Linear_Regression_Task2_200100173.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "89223f98",
   "metadata": {},
   "source": [
    "### Importing useful libraries \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "26f77ebe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# This Python 3 environment comes with many helpful analytics libraries installed\n",
    "# For example, here's several helpful packages to load in\n",
    "import numpy as np   # linear algebra\n",
    "import matplotlib.pyplot as plt # data visualization\n",
    "import pandas as pd  # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31c8220d",
   "metadata": {},
   "source": [
    "### Loading the dataset \n",
    "#### For implementation we will be using house prediction dataset . The dataset can be found [here](https://github.com/vrinda01go/Hellofoss/blob/main/Room_price_data.csv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1c5d873a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"Hostel_Linear-Dataset.csv\")  #import text file "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1ca9aba0",
   "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>Price</th>\n",
       "      <th>Hostel No.</th>\n",
       "      <th>Occupancy</th>\n",
       "      <th>Room Size</th>\n",
       "      <th>Floor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2540.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>686</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2900.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>966</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>788</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2362.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>924</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1098</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Price  Hostel No.  Occupancy  Room Size  Floor\n",
       "0  2540.0           3          1        686      8\n",
       "1  2900.0           3          2        966      5\n",
       "2     NaN           3          1        788      8\n",
       "3  2362.0           3          2        924      2\n",
       "4     NaN           3          2       1098      5"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af08f245",
   "metadata": {},
   "source": [
    "# Visualizing and Cleaning the data\n",
    "\n",
    "We will now be removing the nan values and identical values from the dataset\n",
    "\n",
    "For seeing if there are nan values in the dataset we will use the isna() function and then to remove them we will use the dropna() function. We will need to set additional parameters like rows and columns in the dropna function depending on the number of nan values present for each column\n",
    "\n",
    "Using the sum() function with isna() function we can get to know the number of missing values in each column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2fd4babb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Price         1531\n",
       "Hostel No.       0\n",
       "Occupancy        0\n",
       "Room Size        0\n",
       "Floor            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna()\n",
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83ef03c3",
   "metadata": {},
   "source": [
    "After this we will proceed to remove the nan values \n",
    "\n",
    "Since there are not many nan values in the column 'Price' as compared to the number of rows we will remove the rows which have nan values. \n",
    "\n",
    "Reseting the index after removing the nan values and dropping the old index will also be important"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b65e4503",
   "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>Price</th>\n",
