Assignment3_MLST-Final.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "assert sys.version_info >= (3, 5)\n",
    "import os\n",
    "import sklearn\n",
    "assert sklearn.__version__ >= \"0.20\"\n",
    "# To plot figures\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "mpl.rc('axes', labelsize=14)\n",
    "mpl.rc('xtick', labelsize=12)\n",
    "mpl.rc('ytick', labelsize=12)\n",
    "# Where to save the figures\n",
    "PROJECT_ROOT_DIR = \".\"\n",
    "CHAPTER_ID = \"Assignment3\"\n",
    "IMAGES_PATH = os.path.join(PROJECT_ROOT_DIR, \"images\", CHAPTER_ID)\n",
    "os.makedirs(IMAGES_PATH, exist_ok=True)\n",
    "def save_fig(fig_id, tight_layout=True, fig_extension=\"png\", resolution=300):\n",
    "    path = os.path.join(IMAGES_PATH, fig_id + \".\" + fig_extension)\n",
    "    print(\"Saving figure\", fig_id)\n",
    "    if tight_layout:\n",
    "        plt.tight_layout()\n",
    "    plt.savefig(path, format=fig_extension, dpi=resolution)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set(color_codes=True)\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from scipy import stats\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.metrics import confusion_matrix,accuracy_score,precision_score,recall_score,f1_score\n",
    "from sklearn.model_selection import cross_validate,GridSearchCV,RandomizedSearchCV\n",
    "#from tpot import TPOTClassifier\n",
    "from scipy.stats import randint\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Importing data from csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Load data from CSV files\n",
    "def load_data(_path):\n",
    "    csv_path = os.path.join(\"archive\", _path)\n",
    "    return pd.read_csv(csv_path, header=None, sep=\"\\t\")\n",
    "\n",
    "d_PS1 = load_data(\"PS1.txt\")\n",
    "d_PS2 = load_data(\"PS2.txt\")\n",
    "d_PS3 = load_data(\"PS3.txt\")\n",
    "d_PS4 = load_data(\"PS4.txt\")\n",
    "d_PS5 = load_data(\"PS5.txt\")\n",
    "d_PS6 = load_data(\"PS6.txt\")\n",
    "\n",
    "#d_EPS1 = load_data(\"EPS1.txt\")\n",
    "\n",
    "#d_FS1 = load_data(\"FS1.txt\")\n",
    "#d_FS2 = load_data(\"FS2.txt\")\n",
    "\n",
    "#d_TS1 = load_data(\"TS1.txt\")\n",
    "#d_TS2 = load_data(\"TS2.txt\")\n",
    "#d_TS3 = load_data(\"TS3.txt\")\n",
    "#d_TS4 = load_data(\"TS4.txt\")\n",
    "\n",
    "#d_VS1 = load_data(\"VS1.txt\")\n",
    "\n",
    "#d_CE = load_data(\"CE.txt\")\n",
    "#d_CP = load_data(\"CP.txt\")\n",
    "#d_SE = load_data(\"SE.txt\")\n",
    "\n",
    "d_PROFILE = load_data(\"profile.txt\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#d_PS1.T.iloc[0:,1:3].plot()\n",
    "#d_PS1.T.iloc[0:,501:503].plot()\n",
    "#d_PS1.T.iloc[0:,651:654].plot()\n",
    "#d_PS1.T.iloc[0:,1001:1003].plot()\n",
    "#save_fig(\"pressure_plots\")\n",
    "#plt.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#d_PS1.T.iloc[0:,0:700].plot(figsize=(15,15))\n",
    "#save_fig(\"pressure_plots\")\n",
    "#plt.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Preperations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "ps1_mean=d_PS1.mean(axis=1)\n",
    "ps2_mean=d_PS2.mean(axis=1)\n",
    "ps3_mean=d_PS3.mean(axis=1)\n",
    "ps4_mean=d_PS4.mean(axis=1)\n",
    "ps5_mean=d_PS5.mean(axis=1)\n",
    "ps6_mean=d_PS6.mean(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "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>PS1</th>\n",
       "      <th>PS2</th>\n",
       "      <th>PS3</th>\n",
       "      <th>PS4</th>\n",
       "      <th>PS5</th>\n",
       "      <th>PS6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>160.673492</td>\n",
       "      <td>109.466914</td>\n",
       "      <td>1.991475</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.842170</td>\n",
       "      <td>9.728097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>160.603320</td>\n",
       "      <td>109.354890</td>\n",
       "      <td>1.976234</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.635142</td>\n",
       "      <td>9.529488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>160.347720</td>\n",
       "      <td>109.158845</td>\n",
       "      <td>1.972224</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.530548</td>\n",
       "      <td>9.427949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>160.188088</td>\n",
       "      <td>109.064807</td>\n",
       "      <td>1.946575</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.438827</td>\n",
       "      <td>9.337430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>160.000472</td>\n",
       "      <td>108.931434</td>\n",
       "      <td>1.922707</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.358762</td>\n",
       "      <td>9.260636</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          PS1         PS2       PS3  PS4       PS5       PS6\n",
       "0  160.673492  109.466914  1.991475  0.0  9.842170  9.728097\n",
       "1  160.603320  109.354890  1.976234  0.0  9.635142  9.529488\n",
       "2  160.347720  109.158845  1.972224  0.0  9.530548  9.427949\n",
       "3  160.188088  109.064807  1.946575  0.0  9.438827  9.337430\n",
       "4  160.000472  108.931434  1.922707  0.0  9.358762  9.260636"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pressure=pd.DataFrame({\"PS1\":ps1_mean,\"PS2\":ps2_mean,\"PS3\":ps3_mean,\"PS4\":ps4_mean,\"PS5\":ps5_mean,\"PS6\":ps6_mean})\n",
    "df_pressure.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_pressure.plot(subplots=True,figsize=(15,15))\n",
    "#save_fig(\"pressure_plots\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        0.000000\n",
       "1        0.000000\n",
       "2        0.000000\n",
       "3        0.000000\n",
       "4        0.000000\n",
       "          ...    \n",
       "2200    10.202473\n",
       "2201    10.197919\n",
       "2202    10.196824\n",
       "2203    10.198588\n",
       "2204    10.203126\n",
       "Length: 2205, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#At first it looked like PS4 only contained close to 0 values, but after checking the mean i concluded that i needed to keep it.\n",
    "#df.drop([\"PS4\"], axis=1)\n",
    "ps4_mean\n",
    "#This drift from 0 to 10 over the course of 2204 runs might indicate that something is drifting from normal conditions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2200</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2201</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2202</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2203</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2204</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2205 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        0    1  2    3  4\n",
       "0       3  100  0  130  1\n",
       "1       3  100  0  130  1\n",
       "2       3  100  0  130  1\n",
       "3       3  100  0  130  1\n",
       "4       3  100  0  130  1\n",
       "...   ...  ... ..  ... ..\n",
       "2200  100  100  0   90  0\n",
       "2201  100  100  0   90  0\n",
       "2202  100  100  0   90  0\n",
