Polynomial regression - Multiple Linear Regression.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "917d9699",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib as mpl\n",
    "import tensorflow as tf\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import metrics\n",
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from mpl_toolkits.mplot3d import Axes3D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f56c5c23",
   "metadata": {},
   "outputs": [],
   "source": [
    "class ts_preprocessor(object):\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.scaleFactor = 1\n",
    "    \n",
    "    def fit_transform(self, data):\n",
    "        self.scaleFactor = data.max() + 1 if self.scaleFactor == 1.0 else self.scaleFactor\n",
    "        return np.array([(x / self.scaleFactor) for x in data])\n",
    "    \n",
    "    def inverse_transform(self, data):\n",
    "        return [(x * self.scaleFactor) for x in data]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9fe91a8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5b082b2c",
   "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>side 1</th>\n",
       "      <th>side 2</th>\n",
       "      <th>side 3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12</td>\n",
       "      <td>16</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>15</td>\n",
       "      <td>20</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>18</td>\n",
       "      <td>24</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>994</th>\n",
       "      <td>2988</td>\n",
       "      <td>3984</td>\n",
       "      <td>4980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>2991</td>\n",
       "      <td>3988</td>\n",
       "      <td>4985</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>2994</td>\n",
       "      <td>3992</td>\n",
       "      <td>4990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>2997</td>\n",
       "      <td>3996</td>\n",
       "      <td>4995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>3000</td>\n",
       "      <td>4000</td>\n",
       "      <td>5000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>999 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     side 1  side 2  side 3\n",
       "0         3       4       5\n",
       "1         9      12      15\n",
       "2        12      16      20\n",
       "3        15      20      25\n",
       "4        18      24      30\n",
       "..      ...     ...     ...\n",
       "994    2988    3984    4980\n",
       "995    2991    3988    4985\n",
       "996    2994    3992    4990\n",
       "997    2997    3996    4995\n",
       "998    3000    4000    5000\n",
       "\n",
       "[999 rows x 3 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('rht_new.csv')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b1acd13a",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df[['side 1','side 2']]\n",
    "y = df['side 3']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "29b23625",
   "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>side 1</th>\n",
       "      <th>side 2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>15</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>18</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>994</th>\n",
       "      <td>2988</td>\n",
       "      <td>3984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>2991</td>\n",
       "      <td>3988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>2994</td>\n",
       "      <td>3992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>2997</td>\n",
       "      <td>3996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>3000</td>\n",
       "      <td>4000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>999 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     side 1  side 2\n",
       "0         3       4\n",
       "1         9      12\n",
       "2        12      16\n",
       "3        15      20\n",
       "4        18      24\n",
       "..      ...     ...\n",
       "994    2988    3984\n",
       "995    2991    3988\n",
       "996    2994    3992\n",
       "997    2997    3996\n",
       "998    3000    4000\n",
       "\n",
       "[999 rows x 2 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2faefe82",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mpl.rcParams['legend.fontsize'] = 12\n",
    " \n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(projection ='3d')\n",
    " \n",
    "ax.scatter(X['side 1'], X['side 2'], y, label ='y', s = 5)\n",
    "ax.set_xlabel('Side 1')\n",
    "ax.set_ylabel('Side 2')\n",
    "ax.set_zlabel('Side 3')\n",
    "ax.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "510402da",
   "metadata": {},
   "source": [
    "**Clear straight-line --> linear relationship**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "50c05605",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_linear = LinearRegression()\n",
    "model_linear.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e04288f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set the degree equals to 3\n",