       "      <th>Hostel No.</th>\n",
       "      <th>Occupancy</th>\n",
       "      <th>Room Size</th>\n",
       "      <th>Floor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2540.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>686</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2900.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>966</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2362.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>924</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1432.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>706</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1702.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1038</td>\n",
       "      <td>3</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>5002</th>\n",
       "      <td>1624.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>674</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5003</th>\n",
       "      <td>1470.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>626</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5004</th>\n",
       "      <td>1020.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1006</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5005</th>\n",
       "      <td>2400.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>938</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5006</th>\n",
       "      <td>950.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1053</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5007 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Price  Hostel No.  Occupancy  Room Size  Floor\n",
       "0     2540.0           3          1        686      8\n",
       "1     2900.0           3          2        966      5\n",
       "2     2362.0           3          2        924      2\n",
       "3     1432.0           2          1        706      3\n",
       "4     1702.0           2          2       1038      3\n",
       "...      ...         ...        ...        ...    ...\n",
       "5002  1624.0           2          1        674      3\n",
       "5003  1470.0           2          1        626      3\n",
       "5004  1020.0           3          2       1006      3\n",
       "5005  2400.0           2          2        938      1\n",
       "5006   950.0           3          2       1053      2\n",
       "\n",
       "[5007 rows x 5 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#fill\n",
    "df = df.dropna(axis=0, how='any', thresh=None, subset=None, inplace=False)\n",
    "df.reset_index(inplace = True)\n",
    "del df['index']\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "40784889",
   "metadata": {},
   "source": [
    "Now we can use the drop_duplicate function to remove the duplicate values\n",
    "\n",
    "This function has a parameter calle 'keep' where we specifiy to drop and which value to keep\n",
    "\n",
    "For this excercise we will keep the first values and drop the rest of the duplicates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "75fa3dc8",
   "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>Price</th>\n",
       "      <th>Hostel No.</th>\n",
       "      <th>Occupancy</th>\n",
       "      <th>Room Size</th>\n",
       "      <th>Floor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2540.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>686</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2900.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>966</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2362.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>924</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1432.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>706</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1702.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1038</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Price  Hostel No.  Occupancy  Room Size  Floor\n",