       "2203  100  100  0   90  0\n",
       "2204  100  100  0   90  0\n",
       "\n",
       "[2205 rows x 5 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking the profile shape, (.shape not needed in jupyter notebook)\n",
    "d_PROFILE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>Cooler_Condition</th>\n",
       "      <th>Valve_Condition</th>\n",
       "      <th>Internal_Pump_Leakage</th>\n",
       "      <th>Hydraulic_Accumulator</th>\n",
       "      <th>Stable_Flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2200</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2201</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2202</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2203</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2204</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2205 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Cooler_Condition  Valve_Condition  Internal_Pump_Leakage  \\\n",
       "0                    3              100                      0   \n",
       "1                    3              100                      0   \n",
       "2                    3              100                      0   \n",
       "3                    3              100                      0   \n",
       "4                    3              100                      0   \n",
       "...                ...              ...                    ...   \n",
       "2200               100              100                      0   \n",
       "2201               100              100                      0   \n",
       "2202               100              100                      0   \n",
       "2203               100              100                      0   \n",
       "2204               100              100                      0   \n",
       "\n",
       "      Hydraulic_Accumulator  Stable_Flag  \n",
       "0                       130            1  \n",
       "1                       130            1  \n",
       "2                       130            1  \n",
       "3                       130            1  \n",
       "4                       130            1  \n",
       "...                     ...          ...  \n",
       "2200                     90            0  \n",
       "2201                     90            0  \n",
       "2202                     90            0  \n",
       "2203                     90            0  \n",
       "2204                     90            0  \n",
       "\n",
       "[2205 rows x 5 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_profile=pd.DataFrame({\"Cooler_Condition\":d_PROFILE.iloc[:,-5],\"Valve_Condition\":d_PROFILE.iloc[:,-4],\n",
    "                         \"Internal_Pump_Leakage\":d_PROFILE.iloc[:,-3],\"Hydraulic_Accumulator\":d_PROFILE.iloc[:,-2],\n",
    "                         \"Stable_Flag\":d_PROFILE.iloc[:,-1]})\n",
    "df_profile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "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>PS1</th>\n",
       "      <th>PS2</th>\n",
       "      <th>PS3</th>\n",
       "      <th>PS4</th>\n",
       "      <th>PS5</th>\n",
       "      <th>PS6</th>\n",
       "      <th>Cooler_Condition</th>\n",
       "      <th>Valve_Condition</th>\n",
       "      <th>Internal_Pump_Leakage</th>\n",
       "      <th>Hydraulic_Accumulator</th>\n",
       "      <th>Stable_Flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>160.673492</td>\n",
       "      <td>109.466914</td>\n",
       "      <td>1.991475</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.842170</td>\n",
       "      <td>9.728097</td>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>160.603320</td>\n",
       "      <td>109.354890</td>\n",
       "      <td>1.976234</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.635142</td>\n",
       "      <td>9.529488</td>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>160.347720</td>\n",
       "      <td>109.158845</td>\n",
       "      <td>1.972224</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.530548</td>\n",
       "      <td>9.427949</td>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>160.188088</td>\n",
       "      <td>109.064807</td>\n",
       "      <td>1.946575</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.438827</td>\n",
       "      <td>9.337430</td>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>160.000472</td>\n",
       "      <td>108.931434</td>\n",
       "      <td>1.922707</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.358762</td>\n",
       "      <td>9.260636</td>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2200</th>\n",
       "      <td>161.227572</td>\n",
       "      <td>109.779581</td>\n",
       "      <td>2.001438</td>\n",
       "      <td>10.202473</td>\n",
       "      <td>9.972037</td>\n",
       "      <td>9.850361</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2201</th>\n",
       "      <td>161.206070</td>\n",
       "      <td>109.787481</td>\n",
       "      <td>1.998781</td>\n",
       "      <td>10.197919</td>\n",
       "      <td>9.966184</td>\n",
       "      <td>9.844854</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2202</th>\n",
       "      <td>161.192120</td>\n",
       "      <td>109.756174</td>\n",
       "      <td>1.993436</td>\n",
       "      <td>10.196824</td>\n",
       "      <td>9.964329</td>\n",
       "      <td>9.842628</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2203</th>\n",
       "      <td>161.208917</td>\n",
       "      <td>109.793884</td>\n",
       "      <td>2.007077</td>\n",
       "      <td>10.198588</td>\n",
       "      <td>9.968232</td>\n",
       "      <td>9.846690</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2204</th>\n",
       "      <td>161.217128</td>\n",
       "      <td>109.792177</td>\n",
       "      <td>2.002690</td>\n",
       "      <td>10.203126</td>\n",
       "      <td>9.973638</td>\n",
       "      <td>9.851949</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2205 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             PS1         PS2       PS3        PS4       PS5       PS6  \\\n",
       "0     160.673492  109.466914  1.991475   0.000000  9.842170  9.728097   \n",
       "1     160.603320  109.354890  1.976234   0.000000  9.635142  9.529488   \n",
       "2     160.347720  109.158845  1.972224   0.000000  9.530548  9.427949   \n",
       "3     160.188088  109.064807  1.946575   0.000000  9.438827  9.337430   \n",
       "4     160.000472  108.931434  1.922707   0.000000  9.358762  9.260636   \n",
       "...          ...         ...       ...        ...       ...       ...   \n",
       "2200  161.227572  109.779581  2.001438  10.202473  9.972037  9.850361   \n",
       "2201  161.206070  109.787481  1.998781  10.197919  9.966184  9.844854   \n",
       "2202  161.192120  109.756174  1.993436  10.196824  9.964329  9.842628   \n",
       "2203  161.208917  109.793884  2.007077  10.198588  9.968232  9.846690   \n",
