7815\n",
      "\n",
      "_______________________________________________________\n",
      "accuracy score: 0.7594369801663468\n",
      "#############################\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"#############################\")\n",
    "excercise_3(all_data_1S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_3(all_data_2S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_3(all_data_5S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "excercise_3(all_data_10S_filtered, 50)\n",
    "print(\"#############################\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusions:\n",
    "\n",
    "### According precision and recall metrics, the best answer are with method 1 , where training and test data are taken radomly, by the way the best results appear when the time analyzed is 10 seconds , probably because the activities sequence going the best results over a longer time. \n",
    "\n",
    "### Maybe this metric would be reduced if you take more than 10 seconds or as it`s showing in this report you can reduce the effectiveness if you take a litlle less bit time as well.\n",
    "\n",
    "### If only one place could be selected to put the senssor according to the correlation matrix I would place it on back, which one has a highest level of correlation with other variables (related with acelerometer and innertian) an then this one can give me reliable information , but that would be useful for this analysis. \n",
    "\n",
    "\n",
    "### If the smartphone changed it´s orientation , I suposse the answers of this study doesn´t change , because the sensors of the smartphone (accelerometer, gyroscope, magnetometer), can tell you the position of the phone."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}