    "poly_reg = PolynomialFeatures(degree = 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "974b4663",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.00000000e+00, 3.00000000e+00, 4.00000000e+00, ...,\n",
       "        3.60000000e+01, 4.80000000e+01, 6.40000000e+01],\n",
       "       [1.00000000e+00, 9.00000000e+00, 1.20000000e+01, ...,\n",
       "        9.72000000e+02, 1.29600000e+03, 1.72800000e+03],\n",
       "       [1.00000000e+00, 1.20000000e+01, 1.60000000e+01, ...,\n",
       "        2.30400000e+03, 3.07200000e+03, 4.09600000e+03],\n",
       "       ...,\n",
       "       [1.00000000e+00, 2.99400000e+03, 3.99200000e+03, ...,\n",
       "        3.57844317e+10, 4.77125756e+10, 6.36167675e+10],\n",
       "       [1.00000000e+00, 2.99700000e+03, 3.99600000e+03, ...,\n",
       "        3.58921080e+10, 4.78561440e+10, 6.38081919e+10],\n",
       "       [1.00000000e+00, 3.00000000e+03, 4.00000000e+03, ...,\n",
       "        3.60000000e+10, 4.80000000e+10, 6.40000000e+10]])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_poly = poly_reg.fit_transform(X)\n",
    "X_poly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "798802a8",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_linear_2 = LinearRegression()\n",
    "# fit the model\n",
    "model_linear_2.fit(X_poly, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "770b5de2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([4.99999887e+00, 1.49999989e+01, 1.99999989e+01, 2.49999989e+01,\n",
       "       2.99999989e+01, 3.49999989e+01, 3.99999989e+01, 4.49999989e+01,\n",
       "       4.99999989e+01, 5.49999989e+01, 5.99999989e+01, 6.49999989e+01,\n",
       "       6.99999989e+01, 7.49999989e+01, 7.99999989e+01, 8.49999989e+01,\n",
       "       8.99999989e+01, 9.49999989e+01, 9.99999989e+01, 1.04999999e+02,\n",
       "       1.09999999e+02, 1.14999999e+02, 1.19999999e+02, 1.24999999e+02,\n",
       "       1.29999999e+02, 1.34999999e+02, 1.39999999e+02, 1.44999999e+02,\n",
       "       1.49999999e+02, 1.54999999e+02, 1.59999999e+02, 1.64999999e+02,\n",
       "       1.69999999e+02, 1.74999999e+02, 1.79999999e+02, 1.84999999e+02,\n",
       "       1.89999999e+02, 1.94999999e+02, 1.99999999e+02, 2.04999999e+02,\n",
       "       2.09999999e+02, 2.14999999e+02, 2.19999999e+02, 2.24999999e+02,\n",
       "       2.29999999e+02, 2.34999999e+02, 2.39999999e+02, 2.44999999e+02,\n",
       "       2.49999999e+02, 2.54999999e+02, 2.59999999e+02, 2.64999999e+02,\n",
       "       2.69999999e+02, 2.74999999e+02, 2.79999999e+02, 2.84999999e+02,\n",
       "       2.89999999e+02, 2.94999999e+02, 2.99999999e+02, 3.04999999e+02,\n",
       "       3.09999999e+02, 3.14999999e+02, 3.19999999e+02, 3.24999999e+02,\n",
       "       3.29999999e+02, 3.34999999e+02, 3.39999999e+02, 3.44999999e+02,\n",
       "       3.49999999e+02, 3.54999999e+02, 3.59999999e+02, 3.64999999e+02,\n",
       "       3.69999999e+02, 3.74999999e+02, 3.79999999e+02, 3.84999999e+02,\n",
       "       3.89999999e+02, 3.94999999e+02, 3.99999999e+02, 4.04999999e+02,\n",
       "       4.09999999e+02, 4.14999999e+02, 4.19999999e+02, 4.24999999e+02,\n",
       "       4.29999999e+02, 4.34999999e+02, 4.39999999e+02, 4.44999999e+02,\n",
       "       4.49999999e+02, 4.54999999e+02, 4.59999999e+02, 4.64999999e+02,\n",
       "       4.69999999e+02, 4.74999999e+02, 4.79999999e+02, 4.84999999e+02,\n",
       "       4.89999999e+02, 4.94999999e+02, 4.99999999e+02, 5.04999999e+02,\n",
       "       5.09999999e+02, 5.14999999e+02, 5.19999999e+02, 5.24999999e+02,\n",
       "       5.29999999e+02, 5.34999999e+02, 5.39999999e+02, 5.44999999e+02,\n",
       "       5.49999999e+02, 5.54999999e+02, 5.59999999e+02, 5.64999999e+02,\n",
       "       5.69999999e+02, 5.74999999e+02, 5.79999999e+02, 5.84999999e+02,\n",
       "       5.89999999e+02, 5.94999999e+02, 5.99999999e+02, 6.04999999e+02,\n",
       "       6.09999999e+02, 6.14999999e+02, 6.19999999e+02, 6.24999999e+02,\n",
       "       6.29999999e+02, 6.34999999e+02, 6.39999999e+02, 6.44999999e+02,\n",
       "       6.49999999e+02, 6.54999999e+02, 6.59999999e+02, 6.64999999e+02,\n",
       "       6.69999999e+02, 6.74999999e+02, 6.79999999e+02, 6.84999999e+02,\n",
       "       6.89999999e+02, 6.94999999e+02, 6.99999999e+02, 7.04999999e+02,\n",
       "       7.09999999e+02, 7.14999999e+02, 7.19999999e+02, 7.24999999e+02,\n",
       "       7.29999999e+02, 7.34999999e+02, 7.39999999e+02, 7.44999999e+02,\n",
       "       7.49999999e+02, 7.54999999e+02, 7.59999999e+02, 7.64999999e+02,\n",
       "       7.69999999e+02, 7.74999999e+02, 7.79999999e+02, 7.84999999e+02,\n",
       "       7.89999999e+02, 7.94999999e+02, 7.99999999e+02, 8.04999999e+02,\n",
       "       8.09999999e+02, 8.14999999e+02, 8.19999999e+02, 8.24999999e+02,\n",
       "       8.29999999e+02, 8.34999999e+02, 8.39999999e+02, 8.44999999e+02,\n",
       "       8.49999999e+02, 8.54999999e+02, 8.59999999e+02, 8.64999999e+02,\n",
       "       8.69999999e+02, 8.74999999e+02, 8.79999999e+02, 8.84999999e+02,\n",
       "       8.89999999e+02, 8.94999999e+02, 8.99999999e+02, 9.04999999e+02,\n",
       "       9.09999999e+02, 9.14999999e+02, 9.19999999e+02, 9.24999999e+02,\n",
       "       9.29999999e+02, 9.34999999e+02, 9.39999999e+02, 9.44999999e+02,\n",