       "2204  161.217128  109.792177  2.002690  10.203126  9.973638  9.851949   \n",
       "\n",
       "      Cooler_Condition  Valve_Condition  Internal_Pump_Leakage  \\\n",
       "0                    3              100                      0   \n",
       "1                    3              100                      0   \n",
       "2                    3              100                      0   \n",
       "3                    3              100                      0   \n",
       "4                    3              100                      0   \n",
       "...                ...              ...                    ...   \n",
       "2200               100              100                      0   \n",
       "2201               100              100                      0   \n",
       "2202               100              100                      0   \n",
       "2203               100              100                      0   \n",
       "2204               100              100                      0   \n",
       "\n",
       "      Hydraulic_Accumulator  Stable_Flag  \n",
       "0                       130            1  \n",
       "1                       130            1  \n",
       "2                       130            1  \n",
       "3                       130            1  \n",
       "4                       130            1  \n",
       "...                     ...          ...  \n",
       "2200                     90            0  \n",
       "2201                     90            0  \n",
       "2202                     90            0  \n",
       "2203                     90            0  \n",
       "2204                     90            0  \n",
       "\n",
       "[2205 rows x 11 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.concat([df_pressure,df_profile], axis=1)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2205 entries, 0 to 2204\n",
      "Data columns (total 11 columns):\n",
      " #   Column                 Non-Null Count  Dtype  \n",
      "---  ------                 --------------  -----  \n",
      " 0   PS1                    2205 non-null   float64\n",
      " 1   PS2                    2205 non-null   float64\n",
      " 2   PS3                    2205 non-null   float64\n",
      " 3   PS4                    2205 non-null   float64\n",
      " 4   PS5                    2205 non-null   float64\n",
      " 5   PS6                    2205 non-null   float64\n",
      " 6   Cooler_Condition       2205 non-null   int64  \n",
      " 7   Valve_Condition        2205 non-null   int64  \n",
      " 8   Internal_Pump_Leakage  2205 non-null   int64  \n",
      " 9   Hydraulic_Accumulator  2205 non-null   int64  \n",
      " 10  Stable_Flag            2205 non-null   int64  \n",
      "dtypes: float64(6), int64(5)\n",
      "memory usage: 189.6 KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>PS1</th>\n",
       "      <th>PS2</th>\n",
       "      <th>PS3</th>\n",
       "      <th>PS4</th>\n",
       "      <th>PS5</th>\n",
       "      <th>PS6</th>\n",
       "      <th>Cooler_Condition</th>\n",
       "      <th>Valve_Condition</th>\n",
       "      <th>Internal_Pump_Leakage</th>\n",
       "      <th>Hydraulic_Accumulator</th>\n",
       "      <th>Stable_Flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "      <td>2205.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>160.485315</td>\n",
       "      <td>109.379906</td>\n",
       "      <td>1.753227</td>\n",
       "      <td>2.600266</td>\n",
       "      <td>9.163320</td>\n",
       "      <td>9.079363</td>\n",
       "      <td>41.240816</td>\n",
       "      <td>90.693878</td>\n",
       "      <td>0.669388</td>\n",
       "      <td>107.199546</td>\n",
       "      <td>0.342857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.699425</td>\n",
       "      <td>4.986585</td>\n",
       "      <td>0.251902</td>\n",
       "      <td>4.279355</td>\n",
       "      <td>0.576296</td>\n",
       "      <td>0.549589</td>\n",
       "      <td>42.383143</td>\n",
       "      <td>10.681802</td>\n",
       "      <td>0.817233</td>\n",
       "      <td>16.435848</td>\n",
       "      <td>0.474772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>155.391547</td>\n",
       "      <td>104.406307</td>\n",
       "      <td>0.840252</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.365800</td>\n",
       "      <td>8.321527</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>73.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>158.100195</td>\n",
       "      <td>106.962382</td>\n",
       "      <td>1.729733</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.547239</td>\n",
       "      <td>8.487167</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>158.960895</td>\n",
       "      <td>107.730169</td>\n",
       "      <td>1.779631</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.115781</td>\n",
       "      <td>9.031516</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>161.000735</td>\n",
       "      <td>109.421612</td>\n",
       "      <td>1.932047</td>\n",
       "      <td>3.503266</td>\n",
       "      <td>9.844351</td>\n",
       "      <td>9.729275</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>130.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>180.922708</td>\n",
       "      <td>131.589089</td>\n",
       "      <td>2.023398</td>\n",
       "      <td>10.207068</td>\n",
       "      <td>9.978510</td>\n",
       "      <td>9.856591</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>130.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               PS1          PS2          PS3          PS4          PS5  \\\n",
       "count  2205.000000  2205.000000  2205.000000  2205.000000  2205.000000   \n",
       "mean    160.485315   109.379906     1.753227     2.600266     9.163320   \n",
       "std       4.699425     4.986585     0.251902     4.279355     0.576296   \n",
       "min     155.391547   104.406307     0.840252     0.000000     8.365800   \n",
       "25%     158.100195   106.962382     1.729733     0.000000     8.547239   \n",
       "50%     158.960895   107.730169     1.779631     0.000000     9.115781   \n",
       "75%     161.000735   109.421612     1.932047     3.503266     9.844351   \n",
       "max     180.922708   131.589089     2.023398    10.207068     9.978510   \n",
       "\n",
       "               PS6  Cooler_Condition  Valve_Condition  Internal_Pump_Leakage  \\\n",
       "count  2205.000000       2205.000000      2205.000000            2205.000000   \n",
       "mean      9.079363         41.240816        90.693878               0.669388   \n",
       "std       0.549589         42.383143        10.681802               0.817233   \n",
       "min       8.321527          3.000000        73.000000               0.000000   \n",
       "25%       8.487167          3.000000        80.000000               0.000000   \n",
       "50%       9.031516         20.000000       100.000000               0.000000   \n",