       "0  2540.0           3          1        686      8\n",
       "1  2900.0           3          2        966      5\n",
       "2  2362.0           3          2        924      2\n",
       "3  1432.0           2          1        706      3\n",
       "4  1702.0           2          2       1038      3"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#fill\n",
    "df = df.drop_duplicates(keep = 'first')\n",
    "df = df.reset_index(drop = True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a007a33f",
   "metadata": {},
   "source": [
    "For visualizing the data we will first start with looking at the distribution of different columns to see if there are enough number for each category in every column and dropping them if the data is biased for one category more than the other"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b325df62",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEICAYAAACzliQjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAZIklEQVR4nO3df5BV5Z3n8fcnYJwMrfgD04tAAqnC1ALOsKHLcZJNtrvirmicYHZndmDZKJPsoq47m9QmO+LMVHQnRZXzg2RXHc2SwUJXhg4VR3FUZnUSe81sJAYskgZ/JKidBLCgItjYhiID+e4f93Q4trdv33tu33O7eT6vqlt97/Occ8/3PDx8+txzz72tiMDMzNLwjnYXYGZm5XHom5klxKFvZpYQh76ZWUIc+mZmCXHom5klxKFvNk4kzZUUkqa2uxaz0Tj0bdKTNCDpshFtqyT9Q5PP2y1pX3PVveX5BiQdlDQt1/YfJPWN1zbMxuLQNyvXVOAz7S7C0uXQtyRI+qeS+iS9LmmPpI/n+q6U9JykNyTtl/T57Gh8G3ChpKHsdqGkd0haI+klSa9J2iLpvAZK+XPg85LOGaXOD0r6rqTB7OcHm9pxsxEc+nbak3QG8LfA48C7gd8HNkl6f7bIBuC6iDgLWAR8MyLeBK4ADkRER3Y7APwX4GrgXwAXAkeAv2ygnB1AH/D5KnWeBzwK3A6cD3wJeFTS+Q3tsFkNDn07XTyUHcW/Lul14K5c36VAB3BbRPw8Ir4JPAKsyPr/EVgg6eyIOBIRz9bYznXAH0XEvog4DtwK/HaDb95+Afh9SReMaP8Y8MOI+N8RcSIiNgMvAL/VwHOb1eTQt9PF1RFxzvAN+E+5vguBn0TEL3JtPwJmZff/DXAl8CNJ/1fSb9bYznuBB3O/XJ4HTgKd9RYaEbup/NJZM6LrwqyuvHydZk1z6FsKDgBzJOXn+3uA/QAR8d2IWEbl1M9DwJZsmWpfQfsT4Ir8L5iI+JWI2N9gTbcA/5G3BvoBKr9U8n5Zp9l4cOhbCr4DvAn8gaQzJHVTOWXSK+mdklZKmh4R/wgcpXLkDnAQOF/S9NxzfQVYK+m9AJIukLSs0YIiYi/wNSrvEQx7DLhI0r+TNFXS7wILqLwqMBsXDn077UXEz4GPU3lj9qdUzvdfExEvZIt8EhiQdBS4Hvj32XovAJuBl7PTORcC/xN4GHhc0hvAduA3Cpb2J8Avr9mPiNeAq4DPAa8BfwBcFRE/BciuOlpZcFtmAMh/RMXMLB0+0jczS4hD38wsIQ59M7OEOPTNzBIy4b8CdsaMGTF37txC67755ptMmzZt7AVL5roa47oa47oac7rWtXPnzp9GxMhPfUNETOjbkiVLoqgnn3yy8Lqt5Loa47oa47oac7rWBeyIKpnq0ztmZglx6JuZJcShb2aWEIe+mVlCHPpmZglx6JuZJcShb2aWEIe+mVlCHPpmZgmZ8F/DYDZR9e8fZNWaR9uy7YHbPtaW7drk5yN9M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS8iYoS/pHkmHJO3OtX1N0q7sNiBpV9Y+V9KxXN9XcusskdQvaa+k2yWpJXtkZmajqufDWRuBO4H7hhsi4neH70taBwzmln8pIhZXeZ67gdXAduAxYCmwreGKzcyssDGP9CPiKeBwtb7saP3fAptrPYekmcDZEfF09rcb7wOubrhaMzNriioZPMZC0lzgkYhYNKL9I8CXIqIrt9we4AfAUeCPI+JbkrqA2yLismy5DwM3RcRVo2xvNZVXBXR2di7p7e0ttHNDQ0N0dHQUWreVXFdjJmpdhw4PcvBYe7Z98azpo/ZN1PGarHX17x8cta+V5k2f0tR49fT07BzO5rxmv3tnBW89yn8VeE9EvCZpCfCQpIVAtfP3o/62iYj1wHqArq6u6O7uLlRcX18fRddtJdfVmIla1x2btrKuvz1fXzWwsnvUvok6XpO1rnZ9v9LGpdNaMl6FZ6ykqcC/BpYMt0XEceB4dn+npJeAi4B9wOzc6rOBA0W3bWZmxTRzyeZlwAsRsW+4QdIFkqZk998HzAdejohXgTckXZq9D3ANsLWJbZuZWQH1XLK5GXgaeL+kfZI+nXUt5+1v4H4E+L6k7wFfB66PiOE3gW8A/grYC7yEr9wxMyvdmKd3ImLFKO2rqrQ9ADwwyvI7gEXV+szMrBz+RK6ZWUIc+mZmCXHom5klxKFvZpYQh76ZWUIc+mZmCXHom5klxKFvZpYQh76ZWUIc+mZmCXHom5klxKFvZpYQh76ZWUIc+mZmCXHom5klxKFvZpYQh76ZWUIc+mZmCXHom5klpJ4/jH6PpEOSdufabpW0X9Ku7HZlru9mSXslvSjp8lz7Ekn9Wd/tkjT+u2NmZrXUc6S/EVhapf3LEbE4uz0GIGkBsBxYmK1zl6Qp2fJ3A6uB+dmt2nOamVkLjRn6EfEUcLjO51sG9EbE8Yh4BdgLXCJpJnB2RDwdEQHcB1xdsGYzMytIlQweYyFpLvBIRCzKHt8KrAKOAjuAz0XEEUl3Atsj4v5suQ3ANmAAuC0iLsvaPwzcFBFXjbK91VReFdDZ2bmkt7e30M4NDQ3R0dFRaN1Wcl2Nmah1HTo8yMFj7dn2xbOmj9o3UcdrstbVv3+wxGpOmTd9SlPj1dPTszMiuka2Ty34fHcDXwQi+7kO+BRQ7Tx91GivKiLWA+sBurq6oru7u1CRfX19FF23lVxXYyZqXXds2sq6/qL/hZozsLJ71L6JOl6Tta5Vax4tr5icjUuntWS8Cl29ExEHI+JkRPwC+CpwSda1D5iTW3Q2cCBrn12l3czMSlQo9LNz9MM+AQxf2fMwsFzSmZLmUXnD9pmIeBV4Q9Kl2VU71wBbm6jbzMwKGPO1qaTNQDcwQ9I+4BagW9JiKqdoBoDrACJij6QtwHPACeDGiDiZPdUNVK4EeheV8/zbxnE/zMysDmOGfkSsqNK8ocbya4G1Vdp3AIsaqs7MzMaVP5FrZpYQh76ZWUIc+mZmCXHom5klxKFvZpYQh76ZWULa8xnykvTvH2zLR6gHbvtY6ds0M6uHj/TNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEjBn6ku6RdEjS7lzbn0t6QdL3JT0o6Zysfa6kY5J2Zbev5NZZIqlf0l5Jt0tSS/bIzMxGVc+R/kZg6Yi2J4BFEfFrwA+Am3N9L0XE4ux2fa79bmA1MD+7jXxOMzNrsTFDPyKeAg6PaHs8Ik5kD7cDs2s9h6SZwNkR8XREBHAfcHWhis3MrDBVMniMhaS5wCMRsahK398CX4uI+7Pl9lA5+j8K/HFEfEtSF3BbRFyWrfNh4KaIuGqU7a2m8qqAzs7OJb29vUX2jUOHBzl4rNCqTbl41vSa/UNDQ3R0dJRUTf1cV2PaNb+g9hybqOM1Wevq3z9YYjWnzJs+panx6unp2RkRXSPbm/ojKpL+CDgBbMqaXgXeExGvSVoCPCRpIVDt/P2ov20iYj2wHqCrqyu6u7sL1XfHpq2s6y//78QMrOyu2d/X10fRfWol19WYds0vqD3HJup4Tda62vGHmAA2Lp3WkvEqPGMlXQtcBXw0O2VDRBwHjmf3d0p6CbgI2MdbTwHNBg4U3baZmRVT6JJNSUuBm4CPR8TPcu0XSJqS3X8flTdsX46IV4E3JF2aXbVzDbC16erNzKwhYx7pS9oMdAMzJO0DbqFytc6ZwBPZlZfbsyt1PgL8iaQTwEng+ogYfhP4BipXAr0L2JbdzMysRGOGfkSsqNK8YZRlHwAeGKVvB/C2N4LNzKw8/kSumVlCHPpmZglx6JuZJcShb2aWEIe+mVlCHPpmZglx6JuZJcShb2aWEIe+mVlCHPpmZglx6JuZJcShb2aWEIe+mVlCHPpmZglx6JuZJcShb2aWEIe+mVlCHPpmZglx6JuZJWTM0Jd0j6RDknbn2s6T9ISkH2Y/z8313Sxpr6QXJV2ea18iqT/ru13ZX1Q3M7Py1HOkvxFYOqJtDfCNiJgPfCN7jKQFwHJgYbbOXZKmZOvcDawG5me3kc9pZmYtNmboR8RTwOERzcuAe7P79wJX59p7I+J4RLwC7AUukTQTODsino6IAO7LrWNmZiVRJYPHWEiaCzwSEYuyx69HxDm5/iMRca6kO4HtEXF/1r4B2AYMALdFxGVZ+4eBmyLiqlG2t5rKqwI6OzuX9Pb2Ftq5Q4cHOXis0KpNuXjW9Jr9Q0NDdHR0lFRN/VxXY9o1v6D2HJuo4zVZ6+rfP1hiNafMmz6lqfHq6enZGRFdI9unNlXV21U7Tx812quKiPXAeoCurq7o7u4uVMwdm7ayrn+8d3FsAyu7a/b39fVRdJ9ayXU1pl3zC2rPsYk6XpO1rlVrHi2vmJyNS6e1ZLyKXr1zMDtlQ/bzUNa+D5iTW242cCBrn12l3czMSlQ09B8Grs3uXwtszbUvl3SmpHlU3rB9JiJeBd6QdGl21c41uXXMzKwkY742lbQZ6AZmSNoH3ALcBmyR9Gngx8DvAETEHklbgOeAE8CNEXEye6obqFwJ9C4q5/m3jeuemJnZmMYM/YhYMUrXR0dZfi2wtkr7DmBRQ9WZmdm48idyzcwS4tA3M0uIQ9/MLCEOfTOzhDj0zcwS4tA3M0uIQ9/MLCEOfTOzhDj0zcwS4tA3M0uIQ9/MLCEOfTOzhDj0zcwS4tA3M0uIQ9/MLCEOfTOzhDj0zcwS4tA3M0uIQ9/MLCGFQ1/S+yXtyt2OSvqspFsl7c+1X5lb52ZJeyW9KOny8dkFMzOr15h/GH00EfEisBhA0hRgP/Ag8HvAlyPiL/LLS1oALAcWAhcCfy/poog4WbQGMzNrzHid3vko8FJE/KjGMsuA3og4HhGvAHuBS8Zp+2ZmVgdFRPNPIt0DPBsRd0q6FVgFHAV2AJ+LiCOS7gS2R8T92TobgG0R8fUqz7caWA3Q2dm5pLe3t1Bdhw4PcvBYoVWbcvGs6TX7h4aG6OjoKKma+rmuxrRrfkHtOTZRx2uy1tW/f7DEak6ZN31KU+PV09OzMyK6RrYXPr0zTNI7gY8DN2dNdwNfBCL7uQ74FKAqq1f9jRMR64H1AF1dXdHd3V2otjs2bWVdf9O72LCBld01+/v6+ii6T63kuhrTrvkFtefYRB2vyVrXqjWPlldMzsal01oyXuNxeucKKkf5BwEi4mBEnIyIXwBf5dQpnH3AnNx6s4ED47B9MzOr03iE/gpg8/ADSTNzfZ8Admf3HwaWSzpT0jxgPvDMOGzfzMzq1NRrU0m/CvxL4Lpc859JWkzl1M3AcF9E7JG0BXgOOAHc6Ct3zMzK1VToR8TPgPNHtH2yxvJrgbXNbNPMzIrzJ3LNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIU2FvqQBSf2SdknakbWdJ+kJST/Mfp6bW/5mSXslvSjp8maLNzOzxozHkX5PRCyOiK7s8RrgGxExH/hG9hhJC4DlwEJgKXCXpCnjsH0zM6tTK07vLAPuze7fC1yda++NiOMR8QqwF7ikBds3M7NRKCKKryy9AhwBAvhfEbFe0usRcU5umSMRca6kO4HtEXF/1r4B2BYRX6/yvKuB1QCdnZ1Lent7C9V36PAgB48VWrUpF8+aXrN/aGiIjo6Okqqpn+tqTLvmF9SeYxN1vCZrXf37B0us5pR506c0NV49PT07c2dgfmlqU1XBhyLigKR3A09IeqHGsqrSVvU3TkSsB9YDdHV1RXd3d6Hi7ti0lXX9ze5i4wZWdtfs7+vro+g+tZLraky75hfUnmMTdbwma12r1jxaXjE5G5dOa8l4NXV6JyIOZD8PAQ9SOV1zUNJMgOznoWzxfcCc3OqzgQPNbN/MzBpTOPQlTZN01vB94F8Bu4GHgWuzxa4Ftmb3HwaWSzpT0jxgPvBM0e2bmVnjmnlt2gk8KGn4ef46Iv5O0neBLZI+DfwY+B2AiNgjaQvwHHACuDEiTjZVvZmZNaRw6EfEy8CvV2l/DfjoKOusBdYW3aaZmTXHn8g1M0uIQ9/MLCEOfTOzhDj0zcwS4tA3M0uIQ9/MLCEOfTOzhDj0zcwS4tA3M0uIQ9/MLCEOfTOzhDj0zcwS4tA3M0uIQ9/MLCEOfTOzhDj0zcwS4tA3M0uIQ9/MLCEOfTOzhBQOfUlzJD0p6XlJeyR9Jmu/VdJ+Sbuy25W5dW6WtFfSi5IuH48dMDOz+hX+w+jACeBzEfGspLOAnZKeyPq+HBF/kV9Y0gJgObAQuBD4e0kXRcTJJmowM7MGFD7Sj4hXI+LZ7P4bwPPArBqrLAN6I+J4RLwC7AUuKbp9MzNrnCKi+SeR5gJPAYuA/wqsAo4CO6i8Gjgi6U5ge0Tcn62zAdgWEV+v8nyrgdUAnZ2dS3p7ewvVdejwIAePFVq1KRfPml6zf2hoiI6OjpKqqZ/raky75hfUnmMTdbwma139+wdLrOaUedOnNDVePT09OyOia2R7M6d3AJDUATwAfDYijkq6G/giENnPdcCnAFVZvepvnIhYD6wH6Orqiu7u7kK13bFpK+v6m97Fhg2s7K7Z39fXR9F9aiXX1Zh2zS+oPccm6nhN1rpWrXm0vGJyNi6d1pLxaurqHUlnUAn8TRHxNwARcTAiTkbEL4CvcuoUzj5gTm712cCBZrZvZmaNaebqHQEbgOcj4ku59pm5xT4B7M7uPwwsl3SmpHnAfOCZots3M7PGNfPa9EPAJ4F+Sbuytj8EVkhaTOXUzQBwHUBE7JG0BXiOypU/N/rKHTOzchUO/Yj4B6qfp3+sxjprgbVFt2lmZs3xJ3LNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIQ59M7OEOPTNzBLi0DczS4hD38wsIaWHvqSlkl6UtFfSmrK3b2aWslJDX9IU4C+BK4AFwApJC8qswcwsZWUf6V8C7I2IlyPi50AvsKzkGszMkjW15O3NAn6Se7wP+I2RC0laDazOHg5JerHg9mYAPy24bmH60zEXaUtddXBdjWlbXWPMMY9XYyZkXT1/2nRd763WWHboq0pbvK0hYj2wvumNSTsioqvZ5xlvrqsxrqsxrqsxqdVV9umdfcCc3OPZwIGSazAzS1bZof9dYL6keZLeCSwHHi65BjOzZJV6eiciTkj6z8D/AaYA90TEnhZusulTRC3iuhrjuhrjuhqTVF2KeNspdTMzO035E7lmZglx6JuZJWRShr6keyQdkrR7lH5Juj37qofvS/pArq9lXwNRR10rs3q+L+nbkn491zcgqV/SLkk7Sq6rW9Jgtu1dkr6Q62vneP23XE27JZ2UdF7W18rxmiPpSUnPS9oj6TNVlil9jtVZV+lzrM66Sp9jddZV+hyT9CuSnpH0vayu/15lmdbNr4iYdDfgI8AHgN2j9F8JbKPyuYBLge9k7VOAl4D3Ae8EvgcsKLGuDwLnZvevGK4rezwAzGjTeHUDj1Rpb+t4jVj2t4BvljReM4EPZPfPAn4wcr/bMcfqrKv0OVZnXaXPsXrqasccy+ZMR3b/DOA7wKVlza9JeaQfEU8Bh2sssgy4Lyq2A+dImkmLvwZirLoi4tsRcSR7uJ3K5xRaro7xGk1bx2uEFcDm8dp2LRHxakQ8m91/A3ieyqfJ80qfY/XU1Y45Vud4jaat4zVCKXMsmzND2cMzstvIK2paNr8mZejXodrXPcyq0d4On6bym3xYAI9L2qnK11CU7Tezl5vbJC3M2ibEeEn6VWAp8ECuuZTxkjQX+GdUjsby2jrHatSVV/ocG6Outs2xscar7DkmaYqkXcAh4ImIKG1+lf01DGUZ7ese6voaiFaT1EPlP+Q/zzV/KCIOSHo38ISkF7Ij4TI8C7w3IoYkXQk8BMxngowXlZfd/y8i8q8KWj5ekjqohMBnI+LoyO4qq5Qyx8aoa3iZ0ufYGHW1bY7VM16UPMci4iSwWNI5wIOSFkVE/r2tls2v0/VIf7Sve2j710BI+jXgr4BlEfHacHtEHMh+HgIepPIyrhQRcXT45WZEPAacIWkGE2C8MssZ8bK71eMl6QwqQbEpIv6myiJtmWN11NWWOTZWXe2aY/WMV6b0OZY99+tAH5VXGXmtm1/j9eZE2TdgLqO/Mfkx3vomyDNZ+1TgZWAep94EWVhiXe8B9gIfHNE+DTgrd//bwNIS6/onnPqg3iXAj7Oxa+t4Zf3TqZz3n1bWeGX7fh/wP2osU/ocq7Ou0udYnXWVPsfqqasdcwy4ADgnu/8u4FvAVWXNr0l5ekfSZipXA8yQtA+4hcqbIUTEV4DHqLz7vRf4GfB7WV9Lvwaijrq+AJwP3CUJ4ERUvkWvk8pLPKj8o/51RPxdiXX9NnCDpBPAMWB5VGZYu8cL4BPA4xHxZm7Vlo4X8CHgk0B/dt4V4A+pBGo751g9dbVjjtVTVzvmWD11QflzbCZwryp/VOodwJaIeETS9bm6Wja//DUMZmYJOV3P6ZuZWRUOfTOzhDj0zcwS4tA3M0uIQ9/MLCEOfTOzhDj0zcwS8v8B7IAMVeZyJMQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "columns = df.columns\n",
    "df.plot()\n",
    "for column in columns:\n",
    "    df.hist(column)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9811a731",
   "metadata": {},
   "source": [
    "We can clearly notice that for the Occupancy column the (occupancy) = 4 has a really low set of data points as compared to others. Hence we can proceed in dropping those rows where the occupancy is 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "825783c0",
   "metadata": {},
   "outputs": [],
   "source": [
    "index_drp = df[df[\"Occupancy\"]==4].index\n",
    "df.drop(index_drp, inplace = True)\n",
    "df = df.reset_index(drop= True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "30c64310",
   "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>Price</th>\n",
       "      <th>Hostel No.</th>\n",
       "      <th>Occupancy</th>\n",
       "      <th>Room Size</th>\n",
       "      <th>Floor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2540.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>686</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2900.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>966</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2362.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>924</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1432.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>706</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1702.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1038</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Price  Hostel No.  Occupancy  Room Size  Floor\n",