       "       9.49999999e+02, 9.54999999e+02, 9.59999999e+02, 9.64999999e+02,\n",
       "       9.69999999e+02, 9.74999999e+02, 9.79999999e+02, 9.84999999e+02,\n",
       "       9.89999999e+02, 9.94999999e+02, 9.99999999e+02, 1.00500000e+03,\n",
       "       1.01000000e+03, 1.01500000e+03, 1.02000000e+03, 1.02500000e+03,\n",
       "       1.03000000e+03, 1.03500000e+03, 1.04000000e+03, 1.04500000e+03,\n",
       "       1.05000000e+03, 1.05500000e+03, 1.06000000e+03, 1.06500000e+03,\n",
       "       1.07000000e+03, 1.07500000e+03, 1.08000000e+03, 1.08500000e+03,\n",
       "       1.09000000e+03, 1.09500000e+03, 1.10000000e+03, 1.10500000e+03,\n",
       "       1.11000000e+03, 1.11500000e+03, 1.12000000e+03, 1.12500000e+03,\n",
       "       1.13000000e+03, 1.13500000e+03, 1.14000000e+03, 1.14500000e+03,\n",
       "       1.15000000e+03, 1.15500000e+03, 1.16000000e+03, 1.16500000e+03,\n",
       "       1.17000000e+03, 1.17500000e+03, 1.18000000e+03, 1.18500000e+03,\n",
       "       1.19000000e+03, 1.19500000e+03, 1.20000000e+03, 1.20500000e+03,\n",
       "       1.21000000e+03, 1.21500000e+03, 1.22000000e+03, 1.22500000e+03,\n",
       "       1.23000000e+03, 1.23500000e+03, 1.24000000e+03, 1.24500000e+03,\n",
       "       1.25000000e+03, 1.25500000e+03, 1.26000000e+03, 1.26500000e+03,\n",
       "       1.27000000e+03, 1.27500000e+03, 1.28000000e+03, 1.28500000e+03,\n",
       "       1.29000000e+03, 1.29500000e+03, 1.30000000e+03, 1.30500000e+03,\n",
       "       1.31000000e+03, 1.31500000e+03, 1.32000000e+03, 1.32500000e+03,\n",
       "       1.33000000e+03, 1.33500000e+03, 1.34000000e+03, 1.34500000e+03,\n",
       "       1.35000000e+03, 1.35500000e+03, 1.36000000e+03, 1.36500000e+03,\n",
       "       1.37000000e+03, 1.37500000e+03, 1.38000000e+03, 1.38500000e+03,\n",
       "       1.39000000e+03, 1.39500000e+03, 1.40000000e+03, 1.40500000e+03,\n",
       "       1.41000000e+03, 1.41500000e+03, 1.42000000e+03, 1.42500000e+03,\n",
       "       1.43000000e+03, 1.43500000e+03, 1.44000000e+03, 1.44500000e+03,\n",
       "       1.45000000e+03, 1.45500000e+03, 1.46000000e+03, 1.46500000e+03,\n",
       "       1.47000000e+03, 1.47500000e+03, 1.48000000e+03, 1.48500000e+03,\n",
       "       1.49000000e+03, 1.49500000e+03, 1.50000000e+03, 1.50500000e+03,\n",
       "       1.51000000e+03, 1.51500000e+03, 1.52000000e+03, 1.52500000e+03,\n",
       "       1.53000000e+03, 1.53500000e+03, 1.54000000e+03, 1.54500000e+03,\n",
       "       1.55000000e+03, 1.55500000e+03, 1.56000000e+03, 1.56500000e+03,\n",
       "       1.57000000e+03, 1.57500000e+03, 1.58000000e+03, 1.58500000e+03,\n",
       "       1.59000000e+03, 1.59500000e+03, 1.60000000e+03, 1.60500000e+03,\n",
       "       1.61000000e+03, 1.61500000e+03, 1.62000000e+03, 1.62500000e+03,\n",
       "       1.63000000e+03, 1.63500000e+03, 1.64000000e+03, 1.64500000e+03,\n",
       "       1.65000000e+03, 1.65500000e+03, 1.66000000e+03, 1.66500000e+03,\n",
       "       1.67000000e+03, 1.67500000e+03, 1.68000000e+03, 1.68500000e+03,\n",
       "       1.69000000e+03, 1.69500000e+03, 1.70000000e+03, 1.70500000e+03,\n",
       "       1.71000000e+03, 1.71500000e+03, 1.72000000e+03, 1.72500000e+03,\n",
       "       1.73000000e+03, 1.73500000e+03, 1.74000000e+03, 1.74500000e+03,\n",
       "       1.75000000e+03, 1.75500000e+03, 1.76000000e+03, 1.76500000e+03,\n",
       "       1.77000000e+03, 1.77500000e+03, 1.78000000e+03, 1.78500000e+03,\n",
       "       1.79000000e+03, 1.79500000e+03, 1.80000000e+03, 1.80500000e+03,\n",
       "       1.81000000e+03, 1.81500000e+03, 1.82000000e+03, 1.82500000e+03,\n",
       "       1.83000000e+03, 1.83500000e+03, 1.84000000e+03, 1.84500000e+03,\n",
       "       1.85000000e+03, 1.85500000e+03, 1.86000000e+03, 1.86500000e+03,\n",
       "       1.87000000e+03, 1.87500000e+03, 1.88000000e+03, 1.88500000e+03,\n",
       "       1.89000000e+03, 1.89500000e+03, 1.90000000e+03, 1.90500000e+03,\n",
       "       1.91000000e+03, 1.91500000e+03, 1.92000000e+03, 1.92500000e+03,\n",
       "       1.93000000e+03, 1.93500000e+03, 1.94000000e+03, 1.94500000e+03,\n",
       "       1.95000000e+03, 1.95500000e+03, 1.96000000e+03, 1.96500000e+03,\n",
       "       1.97000000e+03, 1.97500000e+03, 1.98000000e+03, 1.98500000e+03,\n",
       "       1.99000000e+03, 1.99500000e+03, 2.00000000e+03, 2.00500000e+03,\n",
       "       2.01000000e+03, 2.01500000e+03, 2.02000000e+03, 2.02500000e+03,\n",
       "       2.03000000e+03, 2.03500000e+03, 2.04000000e+03, 2.04500000e+03,\n",
       "       2.05000000e+03, 2.05500000e+03, 2.06000000e+03, 2.06500000e+03,\n",
       "       2.07000000e+03, 2.07500000e+03, 2.08000000e+03, 2.08500000e+03,\n",
       "       2.09000000e+03, 2.09500000e+03, 2.10000000e+03, 2.10500000e+03,\n",
       "       2.11000000e+03, 2.11500000e+03, 2.12000000e+03, 2.12500000e+03,\n",
       "       2.13000000e+03, 2.13500000e+03, 2.14000000e+03, 2.14500000e+03,\n",
       "       2.15000000e+03, 2.15500000e+03, 2.16000000e+03, 2.16500000e+03,\n",
       "       2.17000000e+03, 2.17500000e+03, 2.18000000e+03, 2.18500000e+03,\n",
       "       2.19000000e+03, 2.19500000e+03, 2.20000000e+03, 2.20500000e+03,\n",