       "75%       9.729275        100.000000       100.000000               1.000000   \n",
       "max       9.856591        100.000000       100.000000               2.000000   \n",
       "\n",
       "       Hydraulic_Accumulator  Stable_Flag  \n",
       "count            2205.000000  2205.000000  \n",
       "mean              107.199546     0.342857  \n",
       "std                16.435848     0.474772  \n",
       "min                90.000000     0.000000  \n",
       "25%                90.000000     0.000000  \n",
       "50%               100.000000     0.000000  \n",
       "75%               130.000000     1.000000  \n",
       "max               130.000000     1.000000  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PS1                        22.084591\n",
       "PS2                        24.866035\n",
       "PS3                         0.063454\n",
       "PS4                        18.312876\n",
       "PS5                         0.332117\n",
       "PS6                         0.302048\n",
       "Cooler_Condition         1796.330820\n",
       "Valve_Condition           114.100893\n",
       "Internal_Pump_Leakage       0.667869\n",
       "Hydraulic_Accumulator     270.137114\n",
       "Stable_Flag                 0.225408\n",
       "dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.var()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>PS1</th>\n",
       "      <th>PS2</th>\n",
       "      <th>PS3</th>\n",
       "      <th>PS4</th>\n",
       "      <th>PS5</th>\n",
       "      <th>PS6</th>\n",
       "      <th>Cooler_Condition</th>\n",
       "      <th>Valve_Condition</th>\n",
       "      <th>Internal_Pump_Leakage</th>\n",
       "      <th>Hydraulic_Accumulator</th>\n",
       "      <th>Stable_Flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PS1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.995294</td>\n",
       "      <td>-0.719679</td>\n",
       "      <td>0.043416</td>\n",
       "      <td>-0.125801</td>\n",
       "      <td>-0.124423</td>\n",
       "      <td>-0.002000</td>\n",
       "      <td>-0.152883</td>\n",
       "      <td>0.311859</td>\n",
       "      <td>-0.198940</td>\n",
       "      <td>-0.232049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS2</th>\n",
       "      <td>0.995294</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.767557</td>\n",
       "      <td>-0.020001</td>\n",
       "      <td>-0.200835</td>\n",
       "      <td>-0.199468</td>\n",
       "      <td>-0.075386</td>\n",
       "      <td>-0.098491</td>\n",
       "      <td>0.309810</td>\n",
       "      <td>-0.180580</td>\n",
       "      <td>-0.206720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS3</th>\n",
       "      <td>-0.719679</td>\n",
       "      <td>-0.767557</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.484111</td>\n",
       "      <td>0.755438</td>\n",
       "      <td>0.754532</td>\n",
       "      <td>0.662852</td>\n",
       "      <td>0.163278</td>\n",
       "      <td>-0.355070</td>\n",
       "      <td>-0.006242</td>\n",
       "      <td>0.237027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS4</th>\n",
       "      <td>0.043416</td>\n",
       "      <td>-0.020001</td>\n",
       "      <td>0.484111</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.744478</td>\n",
       "      <td>0.744778</td>\n",
       "      <td>0.832446</td>\n",
       "      <td>-0.146385</td>\n",
       "      <td>0.154950</td>\n",
       "      <td>-0.001021</td>\n",
       "      <td>-0.244314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS5</th>\n",
       "      <td>-0.125801</td>\n",
       "      <td>-0.200835</td>\n",
       "      <td>0.755438</td>\n",
       "      <td>0.744478</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.999993</td>\n",
       "      <td>0.949962</td>\n",
       "      <td>0.020317</td>\n",
       "      <td>-0.016579</td>\n",
       "      <td>-0.209262</td>\n",
       "      <td>0.021458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS6</th>\n",
       "      <td>-0.124423</td>\n",
       "      <td>-0.199468</td>\n",
       "      <td>0.754532</td>\n",
       "      <td>0.744778</td>\n",
       "      <td>0.999993</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.950316</td>\n",
       "      <td>0.020284</td>\n",
       "      <td>-0.016262</td>\n",
       "      <td>-0.210282</td>\n",
       "      <td>0.021353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cooler_Condition</th>\n",
       "      <td>-0.002000</td>\n",
       "      <td>-0.075386</td>\n",
       "      <td>0.662852</td>\n",
       "      <td>0.832446</td>\n",
       "      <td>0.949962</td>\n",
       "      <td>0.950316</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.004952</td>\n",
       "      <td>-0.004656</td>\n",
       "      <td>-0.203954</td>\n",
       "      <td>-0.004105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Valve_Condition</th>\n",
       "      <td>-0.152883</td>\n",
       "      <td>-0.098491</td>\n",
       "      <td>0.163278</td>\n",
       "      <td>-0.146385</td>\n",
       "      <td>0.020317</td>\n",
       "      <td>0.020284</td>\n",
       "      <td>0.004952</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.352610</td>\n",
       "      <td>-0.082222</td>\n",
       "      <td>0.629433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Internal_Pump_Leakage</th>\n",
       "      <td>0.311859</td>\n",
       "      <td>0.309810</td>\n",
       "      <td>-0.355070</td>\n",
       "      <td>0.154950</td>\n",
       "      <td>-0.016579</td>\n",
       "      <td>-0.016262</td>\n",
       "      <td>-0.004656</td>\n",
       "      <td>-0.352610</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.077303</td>\n",
       "      <td>-0.549678</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hydraulic_Accumulator</th>\n",
       "      <td>-0.198940</td>\n",
       "      <td>-0.180580</td>\n",
       "      <td>-0.006242</td>\n",
       "      <td>-0.001021</td>\n",
       "      <td>-0.209262</td>\n",
       "      <td>-0.210282</td>\n",
       "      <td>-0.203954</td>\n",
       "      <td>-0.082222</td>\n",
       "      <td>0.077303</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.120817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stable_Flag</th>\n",
       "      <td>-0.232049</td>\n",
       "      <td>-0.206720</td>\n",
       "      <td>0.237027</td>\n",
       "      <td>-0.244314</td>\n",
       "      <td>0.021458</td>\n",
       "      <td>0.021353</td>\n",
       "      <td>-0.004105</td>\n",
       "      <td>0.629433</td>\n",
       "      <td>-0.549678</td>\n",
       "      <td>-0.120817</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            PS1       PS2       PS3       PS4       PS5  \\\n",
       "PS1                    1.000000  0.995294 -0.719679  0.043416 -0.125801   \n",
       "PS2                    0.995294  1.000000 -0.767557 -0.020001 -0.200835   \n",
       "PS3                   -0.719679 -0.767557  1.000000  0.484111  0.755438   \n",