       "0  2540.0           3          1        686      8\n",
       "1  2900.0           3          2        966      5\n",
       "2  2362.0           3          2        924      2\n",
       "3  1432.0           2          1        706      3\n",
       "4  1702.0           2          2       1038      3"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f333875b",
   "metadata": {},
   "source": [
    "We will now write the columns between categorical and numerical\n",
    "\n",
    "categorical = Hostel No, occupancy, floor\n",
    "\n",
    "Numerical = price, occupancy, roomsize, floor, hostel No.\n",
    "\n",
    "Remember that we can treat Hostel Number and occupancy as numerical or categorical. For this notebook we will treat them as categorical for data visualization and numerical for the regression"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f34ca6a",
   "metadata": {},
   "source": [
    "We will also plot the scatter plots and the correlation map to analyse the relation ships between different numerical columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f4a3ab6e",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "categorical = ['Hostel No.', 'Occupancy', 'Floor']\n",
    "numerical = [ 'Price', 'Room Size']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "df2b588a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for column1 in numerical:\n",
    "    for column2 in numerical:\n",
    "        if(column1 != column2):\n",
    "            sns.set(rc={\"figure.figsize\":(8, 8)}) \n",
    "            sns.scatterplot(x=column1, y=column2, data=df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "625b89a8",
   "metadata": {},
   "source": [
    "We can notice that there are no linear relation present between the numerical columns. Hence no need to drop anything"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a7f95fa",
   "metadata": {},
   "source": [
    "Now we will plot box plots of categorical and numerical columns to get more information about the number of outliers and the distrubtion."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e73f0769",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for c in categorical:\n",
    "    for n in numerical:\n",
    "        sns.boxplot(x= c, y= n, data=df)\n",
    "        plt.xlabel(c)\n",
    "        plt.ylabel(n)\n",
    "        plt.show()\n",
    "        #fill\n",
    "        # we can use the boxplot function of pandas (refer to the documentation)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a74d69fd",
   "metadata": {},
   "source": [
    "Now that we have analysed our data we can proceed to normalixing our data and regression"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "77d8c6ba",
   "metadata": {},
   "source": [
    "### Importing useful libraries \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "fffac537",
   "metadata": {},
   "outputs": [],
   "source": [
    "# This Python 3 environment comes with many helpful analytics libraries installed\n",
    "# For example, here's several helpful packages to load in\n",
    "import numpy as np   # linear algebra\n",
    "import matplotlib.pyplot as plt # data visualization\n",
    "import pandas as pd  # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "from sklearn import preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "04cd7e4e",
   "metadata": {},
   "source": [
    "### Loading the dataset \n",
    "#### For implementation we will be using house prediction dataset . The dataset can be found [here](https://github.com/vrinda01go/Hellofoss/blob/main/Insti_data.csv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "4ac8e74b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2.540e+03, 3.000e+00, 1.000e+00, 6.860e+02, 8.000e+00],\n",