       "       2.21000000e+03, 2.21500000e+03, 2.22000000e+03, 2.22500000e+03,\n",
       "       2.23000000e+03, 2.23500000e+03, 2.24000000e+03, 2.24500000e+03,\n",
       "       2.25000000e+03, 2.25500000e+03, 2.26000000e+03, 2.26500000e+03,\n",
       "       2.27000000e+03, 2.27500000e+03, 2.28000000e+03, 2.28500000e+03,\n",
       "       2.29000000e+03, 2.29500000e+03, 2.30000000e+03, 2.30500000e+03,\n",
       "       2.31000000e+03, 2.31500000e+03, 2.32000000e+03, 2.32500000e+03,\n",
       "       2.33000000e+03, 2.33500000e+03, 2.34000000e+03, 2.34500000e+03,\n",
       "       2.35000000e+03, 2.35500000e+03, 2.36000000e+03, 2.36500000e+03,\n",
       "       2.37000000e+03, 2.37500000e+03, 2.38000000e+03, 2.38500000e+03,\n",
       "       2.39000000e+03, 2.39500000e+03, 2.40000000e+03, 2.40500000e+03,\n",
       "       2.41000000e+03, 2.41500000e+03, 2.42000000e+03, 2.42500000e+03,\n",
       "       2.43000000e+03, 2.43500000e+03, 2.44000000e+03, 2.44500000e+03,\n",
       "       2.45000000e+03, 2.45500000e+03, 2.46000000e+03, 2.46500000e+03,\n",
       "       2.47000000e+03, 2.47500000e+03, 2.48000000e+03, 2.48500000e+03,\n",
       "       2.49000000e+03, 2.49500000e+03, 2.50000000e+03, 2.50500000e+03,\n",
       "       2.51000000e+03, 2.51500000e+03, 2.52000000e+03, 2.52500000e+03,\n",
       "       2.53000000e+03, 2.53500000e+03, 2.54000000e+03, 2.54500000e+03,\n",
       "       2.55000000e+03, 2.55500000e+03, 2.56000000e+03, 2.56500000e+03,\n",
       "       2.57000000e+03, 2.57500000e+03, 2.58000000e+03, 2.58500000e+03,\n",
       "       2.59000000e+03, 2.59500000e+03, 2.60000000e+03, 2.60500000e+03,\n",
       "       2.61000000e+03, 2.61500000e+03, 2.62000000e+03, 2.62500000e+03,\n",
       "       2.63000000e+03, 2.63500000e+03, 2.64000000e+03, 2.64500000e+03,\n",
       "       2.65000000e+03, 2.65500000e+03, 2.66000000e+03, 2.66500000e+03,\n",
       "       2.67000000e+03, 2.67500000e+03, 2.68000000e+03, 2.68500000e+03,\n",
       "       2.69000000e+03, 2.69500000e+03, 2.70000000e+03, 2.70500000e+03,\n",
       "       2.71000000e+03, 2.71500000e+03, 2.72000000e+03, 2.72500000e+03,\n",
       "       2.73000000e+03, 2.73500000e+03, 2.74000000e+03, 2.74500000e+03,\n",
       "       2.75000000e+03, 2.75500000e+03, 2.76000000e+03, 2.76500000e+03,\n",
       "       2.77000000e+03, 2.77500000e+03, 2.78000000e+03, 2.78500000e+03,\n",
       "       2.79000000e+03, 2.79500000e+03, 2.80000000e+03, 2.80500000e+03,\n",
       "       2.81000000e+03, 2.81500000e+03, 2.82000000e+03, 2.82500000e+03,\n",
       "       2.83000000e+03, 2.83500000e+03, 2.84000000e+03, 2.84500000e+03,\n",
       "       2.85000000e+03, 2.85500000e+03, 2.86000000e+03, 2.86500000e+03,\n",
       "       2.87000000e+03, 2.87500000e+03, 2.88000000e+03, 2.88500000e+03,\n",
       "       2.89000000e+03, 2.89500000e+03, 2.90000000e+03, 2.90500000e+03,\n",
       "       2.91000000e+03, 2.91500000e+03, 2.92000000e+03, 2.92500000e+03,\n",
       "       2.93000000e+03, 2.93500000e+03, 2.94000000e+03, 2.94500000e+03,\n",
       "       2.95000000e+03, 2.95500000e+03, 2.96000000e+03, 2.96500000e+03,\n",
       "       2.97000000e+03, 2.97500000e+03, 2.98000000e+03, 2.98500000e+03,\n",
       "       2.99000000e+03, 2.99500000e+03, 3.00000000e+03, 3.00500000e+03,\n",
       "       3.01000000e+03, 3.01500000e+03, 3.02000000e+03, 3.02500000e+03,\n",
       "       3.03000000e+03, 3.03500000e+03, 3.04000000e+03, 3.04500000e+03,\n",
       "       3.05000000e+03, 3.05500000e+03, 3.06000000e+03, 3.06500000e+03,\n",
       "       3.07000000e+03, 3.07500000e+03, 3.08000000e+03, 3.08500000e+03,\n",
       "       3.09000000e+03, 3.09500000e+03, 3.10000000e+03, 3.10500000e+03,\n",
       "       3.11000000e+03, 3.11500000e+03, 3.12000000e+03, 3.12500000e+03,\n",
       "       3.13000000e+03, 3.13500000e+03, 3.14000000e+03, 3.14500000e+03,\n",
       "       3.15000000e+03, 3.15500000e+03, 3.16000000e+03, 3.16500000e+03,\n",
       "       3.17000000e+03, 3.17500000e+03, 3.18000000e+03, 3.18500000e+03,\n",
       "       3.19000000e+03, 3.19500000e+03, 3.20000000e+03, 3.20500000e+03,\n",
       "       3.21000000e+03, 3.21500000e+03, 3.22000000e+03, 3.22500000e+03,\n",
       "       3.23000000e+03, 3.23500000e+03, 3.24000000e+03, 3.24500000e+03,\n",
       "       3.25000000e+03, 3.25500000e+03, 3.26000000e+03, 3.26500000e+03,\n",
       "       3.27000000e+03, 3.27500000e+03, 3.28000000e+03, 3.28500000e+03,\n",
       "       3.29000000e+03, 3.29500000e+03, 3.30000000e+03, 3.30500000e+03,\n",
       "       3.31000000e+03, 3.31500000e+03, 3.32000000e+03, 3.32500000e+03,\n",
       "       3.33000000e+03, 3.33500000e+03, 3.34000000e+03, 3.34500000e+03,\n",
       "       3.35000000e+03, 3.35500000e+03, 3.36000000e+03, 3.36500000e+03,\n",
       "       3.37000000e+03, 3.37500000e+03, 3.38000000e+03, 3.38500000e+03,\n",
       "       3.39000000e+03, 3.39500000e+03, 3.40000000e+03, 3.40500000e+03,\n",
       "       3.41000000e+03, 3.41500000e+03, 3.42000000e+03, 3.42500000e+03,\n",
       "       3.43000000e+03, 3.43500000e+03, 3.44000000e+03, 3.44500000e+03,\n",
       "       3.45000000e+03, 3.45500000e+03, 3.46000000e+03, 3.46500000e+03,\n",