       "PS4                    0.043416 -0.020001  0.484111  1.000000  0.744478   \n",
       "PS5                   -0.125801 -0.200835  0.755438  0.744478  1.000000   \n",
       "PS6                   -0.124423 -0.199468  0.754532  0.744778  0.999993   \n",
       "Cooler_Condition      -0.002000 -0.075386  0.662852  0.832446  0.949962   \n",
       "Valve_Condition       -0.152883 -0.098491  0.163278 -0.146385  0.020317   \n",
       "Internal_Pump_Leakage  0.311859  0.309810 -0.355070  0.154950 -0.016579   \n",
       "Hydraulic_Accumulator -0.198940 -0.180580 -0.006242 -0.001021 -0.209262   \n",
       "Stable_Flag           -0.232049 -0.206720  0.237027 -0.244314  0.021458   \n",
       "\n",
       "                            PS6  Cooler_Condition  Valve_Condition  \\\n",
       "PS1                   -0.124423         -0.002000        -0.152883   \n",
       "PS2                   -0.199468         -0.075386        -0.098491   \n",
       "PS3                    0.754532          0.662852         0.163278   \n",
       "PS4                    0.744778          0.832446        -0.146385   \n",
       "PS5                    0.999993          0.949962         0.020317   \n",
       "PS6                    1.000000          0.950316         0.020284   \n",
       "Cooler_Condition       0.950316          1.000000         0.004952   \n",
       "Valve_Condition        0.020284          0.004952         1.000000   \n",
       "Internal_Pump_Leakage -0.016262         -0.004656        -0.352610   \n",
       "Hydraulic_Accumulator -0.210282         -0.203954        -0.082222   \n",
       "Stable_Flag            0.021353         -0.004105         0.629433   \n",
       "\n",
       "                       Internal_Pump_Leakage  Hydraulic_Accumulator  \\\n",
       "PS1                                 0.311859              -0.198940   \n",
       "PS2                                 0.309810              -0.180580   \n",
       "PS3                                -0.355070              -0.006242   \n",
       "PS4                                 0.154950              -0.001021   \n",
       "PS5                                -0.016579              -0.209262   \n",
       "PS6                                -0.016262              -0.210282   \n",
       "Cooler_Condition                   -0.004656              -0.203954   \n",
       "Valve_Condition                    -0.352610              -0.082222   \n",
       "Internal_Pump_Leakage               1.000000               0.077303   \n",
       "Hydraulic_Accumulator               0.077303               1.000000   \n",
       "Stable_Flag                        -0.549678              -0.120817   \n",
       "\n",
       "                       Stable_Flag  \n",
       "PS1                      -0.232049  \n",
       "PS2                      -0.206720  \n",
       "PS3                       0.237027  \n",
       "PS4                      -0.244314  \n",
       "PS5                       0.021458  \n",
       "PS6                       0.021353  \n",
       "Cooler_Condition         -0.004105  \n",
       "Valve_Condition           0.629433  \n",
       "Internal_Pump_Leakage    -0.549678  \n",
       "Hydraulic_Accumulator    -0.120817  \n",
       "Stable_Flag               1.000000  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking correlation between the different pressures and the targets (profiles)\n",
    "df.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure heatmap\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x1080 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting a figure for graphical representation of the correlation\n",
    "fig=plt.gcf()\n",
    "fig.set_size_inches((25,15))\n",
    "sns.heatmap(df.corr(),annot=True)\n",
    "save_fig(\"heatmap\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": false
   },
   "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>Cooler_Condition</th>\n",
       "      <th>Valve_Condition</th>\n",
       "      <th>Internal_Pump_Leakage</th>\n",
       "      <th>Hydraulic_Accumulator</th>\n",
       "      <th>Stable_Flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PS6</th>\n",
       "      <td>0.950316</td>\n",
       "      <td>0.020284</td>\n",
       "      <td>-0.016262</td>\n",
       "      <td>-0.210282</td>\n",
       "      <td>0.021353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS5</th>\n",
       "      <td>0.949962</td>\n",
       "      <td>0.020317</td>\n",
       "      <td>-0.016579</td>\n",
       "      <td>-0.209262</td>\n",
       "      <td>0.021458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS4</th>\n",
       "      <td>0.832446</td>\n",
       "      <td>-0.146385</td>\n",
       "      <td>0.154950</td>\n",
       "      <td>-0.001021</td>\n",
       "      <td>-0.244314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS3</th>\n",
       "      <td>0.662852</td>\n",
       "      <td>0.163278</td>\n",
       "      <td>-0.355070</td>\n",
       "      <td>-0.006242</td>\n",
       "      <td>0.237027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS1</th>\n",
       "      <td>-0.002000</td>\n",
       "      <td>-0.152883</td>\n",
       "      <td>0.311859</td>\n",
       "      <td>-0.198940</td>\n",
       "      <td>-0.232049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PS2</th>\n",
       "      <td>-0.075386</td>\n",
       "      <td>-0.098491</td>\n",
       "      <td>0.309810</td>\n",
       "      <td>-0.180580</td>\n",
       "      <td>-0.206720</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Cooler_Condition  Valve_Condition  Internal_Pump_Leakage  \\\n",
       "PS6          0.950316         0.020284              -0.016262   \n",
       "PS5          0.949962         0.020317              -0.016579   \n",
       "PS4          0.832446        -0.146385               0.154950   \n",
       "PS3          0.662852         0.163278              -0.355070   \n",
       "PS1         -0.002000        -0.152883               0.311859   \n",
       "PS2         -0.075386        -0.098491               0.309810   \n",
       "\n",
       "     Hydraulic_Accumulator  Stable_Flag  \n",
       "PS6              -0.210282     0.021353  \n",
       "PS5              -0.209262     0.021458  \n",
       "PS4              -0.001021    -0.244314  \n",
       "PS3              -0.006242     0.237027  \n",
       "PS1              -0.198940    -0.232049  \n",
       "PS2              -0.180580    -0.206720  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Sorting the pressures - target correlation (from highest to lowest)\n",
    "df_final_corr=df.corr().sort_values([\"Cooler_Condition\",\"Valve_Condition\",\"Internal_Pump_Leakage\",\"Hydraulic_Accumulator\",\"Stable_Flag\"],ascending=False)\n",
    "df_final_corr_target=df_final_corr[[\"Cooler_Condition\",\"Valve_Condition\",\"Internal_Pump_Leakage\",\"Hydraulic_Accumulator\",\"Stable_Flag\"]]\n",
    "df_final_corr_target.drop([\"Cooler_Condition\",\"Valve_Condition\",\"Internal_Pump_Leakage\",\"Hydraulic_Accumulator\",\"Stable_Flag\"],axis=0,inplace=True)\n",
    "df_final_corr_target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysing Correlation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PS1 and PS2 have low to non positive or negative correlation with the Cooler_Condition target. These two datasets could therefore be dropped from the dataframe when focusing on Cooler_Condition target."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>PS3</th>\n",
       "      <th>PS4</th>\n",
       "      <th>PS5</th>\n",
       "      <th>PS6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.991475</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.842170</td>\n",
       "      <td>9.728097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.976234</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.635142</td>\n",
       "      <td>9.529488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.972224</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.530548</td>\n",
       "      <td>9.427949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.946575</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.438827</td>\n",
       "      <td>9.337430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.922707</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.358762</td>\n",
       "      <td>9.260636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2200</th>\n",
       "      <td>2.001438</td>\n",
       "      <td>10.202473</td>\n",
       "      <td>9.972037</td>\n",
       "      <td>9.850361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2201</th>\n",
       "      <td>1.998781</td>\n",
       "      <td>10.197919</td>\n",
       "      <td>9.966184</td>\n",
       "      <td>9.844854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2202</th>\n",
       "      <td>1.993436</td>\n",
       "      <td>10.196824</td>\n",
       "      <td>9.964329</td>\n",
       "      <td>9.842628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2203</th>\n",
       "      <td>2.007077</td>\n",
       "      <td>10.198588</td>\n",
       "      <td>9.968232</td>\n",
       "      <td>9.846690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2204</th>\n",
       "      <td>2.002690</td>\n",
       "      <td>10.203126</td>\n",
       "      <td>9.973638</td>\n",
       "      <td>9.851949</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2205 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           PS3        PS4       PS5       PS6\n",
       "0     1.991475   0.000000  9.842170  9.728097\n",
       "1     1.976234   0.000000  9.635142  9.529488\n",
       "2     1.972224   0.000000  9.530548  9.427949\n",
       "3     1.946575   0.000000  9.438827  9.337430\n",
       "4     1.922707   0.000000  9.358762  9.260636\n",
       "...        ...        ...       ...       ...\n",
       "2200  2.001438  10.202473  9.972037  9.850361\n",
       "2201  1.998781  10.197919  9.966184  9.844854\n",
       "2202  1.993436  10.196824  9.964329  9.842628\n",
       "2203  2.007077  10.198588  9.968232  9.846690\n",
       "2204  2.002690  10.203126  9.973638  9.851949\n",
       "\n",
       "[2205 rows x 4 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pressure_final = df_pressure.drop(columns=['PS1', 'PS2'])\n",
    "df_pressure_final"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>PS3</th>\n",
       "      <th>PS4</th>\n",
       "      <th>PS5</th>\n",
       "      <th>PS6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.991475</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.842170</td>\n",
       "      <td>9.728097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.976234</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.635142</td>\n",
       "      <td>9.529488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.972224</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.530548</td>\n",
       "      <td>9.427949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.946575</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.438827</td>\n",
       "      <td>9.337430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.922707</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.358762</td>\n",
       "      <td>9.260636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2200</th>\n",
       "      <td>2.001438</td>\n",
       "      <td>10.202473</td>\n",
       "      <td>9.972037</td>\n",
       "      <td>9.850361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2201</th>\n",
       "      <td>1.998781</td>\n",
       "      <td>10.197919</td>\n",
       "      <td>9.966184</td>\n",
       "      <td>9.844854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2202</th>\n",
       "      <td>1.993436</td>\n",
       "      <td>10.196824</td>\n",
       "      <td>9.964329</td>\n",
       "      <td>9.842628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2203</th>\n",
       "      <td>2.007077</td>\n",
       "      <td>10.198588</td>\n",
       "      <td>9.968232</td>\n",
       "      <td>9.846690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2204</th>\n",
       "      <td>2.002690</td>\n",
       "      <td>10.203126</td>\n",
       "      <td>9.973638</td>\n",
       "      <td>9.851949</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2205 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           PS3        PS4       PS5       PS6\n",
       "0     1.991475   0.000000  9.842170  9.728097\n",
       "1     1.976234   0.000000  9.635142  9.529488\n",
       "2     1.972224   0.000000  9.530548  9.427949\n",
       "3     1.946575   0.000000  9.438827  9.337430\n",
       "4     1.922707   0.000000  9.358762  9.260636\n",
       "...        ...        ...       ...       ...\n",
       "2200  2.001438  10.202473  9.972037  9.850361\n",
       "2201  1.998781  10.197919  9.966184  9.844854\n",
       "2202  1.993436  10.196824  9.964329  9.842628\n",
       "2203  2.007077  10.198588  9.968232  9.846690\n",
       "2204  2.002690  10.203126  9.973638  9.851949\n",
       "\n",
       "[2205 rows x 4 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x=df_pressure_final\n",
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>Cooler_Condition</th>\n",
       "      <th>Valve_Condition</th>\n",
       "      <th>Internal_Pump_Leakage</th>\n",
       "      <th>Hydraulic_Accumulator</th>\n",
       "      <th>Stable_Flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2200</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2201</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2202</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2203</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2204</th>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2205 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Cooler_Condition  Valve_Condition  Internal_Pump_Leakage  \\\n",
       "0                    3              100                      0   \n",
       "1                    3              100                      0   \n",