       "       [2.900e+03, 3.000e+00, 2.000e+00, 9.660e+02, 5.000e+00],\n",
       "       [2.362e+03, 3.000e+00, 2.000e+00, 9.240e+02, 2.000e+00],\n",
       "       ...,\n",
       "       [1.020e+03, 3.000e+00, 2.000e+00, 1.006e+03, 3.000e+00],\n",
       "       [2.400e+03, 2.000e+00, 2.000e+00, 9.380e+02, 1.000e+00],\n",
       "       [9.500e+02, 3.000e+00, 2.000e+00, 1.053e+03, 2.000e+00]])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = np.array(df, dtype=float)\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "627a5a3a",
   "metadata": {},
   "source": [
    "#### Since our dataset has four features i.e Hostel No. , Occupancy, Room Size and Floor ,our hypothesis function becomes\n",
    "### hθ(x) = θ0 + θ1x1 + θ2x2 +θ3x3 + θ4x4\n",
    "#### where x1 ,x2,x3 and x4 are the two features (i.e. size of house and number of rooms)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58fdf0fe",
   "metadata": {},
   "source": [
    "### So Your task is to define hypothesis function having 4 features and a corresponding cost function "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "34dc5643",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define and complete hypothesis function\n",
    "#will use dot function of numpy for that as it will dot product theta which is 1*n array with n*1 array to give a single answer\n",
    "def hypothesis_fn(theta,X):\n",
    "    return X.dot(theta)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6fae141b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define and complete cost function\n",
    "def cost_out(theta,X,Y):\n",
    "    predict = hypothesis_fn(theta,X)\n",
    "    return (1/(2*len(Y)))*np.sum(np.square(predict-Y))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "178eb11f",
   "metadata": {},
   "source": [
    "### Gradient Descent \n",
    "#### So we have our hypothesis function and we have a way of measuring how well it fits into the data. Now we need to estimate the parameters in the hypothesis function. That's where gradient descent comes in.\n",
    "### Your next task is to define gradient descent function having some specific value of learning rate and number of epochs.\n",
    "#### Note that learning rate should be neither very high nor very low .Why?\n",
    "#### Check out exact reason [here](https://towardsdatascience.com/understanding-learning-rates-and-how-it-improves-performance-in-deep-learning-d0d4059c1c10)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "167b5bdc",
   "metadata": {},
   "outputs": [],
   "source": [
    "#define and complete Gradient Descent function \n",
    "def grad_descent(theta,X,Y,alpha,itr):\n",
    "    m=len(Y)\n",
    "    costs=[]\n",
    "    for i in range(itr):\n",
    "        predict = hypothesis_fn(theta,X)\n",
    "        theta = theta - (1/m)*alpha*np.sum(((predict -Y).dot(X)))\n",
    "        costs.append(cost_out(theta,X,Y))\n",
    "        \n",
    "    plt.plot(costs)\n",
    "    plt.title(\"Costs changing with variation in Epoch\")\n",
    "    plt.xlabel(\"Epoch\")\n",
    "    plt.ylabel(\"Cost\")\n",
    "    plt.show()\n",
    "        \n",
    "    return theta"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "724e407a",
   "metadata": {},
   "source": [
    "### Now we want to visualize how our cost function varies with number of epochs .So your next task is to plot graph of updated costs vs number of epochs "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e82f6ebf",
   "metadata": {},
   "source": [
    "#### After plotting above graph you will notice that your cost function decreases with epochs.\n",
    "#### Perfect! This is all what we wanted to seek by doing linear regression. \n",
    "\n",