       "       3.47000000e+03, 3.47500000e+03, 3.48000000e+03, 3.48500000e+03,\n",
       "       3.49000000e+03, 3.49500000e+03, 3.50000000e+03, 3.50500000e+03,\n",
       "       3.51000000e+03, 3.51500000e+03, 3.52000000e+03, 3.52500000e+03,\n",
       "       3.53000000e+03, 3.53500000e+03, 3.54000000e+03, 3.54500000e+03,\n",
       "       3.55000000e+03, 3.55500000e+03, 3.56000000e+03, 3.56500000e+03,\n",
       "       3.57000000e+03, 3.57500000e+03, 3.58000000e+03, 3.58500000e+03,\n",
       "       3.59000000e+03, 3.59500000e+03, 3.60000000e+03, 3.60500000e+03,\n",
       "       3.61000000e+03, 3.61500000e+03, 3.62000000e+03, 3.62500000e+03,\n",
       "       3.63000000e+03, 3.63500000e+03, 3.64000000e+03, 3.64500000e+03,\n",
       "       3.65000000e+03, 3.65500000e+03, 3.66000000e+03, 3.66500000e+03,\n",
       "       3.67000000e+03, 3.67500000e+03, 3.68000000e+03, 3.68500000e+03,\n",
       "       3.69000000e+03, 3.69500000e+03, 3.70000000e+03, 3.70500000e+03,\n",
       "       3.71000000e+03, 3.71500000e+03, 3.72000000e+03, 3.72500000e+03,\n",
       "       3.73000000e+03, 3.73500000e+03, 3.74000000e+03, 3.74500000e+03,\n",
       "       3.75000000e+03, 3.75500000e+03, 3.76000000e+03, 3.76500000e+03,\n",
       "       3.77000000e+03, 3.77500000e+03, 3.78000000e+03, 3.78500000e+03,\n",
       "       3.79000000e+03, 3.79500000e+03, 3.80000000e+03, 3.80500000e+03,\n",
       "       3.81000000e+03, 3.81500000e+03, 3.82000000e+03, 3.82500000e+03,\n",
       "       3.83000000e+03, 3.83500000e+03, 3.84000000e+03, 3.84500000e+03,\n",
       "       3.85000000e+03, 3.85500000e+03, 3.86000000e+03, 3.86500000e+03,\n",
       "       3.87000000e+03, 3.87500000e+03, 3.88000000e+03, 3.88500000e+03,\n",
       "       3.89000000e+03, 3.89500000e+03, 3.90000000e+03, 3.90500000e+03,\n",
       "       3.91000000e+03, 3.91500000e+03, 3.92000000e+03, 3.92500000e+03,\n",
       "       3.93000000e+03, 3.93500000e+03, 3.94000000e+03, 3.94500000e+03,\n",
       "       3.95000000e+03, 3.95500000e+03, 3.96000000e+03, 3.96500000e+03,\n",
       "       3.97000000e+03, 3.97500000e+03, 3.98000000e+03, 3.98500000e+03,\n",
       "       3.99000000e+03, 3.99500000e+03, 4.00000000e+03, 4.00500000e+03,\n",
       "       4.01000000e+03, 4.01500000e+03, 4.02000000e+03, 4.02500000e+03,\n",
       "       4.03000000e+03, 4.03500000e+03, 4.04000000e+03, 4.04500000e+03,\n",
       "       4.05000000e+03, 4.05500000e+03, 4.06000000e+03, 4.06500000e+03,\n",
       "       4.07000000e+03, 4.07500000e+03, 4.08000000e+03, 4.08500000e+03,\n",
       "       4.09000000e+03, 4.09500000e+03, 4.10000000e+03, 4.10500000e+03,\n",
       "       4.11000000e+03, 4.11500000e+03, 4.12000000e+03, 4.12500000e+03,\n",
       "       4.13000000e+03, 4.13500000e+03, 4.14000000e+03, 4.14500000e+03,\n",
       "       4.15000000e+03, 4.15500000e+03, 4.16000000e+03, 4.16500000e+03,\n",
       "       4.17000000e+03, 4.17500000e+03, 4.18000000e+03, 4.18500000e+03,\n",
       "       4.19000000e+03, 4.19500000e+03, 4.20000000e+03, 4.20500000e+03,\n",
       "       4.21000000e+03, 4.21500000e+03, 4.22000000e+03, 4.22500000e+03,\n",
       "       4.23000000e+03, 4.23500000e+03, 4.24000000e+03, 4.24500000e+03,\n",
       "       4.25000000e+03, 4.25500000e+03, 4.26000000e+03, 4.26500000e+03,\n",
       "       4.27000000e+03, 4.27500000e+03, 4.28000000e+03, 4.28500000e+03,\n",
       "       4.29000000e+03, 4.29500000e+03, 4.30000000e+03, 4.30500000e+03,\n",
       "       4.31000000e+03, 4.31500000e+03, 4.32000000e+03, 4.32500000e+03,\n",
       "       4.33000000e+03, 4.33500000e+03, 4.34000000e+03, 4.34500000e+03,\n",
       "       4.35000000e+03, 4.35500000e+03, 4.36000000e+03, 4.36500000e+03,\n",
       "       4.37000000e+03, 4.37500000e+03, 4.38000000e+03, 4.38500000e+03,\n",
       "       4.39000000e+03, 4.39500000e+03, 4.40000000e+03, 4.40500000e+03,\n",
       "       4.41000000e+03, 4.41500000e+03, 4.42000000e+03, 4.42500000e+03,\n",
       "       4.43000000e+03, 4.43500000e+03, 4.44000000e+03, 4.44500000e+03,\n",
       "       4.45000000e+03, 4.45500000e+03, 4.46000000e+03, 4.46500000e+03,\n",
       "       4.47000000e+03, 4.47500000e+03, 4.48000000e+03, 4.48500000e+03,\n",
       "       4.49000000e+03, 4.49500000e+03, 4.50000000e+03, 4.50500000e+03,\n",
       "       4.51000000e+03, 4.51500000e+03, 4.52000000e+03, 4.52500000e+03,\n",
       "       4.53000000e+03, 4.53500000e+03, 4.54000000e+03, 4.54500000e+03,\n",
       "       4.55000000e+03, 4.55500000e+03, 4.56000000e+03, 4.56500000e+03,\n",
       "       4.57000000e+03, 4.57500000e+03, 4.58000000e+03, 4.58500000e+03,\n",
       "       4.59000000e+03, 4.59500000e+03, 4.60000000e+03, 4.60500000e+03,\n",
       "       4.61000000e+03, 4.61500000e+03, 4.62000000e+03, 4.62500000e+03,\n",
       "       4.63000000e+03, 4.63500000e+03, 4.64000000e+03, 4.64500000e+03,\n",
       "       4.65000000e+03, 4.65500000e+03, 4.66000000e+03, 4.66500000e+03,\n",
       "       4.67000000e+03, 4.67500000e+03, 4.68000000e+03, 4.68500000e+03,\n",
       "       4.69000000e+03, 4.69500000e+03, 4.70000000e+03, 4.70500000e+03,\n",
       "       4.71000000e+03, 4.71500000e+03, 4.72000000e+03, 4.72500000e+03,\n",