       "2                    3              100                      0   \n",
       "3                    3              100                      0   \n",
       "4                    3              100                      0   \n",
       "...                ...              ...                    ...   \n",
       "2200               100              100                      0   \n",
       "2201               100              100                      0   \n",
       "2202               100              100                      0   \n",
       "2203               100              100                      0   \n",
       "2204               100              100                      0   \n",
       "\n",
       "      Hydraulic_Accumulator  Stable_Flag  \n",
       "0                       130            1  \n",
       "1                       130            1  \n",
       "2                       130            1  \n",
       "3                       130            1  \n",
       "4                       130            1  \n",
       "...                     ...          ...  \n",
       "2200                     90            0  \n",
       "2201                     90            0  \n",
       "2202                     90            0  \n",
       "2203                     90            0  \n",
       "2204                     90            0  \n",
       "\n",
       "[2205 rows x 5 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y=df_profile\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_1=y.iloc[:,-5]\n",
    "target_2=y.iloc[:,-4]\n",
    "target = target_1 #Change target used in ML here\n",
    "t_name = \"target1_\" #Change target used in ML here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "100    0.3361\n",
       "3      0.3320\n",
       "20     0.3320\n",
       "Name: Cooler_Condition, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round((target.value_counts(normalize=True)),4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_x,test_x,train_y,test_y=train_test_split(x,target,test_size=0.2,random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "sc=StandardScaler()\n",
    "train_x=sc.fit_transform(train_x)\n",
    "test_x=sc.transform(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test = pd.DataFrame(test_x)\n",
    "from matplotlib import pyplot\n",
    "test.hist()\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Lag en pipeline, IKKE NØDVENDIG -> Ikke nødvendig da det ikke er behov for fylling av NAN verdier eller tillegging av attributer eller OneHotEncoding. StandardScalar brukes direkte.\n",
    "- Kopier alle modeller fra tidligere assignment -> GJORT\n",
    "- Lagre oversikt over best model (csv)\n",
    "- Test tuning av parameter for modellene som ser best ut (dokumenter) -> GJORT\n",
    "- Skriv rapport"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "l_classifier=LogisticRegression(random_state=42)\n",
    "l_classifier.fit(train_x,train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_y=l_classifier.predict(train_x)\n",
    "pred_test_y=l_classifier.predict(test_x)\n",
    "accuracy_score_lr_train=accuracy_score(train_y,pred_train_y)\n",
    "accuracy_score_lr_test=accuracy_score(test_y,pred_test_y)\n",
    "print(\"accuracy_score_lr_train= \",accuracy_score_lr_train)\n",
    "print(\"accuracy_score_lr_test= \",accuracy_score_lr_test)\n",
    "print(\"\\nconfusion_matrix:-\")\n",
    "sns.heatmap(confusion_matrix(test_y,pred_test_y),annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "precision_score_lr_train=precision_score(train_y,pred_train_y,average=\"weighted\")\n",
    "precision_score_lr_test=precision_score(test_y,pred_test_y,average=\"weighted\")\n",
    "print(\"precision_score_lr_train= \",precision_score_lr_train)\n",
    "print(\"precision_score_lr_test= \",precision_score_lr_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores=cross_validate(l_classifier,train_x,train_y,scoring=[\"accuracy\",\"precision_weighted\"],cv=10)\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "accuracy_score_lr_cross=scores[\"test_accuracy\"].mean()\n",
    "print(\"accuracy_score_lr_cross= \",accuracy_score_lr_cross)\n",
    "precision_score_lr_cross=scores[\"test_precision_weighted\"].mean()\n",
    "print(\"precision_score_lr_cross= \",precision_score_lr_cross)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "knn=KNeighborsClassifier(n_neighbors=10,p=2)\n",
    "knn.fit(train_x,train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_y=knn.predict(train_x)\n",
    "pred_test_y=knn.predict(test_x)\n",
    "accuracy_score_knn_train=accuracy_score(train_y,pred_train_y)\n",
    "accuracy_score_knn_test=accuracy_score(test_y,pred_test_y)\n",
    "print(\"accuracy_score_knn_train= \",accuracy_score_knn_train)\n",
    "print(\"accuracy_score_knn_test= \",accuracy_score_knn_test)\n",
    "print(\"\\nconfusion_matrix:-\")\n",
    "sns.heatmap(confusion_matrix(test_y,pred_test_y),annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "precision_score_knn_train=precision_score(train_y,pred_train_y,average=\"weighted\")\n",
    "precision_score_knn_test=precision_score(test_y,pred_test_y,average=\"weighted\")\n",
    "print(\"precision_score_knn_train= \",precision_score_knn_train)\n",
    "print(\"precision_score_knn_test= \",precision_score_knn_test) #overfitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores=cross_validate(knn,train_x,train_y,scoring=[\"accuracy\",\"precision_weighted\"],cv=10)\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "accuracy_score_knn_cross=scores[\"test_accuracy\"].mean()\n",
    "print(\"accuracy_score_knn_cross= \",accuracy_score_knn_cross)\n",
    "precision_score_knn_cross=scores[\"test_precision_weighted\"].mean()\n",
    "print(\"precision_score_knn_cross= \",precision_score_knn_cross)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "svc=SVC(kernel=\"rbf\")\n",
    "svc.fit(train_x,train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_y=svc.predict(train_x)\n",
    "pred_test_y=svc.predict(test_x)\n",
    "accuracy_score_svc_train=accuracy_score(train_y,pred_train_y)\n",
    "accuracy_score_svc_test=accuracy_score(test_y,pred_test_y)\n",
    "print(\"accuracy_score_svc_train= \",accuracy_score_svc_train)\n",
    "print(\"accuracy_score_svc_test= \",accuracy_score_svc_test)\n",
    "print(\"\\nconfusion_matrix:-\")\n",
    "sns.heatmap(confusion_matrix(test_y,pred_test_y),annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "precision_score_svc_train=precision_score(train_y,pred_train_y,average=\"weighted\")\n",
    "precision_score_svc_test=precision_score(test_y,pred_test_y,average=\"weighted\")\n",
    "print(\"precision_score_svc_train= \",precision_score_svc_train)\n",
    "print(\"precision_score_svc_test= \",precision_score_svc_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores=cross_validate(svc,train_x,train_y,scoring=[\"accuracy\",\"precision_weighted\"],cv=10)\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "accuracy_score_svc_cross=scores[\"test_accuracy\"].mean()\n",