    "#### Now it's time to test our model on some test data. \n",
    "\n",
    "#### For this you will define a test function that will take as input Hostel No. , Occupancy, Room Size , Floor and the final theta vector that was returned by our linear regression model and will give us the price of the house. Compute it for any set of features given and final value of theta as given by gradient descent function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "c522bca3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# define and complete test function that will take required inputs .This function should return price of Room \n",
    "X=data[:,1:]\n",
    "X=preprocessing.normalize(X)\n",
    "X=np.c_[np.ones((len(X),1)),X]\n",
    "Y=data[:,0]\n",
    "theta=np.zeros(5)\n",
    "theta_ideal=grad_descent(theta,X,Y,0.01,900)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "947f35b1",
   "metadata": {},
   "source": [
    "#### Now since we have defined all required functions , we can call functions one by one and get our final results .\n",
    "#### Your final task is to use all functions defined above and predict the price of room for some input combinations to check how well your model works."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06e54ef5",
   "metadata": {},
   "source": [
    "#### You can try playing with different values of alpha and epochs and see which combination gives most accurate results but do lookout for overfitting \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "3fc631fd",
   "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>Predicted values</th>\n",
       "      <th>Actual values</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1954.101284</td>\n",
       "      <td>2540.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1947.242187</td>\n",
       "      <td>2900.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1944.563312</td>\n",
       "      <td>2362.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1945.453175</td>\n",
       "      <td>1432.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1943.759426</td>\n",
       "      <td>1702.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4810</th>\n",
       "      <td>1945.842666</td>\n",
       "      <td>1624.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4811</th>\n",
       "      <td>1946.501423</td>\n",
       "      <td>1470.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4812</th>\n",
       "      <td>1944.927129</td>\n",
       "      <td>1020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4813</th>\n",
       "      <td>1942.393242</td>\n",
       "      <td>2400.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4814</th>\n",
       "      <td>1943.666595</td>\n",
       "      <td>950.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4815 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Predicted values  Actual values\n",
       "0          1954.101284         2540.0\n",
       "1          1947.242187         2900.0\n",
       "2          1944.563312         2362.0\n",
       "3          1945.453175         1432.0\n",
       "4          1943.759426         1702.0\n",
       "...                ...            ...\n",
       "4810       1945.842666         1624.0\n",
       "4811       1946.501423         1470.0\n",
       "4812       1944.927129         1020.0\n",
       "4813       1942.393242         2400.0\n",
       "4814       1943.666595          950.0\n",
       "\n",
       "[4815 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def test(X,Y):\n",
    "    display(pd.DataFrame({\"Predicted values\":X.dot(theta_ideal),\"Actual values\":Y}))\n",
    "\n",
    "test(X,Y)\n"
   ]
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}