       "       4.73000000e+03, 4.73500000e+03, 4.74000000e+03, 4.74500000e+03,\n",
       "       4.75000000e+03, 4.75500000e+03, 4.76000000e+03, 4.76500000e+03,\n",
       "       4.77000000e+03, 4.77500000e+03, 4.78000000e+03, 4.78500000e+03,\n",
       "       4.79000000e+03, 4.79500000e+03, 4.80000000e+03, 4.80500000e+03,\n",
       "       4.81000000e+03, 4.81500000e+03, 4.82000000e+03, 4.82500000e+03,\n",
       "       4.83000000e+03, 4.83500000e+03, 4.84000000e+03, 4.84500000e+03,\n",
       "       4.85000000e+03, 4.85500000e+03, 4.86000000e+03, 4.86500000e+03,\n",
       "       4.87000000e+03, 4.87500000e+03, 4.88000000e+03, 4.88500000e+03,\n",
       "       4.89000000e+03, 4.89500000e+03, 4.90000000e+03, 4.90500000e+03,\n",
       "       4.91000000e+03, 4.91500000e+03, 4.92000000e+03, 4.92500000e+03,\n",
       "       4.93000000e+03, 4.93500000e+03, 4.94000000e+03, 4.94500000e+03,\n",
       "       4.95000000e+03, 4.95500000e+03, 4.96000000e+03, 4.96500000e+03,\n",
       "       4.97000000e+03, 4.97500000e+03, 4.98000000e+03, 4.98500000e+03,\n",
       "       4.99000000e+03, 4.99500000e+03, 5.00000000e+03])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_poly_pred = model_linear_2.predict(X_poly)\n",
    "y_poly_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "c90d6be5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculate_rmse(y_true, y_pred):\n",
    "    squared_errors = [(true - pred) ** 2 for true, pred in zip(y_true, y_pred)]\n",
    "    mean_squared_error = np.mean(squared_errors)\n",
    "    rmse = np.sqrt(mean_squared_error)\n",
    "    return rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "7eaceae3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test evaluation for quadratic polynomial regressor:\n",
      "\n",
      "R^2 Score: 1.0\n",
      "RMSE: 6.301193633481666e-07\n",
      "Correlation Coefficient (R): 0.9999999999999993\n"
     ]
    }
   ],
   "source": [
    "#Evaluation\n",
    "print('Test evaluation for quadratic polynomial regressor:\\n')\n",
    "print('R^2 Score:', metrics.r2_score(y, y_poly_pred))\n",
    "print('RMSE:', calculate_rmse(y, y_poly_pred))\n",
    "print('Correlation Coefficient (R):', np.corrcoef(y, y_poly_pred)[0, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "ace08fb1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Please enter the two sides of the triangle\n",
      "\n",
      "Enter Side A:\n",
      "10\n",
      "Enter Side B:\n",
      "15\n",
      "The predicted Third Side is: 18.00\n",
      "\n",
      "Please enter the two sides of the triangle\n",
      "\n",
      "Enter Side A:\n",
      "100\n",
      "Enter Side B:\n",
      "150\n",
      "The predicted Third Side is: 180.00\n",
      "\n",
      "Please enter the two sides of the triangle\n",
      "\n",
      "Enter Side A:\n",
      "58\n",
      "Enter Side B:\n",
      "68\n",
      "The predicted Third Side is: 89.20\n",
      "\n",
      "Please enter the two sides of the triangle\n",
      "\n",
      "Enter Side A:\n",
      "i\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "could not convert string to float: 'i'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_544\\1012094891.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      2\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"Please enter the two sides of the triangle\\n\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"Enter Side A:\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 4\u001b[1;33m     \u001b[0msidea\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfloat\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      5\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"Enter Side B:\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      6\u001b[0m     \u001b[0msideb\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfloat\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: could not convert string to float: 'i'"
     ]
    }
   ],
   "source": [
    "while True:\n",
    "    print(\"Please enter the two sides of the triangle\\n\")\n",
    "    print(\"Enter Side A:\")\n",
    "    sidea = float(input())\n",
    "    print(\"Enter Side B:\")\n",
    "    sideb = float(input())\n",
    "    \n",
    "    # Create a feature array for prediction\n",
    "    feature_array = np.array([[sidea, sideb]])\n",
    "    \n",
    "    # Transform the features using the polynomial regression model\n",
    "    feature_poly = poly_reg.transform(feature_array)\n",
    "    \n",
    "    # Predict the third side using the trained polynomial regression model\n",
    "    prediction = model_linear_2.predict(feature_poly)\n",
    "    \n",
    "    # Print the predicted third side\n",
    "    print(\"The predicted Third Side is: {:.2f}\\n\".format(prediction[0]))\n",
    "    \n",
    "# The code will run until you manually stop it (e.g., by pressing Ctrl+C).\n",
    "print(\"Done\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc5986b5",
   "metadata": {},
   "source": [
    "### Test 1: Dataset has large value range than that of original"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b9f77b33",
   "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>side 1</th>\n",