    "print(\"accuracy_score_svc_cross= \",accuracy_score_svc_cross)\n",
    "precision_score_svc_cross=scores[\"test_precision_weighted\"].mean()\n",
    "print(\"precision_score_svc_cross= \",precision_score_svc_cross)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dtc=DecisionTreeClassifier(criterion=\"entropy\",random_state=42)\n",
    "dtc.fit(train_x,train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_y=dtc.predict(train_x)\n",
    "pred_test_y=dtc.predict(test_x)\n",
    "accuracy_score_dtc_train=accuracy_score(train_y,pred_train_y)\n",
    "accuracy_score_dtc_test=accuracy_score(test_y,pred_test_y)\n",
    "print(\"accuracy_score_dtc_train= \",accuracy_score_dtc_train)\n",
    "print(\"accuracy_score_dtc_test= \",accuracy_score_dtc_test)\n",
    "print(\"\\nconfusion_matrix:-\")\n",
    "sns.heatmap(confusion_matrix(test_y,pred_test_y),annot=True) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "precision_score_dtc_train=precision_score(train_y,pred_train_y,average=\"weighted\")\n",
    "precision_score_dtc_test=precision_score(test_y,pred_test_y,average=\"weighted\")\n",
    "print(\"precision_score_dtc_train= \",precision_score_dtc_train)\n",
    "print(\"precision_score_dtc_test= \",precision_score_dtc_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores=cross_validate(dtc,train_x,train_y,scoring=[\"accuracy\",\"precision_weighted\"],cv=10)\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "accuracy_score_dtc_cross=scores[\"test_accuracy\"].mean()\n",
    "print(\"accuracy_score_dtc_cross= \",accuracy_score_dtc_cross)\n",
    "precision_score_dtc_cross=scores[\"test_precision_weighted\"].mean()\n",
    "print(\"precision_score_dtc_cross= \",precision_score_dtc_cross)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfc=RandomForestClassifier(n_estimators=10,random_state=1)\n",
    "#Using max_features=4 and random_state=42 instead of 1 gives better result.\n",
    "rfc.fit(train_x,train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_y=rfc.predict(train_x)\n",
    "pred_test_y=rfc.predict(test_x)\n",
    "accuracy_score_rfc_train=accuracy_score(train_y,pred_train_y)\n",
    "accuracy_score_rfc_test=accuracy_score(test_y,pred_test_y)\n",
    "print(\"accuracy_score_rfc_train= \",accuracy_score_rfc_train)\n",
    "print(\"accuracy_score_rfc_test= \",accuracy_score_rfc_test)\n",
    "print(\"\\nconfusion_matrix:-\")\n",
    "sns.heatmap(confusion_matrix(test_y,pred_test_y),annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_mse = mean_squared_error(test_y, pred_test_y)\n",
    "final_rmse = np.sqrt(final_mse)\n",
    "final_rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "confidence = 0.95\n",
    "squared_errors = (pred_test_y - test_y) ** 2\n",
    "np.sqrt(stats.t.interval(confidence, len(squared_errors) - 1,\n",
    "                         loc=squared_errors.mean(),\n",
    "                         scale=stats.sem(squared_errors)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "precision_score_rfc_train=precision_score(train_y,pred_train_y,average=\"weighted\")\n",
    "precision_score_rfc_test=precision_score(test_y,pred_test_y,average=\"weighted\")\n",
    "print(\"precision_score_rfc_train= \",precision_score_rfc_train)\n",
    "print(\"precision_score_rfc_test= \",precision_score_rfc_test)     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores=cross_validate(rfc,train_x,train_y,scoring=[\"accuracy\",\"precision_weighted\"],cv=10)\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "accuracy_score_rfc_cross=scores[\"test_accuracy\"].mean()\n",
    "print(\"accuracy_score_rfc_cross= \",accuracy_score_rfc_cross)\n",
    "precision_score_rfc_cross=scores[\"test_precision_weighted\"].mean()\n",
    "print(\"precision_score_rfc_cross= \",precision_score_rfc_cross)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Best Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_train_accuracy=[accuracy_score_lr_train,accuracy_score_knn_train,accuracy_score_svc_train\n",
    "                     ,accuracy_score_dtc_train,accuracy_score_rfc_train]\n",
    "pred_test_accuracy=[accuracy_score_lr_test,accuracy_score_knn_test,accuracy_score_svc_test\n",
    "                    ,accuracy_score_dtc_test,accuracy_score_rfc_test]\n",
    "pred_train_precision=[precision_score_lr_train,precision_score_knn_train,precision_score_svc_train\n",
    "                     ,precision_score_dtc_train,precision_score_rfc_train]\n",
    "pred_test_precision=[precision_score_lr_test,precision_score_knn_test,precision_score_svc_test\n",
    "                     ,precision_score_dtc_test,precision_score_rfc_test]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train_accuracy=pd.DataFrame(pred_train_accuracy,columns=[\"train\"])\n",
    "df_test_accuracy=pd.DataFrame(pred_test_accuracy,columns=[\"test\"])\n",
    "df_train_precision=pd.DataFrame(pred_train_precision,columns=[\"train\"])\n",
    "df_test_precision=pd.DataFrame(pred_test_precision,columns=[\"test\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_accuracy=pd.concat([df_train_accuracy,df_test_accuracy],axis=1)\n",
    "df_precision=pd.concat([df_train_precision,df_test_precision],axis=1)\n",
    "df_best_model=pd.concat([df_accuracy,df_precision],keys=[\"accuracy\",\"precision\"],axis=1)\n",
    "df_best_model.set_axis([\"logistic_reg\",\"knn\",\"svc\",\"decission_tree\",\"random_for\"],axis=0,inplace=True)\n",
    "df_best_model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fine-Tune Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param_grid = [\n",
    "    # try 12 (3×4) combinations of hyperparameters\n",
    "    {'n_estimators': [3, 10, 30], 'max_features': [2, 4, 6, 8]},\n",
    "    # then try 6 (2×3) combinations with bootstrap set as False\n",
    "    {'bootstrap': [False], 'n_estimators': [3, 10], 'max_features': [2, 3, 4]},\n",
    "  ]\n",
    "\n",
    "forest_reg = RandomForestClassifier(random_state=42)\n",
    "# train across 5 folds, that's a total of (12+6)*5=90 rounds of training \n",
    "grid_search = GridSearchCV(forest_reg, param_grid, cv=10,\n",
    "                           scoring='neg_mean_squared_error',\n",
    "                           return_train_score=True)\n",
    "grid_search.fit(train_x,train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_model = grid_search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_predictions = final_model.predict(test_x)\n",
    "\n",
    "final_mse = mean_squared_error(test_y, final_predictions)\n",
    "final_rmse = np.sqrt(final_mse)\n",
    "final_rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.heatmap(confusion_matrix(test_y,final_predictions),annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "confidence = 0.95\n",
    "squared_errors = (final_predictions - test_y) ** 2\n",
    "np.sqrt(stats.t.interval(confidence, len(squared_errors) - 1,\n",
    "                         loc=squared_errors.mean(),\n",
    "                         scale=stats.sem(squared_errors)))"
   ]
  },
  {
   "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}