       "      <th>side 2</th>\n",
       "      <th>side 3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8403</td>\n",
       "      <td>6062</td>\n",
       "      <td>10361.382775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8133</td>\n",
       "      <td>8748</td>\n",
       "      <td>11944.588440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4346</td>\n",
       "      <td>2247</td>\n",
       "      <td>4892.517246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2883</td>\n",
       "      <td>2983</td>\n",
       "      <td>4148.491051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6268</td>\n",
       "      <td>1265</td>\n",
       "      <td>6394.376357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>8018</td>\n",
       "      <td>5214</td>\n",
       "      <td>9564.210370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>8844</td>\n",
       "      <td>6177</td>\n",
       "      <td>10787.569930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>3738</td>\n",
       "      <td>9852</td>\n",
       "      <td>10537.293201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>2806</td>\n",
       "      <td>4195</td>\n",
       "      <td>5046.945710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>1859</td>\n",
       "      <td>7230</td>\n",
       "      <td>7465.171197</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     side 1  side 2        side 3\n",
       "0      8403    6062  10361.382775\n",
       "1      8133    8748  11944.588440\n",
       "2      4346    2247   4892.517246\n",
       "3      2883    2983   4148.491051\n",
       "4      6268    1265   6394.376357\n",
       "..      ...     ...           ...\n",
       "995    8018    5214   9564.210370\n",
       "996    8844    6177  10787.569930\n",
       "997    3738    9852  10537.293201\n",
       "998    2806    4195   5046.945710\n",
       "999    1859    7230   7465.171197\n",
       "\n",
       "[1000 rows x 3 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a dataset with side 1 and side 2\n",
    "new_data = {\n",
    "    'side 1': np.random.randint(1000, 10001, size=1000),  # Generate random values for side 1\n",
    "    'side 2': np.random.randint(1000, 10001, size=1000)   # Generate random values for side 2\n",
    "}\n",
    "\n",
    "# Calculate side 3 using the Pythagorean theorem\n",
    "new_data['side 3'] = np.sqrt(new_data['side 1']**2 + new_data['side 2']**2)\n",
    "\n",
    "# Create a DataFrame\n",
    "test_df = pd.DataFrame(new_data)\n",
    "test_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8cac71bd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Assuming you have already defined your polynomial regression model and poly_reg\n",
    "X_test = test_df[['side 1', 'side 2']]\n",
    "y_test = test_df['side 3']\n",
    "\n",
    "# Transform the test data using the polynomial features\n",
    "X_test_poly = poly_reg.transform(X_test)\n",
    "\n",
    "# Predict side_3_predict using the trained polynomial regression model\n",
    "side_3_predict = model_linear_2.predict(X_test_poly)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e50a0060",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 1987.7736096505632\n",
      "R^2 Score: 0.4515843728152298\n",
      "Correlation Coefficient (R): 0.8440544105076819\n"
     ]
    }
   ],
   "source": [
    "# Calculate RMSE (Root Mean Squared Error)\n",
    "rmse = np.sqrt(metrics.mean_squared_error(y_test, side_3_predict))\n",
    "\n",
    "# Calculate R^2 score (Coefficient of determination)\n",
    "r2 = metrics.r2_score(y_test, side_3_predict)\n",
    "\n",
    "print('RMSE:', rmse)\n",
    "print('R^2 Score:', r2)\n",
    "print('Correlation Coefficient (R):', np.corrcoef(y_test, side_3_predict)[0, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "395921a2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(range(len(y_test)), y_test, color='green', label='Real Side 3')\n",
    "plt.scatter(range(len(side_3_predict)), side_3_predict, color='red', label='Predicted Side 3')\n",
    "\n",
    "plt.xlabel('Data Point')\n",
    "plt.ylabel('Side 3 Value')\n",
    "plt.legend()\n",
    "plt.title('Real vs. Predicted Side 3 Values')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b77bc8d",
   "metadata": {},
   "source": [
    "### Test 2: Dataset has nearly the same value range with value in original"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9d07f873",
   "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>side 1</th>\n",
       "      <th>side 2</th>\n",
       "      <th>side 3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>353</td>\n",
       "      <td>882</td>\n",
       "      <td>950.017368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>584</td>\n",
       "      <td>777</td>\n",
       "      <td>972.000514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>470</td>\n",
       "      <td>11</td>\n",
       "      <td>470.128706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>95</td>\n",
       "      <td>682</td>\n",
       "      <td>688.584781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>891</td>\n",
       "      <td>868</td>\n",
       "      <td>1243.907151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>999</td>\n",
       "      <td>983</td>\n",
       "      <td>1401.531305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>701</td>\n",
       "      <td>507</td>\n",
       "      <td>865.130048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>760</td>\n",
       "      <td>353</td>\n",
       "      <td>837.979117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>780</td>\n",
       "      <td>985</td>\n",
       "      <td>1256.433444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>912</td>\n",
       "      <td>554</td>\n",
       "      <td>1067.080128</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     side 1  side 2       side 3\n",
       "0       353     882   950.017368\n",
       "1       584     777   972.000514\n",
       "2       470      11   470.128706\n",
       "3        95     682   688.584781\n",
       "4       891     868  1243.907151\n",
       "..      ...     ...          ...\n",
       "995     999     983  1401.531305\n",
       "996     701     507   865.130048\n",
       "997     760     353   837.979117\n",
       "998     780     985  1256.433444\n",
       "999     912     554  1067.080128\n",
       "\n",
       "[1000 rows x 3 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create another dataset with side 1 and side 2\n",
    "new_data = {\n",
    "    'side 1': np.random.randint(10, 1001, size=1000),  # Generate random values for side 1\n",
    "    'side 2': np.random.randint(10, 1001, size=1000)   # Generate random values for side 2\n",
    "}\n",
    "\n",
    "# Calculate side 3 using the Pythagorean theorem\n",
    "new_data['side 3'] = np.sqrt(new_data['side 1']**2 + new_data['side 2']**2)\n",
    "\n",
    "# Create a DataFrame\n",
    "test_df = pd.DataFrame(new_data)\n",
    "test_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "533e9733",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Assuming you have already defined your polynomial regression model and poly_reg\n",
    "X_test = test_df[['side 1', 'side 2']]\n",
    "y_test = test_df['side 3']\n",
    "\n",
    "# Transform the test data using the polynomial features\n",
    "X_test_poly = poly_reg.transform(X_test)\n",
    "\n",
    "# Predict side_3_predict using the trained polynomial regression model\n",
    "side_3_predict = model_linear_2.predict(X_test_poly)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "8d57f298",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 96.91490682299869\n",
      "R^2 Score: 0.8752780266793132\n",
      "Correlation Coefficient (R): 0.9645784253215639\n"
     ]
    }
   ],
   "source": [
    "# Calculate RMSE (Root Mean Squared Error)\n",
    "rmse = np.sqrt(metrics.mean_squared_error(y_test, side_3_predict))\n",
    "\n",
    "# Calculate R^2 score (Coefficient of determination)\n",
    "r2 = metrics.r2_score(y_test, side_3_predict)\n",
    "\n",
    "print('RMSE:', rmse)\n",
    "print('R^2 Score:', r2)\n",
    "print('Correlation Coefficient (R):', np.corrcoef(y_test, side_3_predict)[0, 1])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "837dbe22",
   "metadata": {},
   "source": [
    "**When values in new dataset are significantly different from raw data -> higher RMSE**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7d97351",
   "metadata": {},
   "source": [
    "## The polynomial regression model\n",
    "\n",
    "1.  This makes it easier to identify the linear relationship between the independent and dependent variables, which results in a correlation coefficient of 1 and very low error metrics such as MSQ and RMSE. This indicates that the model has nearly 100% accuracy in predicting the correct answer for inputs that follow the **same rule** as the dataset (*side 1 is a multiple of 3, and side 2 is a multiple of 4,  then side 3 is precisely equal to the square root of the sum of the squares of side 1 and side 2, and side 3 is a multiple of 5 as well , know that they multiplied by the common factor*). Therefore, the **model gets the strongest performance on inputs that follow the same rule as the raw dataset and vice versa**. For example: *in training data, the side 1 is smaller than side 2, so when test values have side 1 values much higher than side 2 values, the results have more error than that of neural network.*\n",
    "\n",
    "\n",
    "2. Additionally, when I created a **new dataset with randomly generated values near with raw data** for side 1 and side 2, and calculated side 3 based on the right-angle triangle rule, the **polynomial regression model's performance decreased**, as expected. The RMSE increased (97.36 while it was just 6.3 when applied model with the raw dataset) and the R^2 decreased, but the R^2 score was still high at 0.88 (lower than when applied to the raw dataset – 0.9999). **On the contrary**, RMSE significantly rise and R^2 score reduce (under 0.5) when the input value is much higher than raw data. Therefore, **the higher the input (far with the training data), the more error the result.**\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bd7aa405",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}