HouseRentModel.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "HouseRentModel.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "CXlHam57z7RM",
        "_ZSqSCc5z3wr"
      ],
      "mount_file_id": "1Fs59gPRF-IpNMnagjc4D4CgqTQ9Fjwnt",
      "authorship_tag": "ABX9TyMTX2njyPqT5cd9iOWnIfz4"
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-4VLCKGDRJI9"
      },
      "source": [
        "##Importing required libraries and data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6MT569JnjO75",
        "outputId": "25ce7c46-580c-480d-adc5-fbee18b97ddb"
      },
      "source": [
        "from google.colab import drive\r\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-6YGo_NEsog_"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "%matplotlib inline\n",
        " \n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
        "from sklearn.model_selection import GridSearchCV, RandomizedSearchCV\n",
        "from sklearn.linear_model import LinearRegression, Ridge, Lasso\n",
        "from sklearn.tree import DecisionTreeRegressor\n",
        "from sklearn.ensemble import RandomForestRegressor, GradientBoostingRegressor\n",
        "from xgboost import XGBRegressor\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.metrics import r2_score, mean_squared_error"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZV_idELlsORv"
      },
      "source": [
        "df = pd.read_csv('/content/drive/MyDrive/Datasets/Mumbai_house_rent_99acers/Mumbai_99acers_cleaned.csv')\r\n",
        "logdf = pd.read_csv('/content/drive/MyDrive/Datasets/Mumbai_house_rent_99acers/Mumbai_99acers_logtransformed.csv')"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2fsHj-hKvkry",
        "outputId": "c326002f-54b3-40a4-fb5c-744560f80a87"
      },
      "source": [
        "logdf.info()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 12797 entries, 0 to 12796\n",
            "Data columns (total 10 columns):\n",
            " #   Column         Non-Null Count  Dtype  \n",
            "---  ------         --------------  -----  \n",
            " 0   society        12797 non-null  object \n",
            " 1   type           12797 non-null  object \n",
            " 2   location       12797 non-null  object \n",
            " 3   bedrooms       12797 non-null  int64  \n",
            " 4   built-up area  12797 non-null  float64\n",
            " 5   furnishing     12797 non-null  object \n",
            " 6   age            12797 non-null  object \n",
            " 7   floor          12797 non-null  int64  \n",
            " 8   total floors   12797 non-null  int64  \n",
            " 9   monthly rent   12797 non-null  float64\n",
            "dtypes: float64(2), int64(3), object(5)\n",
            "memory usage: 999.9+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XOXM61MaFFLj"
      },
      "source": [
        "df['area-per-bedroom'] = df['built-up area']/df['bedrooms']"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 306
        },
        "id": "tR-mMHOFFKJZ",
        "outputId": "3f0f3e62-2380-4e85-a4c2-2b25ccb9ccf0"
      },
      "source": [
        "logdf['area-per-bedroom'] = logdf['built-up area']/logdf['bedrooms']\r\n",
        "logdf.head()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>society</th>\n",
              "      <th>type</th>\n",
              "      <th>location</th>\n",
              "      <th>bedrooms</th>\n",
              "      <th>built-up area</th>\n",
              "      <th>furnishing</th>\n",
              "      <th>age</th>\n",
              "      <th>floor</th>\n",
              "      <th>total floors</th>\n",
              "      <th>monthly rent</th>\n",
              "      <th>area-per-bedroom</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>other society</td>\n",
              "      <td>Residential Apartment</td>\n",
              "      <td>Bandra (West)</td>\n",
              "      <td>2</td>\n",
              "      <td>800.0</td>\n",
              "      <td>Furnished</td>\n",
              "      <td>5 to 10 years old</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>11.350407</td>\n",
              "      <td>400.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>other society</td>\n",
              "      <td>Residential Apartment</td>\n",
              "      <td>Khar West</td>\n",
              "      <td>2</td>\n",
              "      <td>1050.0</td>\n",
              "      <td>Furnished</td>\n",
              "      <td>1 to 5 years old</td>\n",
              "      <td>8</td>\n",
              "      <td>12</td>\n",
              "      <td>11.608236</td>\n",
              "      <td>525.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Gajra Bhoomi Symphony</td>\n",
              "      <td>Residential Apartment</td>\n",
              "      <td>Sector-20 Koparkhairane</td>\n",
              "      <td>2</td>\n",
              "      <td>1050.0</td>\n",
              "      <td>Unfurnished</td>\n",
              "      <td>1 to 5 years old</td>\n",
              "      <td>14</td>\n",
              "      <td>17</td>\n",
              "      <td>10.373491</td>\n",
              "      <td>525.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Crescent Bay</td>\n",
              "      <td>Residential Apartment</td>\n",
              "      <td>Parel</td>\n",
              "      <td>3</td>\n",
              "      <td>2275.0</td>\n",
              "      <td>Semifurnished</td>\n",
              "      <td>1 to 5 years old</td>\n",
              "      <td>40</td>\n",
              "      <td>41</td>\n",
              "      <td>11.918391</td>\n",
              "      <td>758.333333</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Ashwini CHS</td>\n",
              "      <td>Residential Apartment</td>\n",
              "      <td>Tilak Nagar</td>\n",
              "      <td>1</td>\n",
              "      <td>550.0</td>\n",
              "      <td>Semifurnished</td>\n",
              "      <td>5 to 10 years old</td>\n",
              "      <td>4</td>\n",
              "      <td>7</td>\n",
              "      <td>10.203592</td>\n",
              "      <td>550.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                 society                   type  ... monthly rent  area-per-bedroom\n",
              "0          other society  Residential Apartment  ...    11.350407        400.000000\n",
              "1          other society  Residential Apartment  ...    11.608236        525.000000\n",
              "2  Gajra Bhoomi Symphony  Residential Apartment  ...    10.373491        525.000000\n",
              "3           Crescent Bay  Residential Apartment  ...    11.918391        758.333333\n",
              "4            Ashwini CHS  Residential Apartment  ...    10.203592        550.000000\n",
              "\n",
              "[5 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dCrEBQlznCtj"
      },
      "source": [
        "## OneHotEncoding Text features"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tx7xliERoweE"
      },
      "source": [
        "temp = pd.get_dummies(df, drop_first=True)\r\n",
        "corr = temp.corr()['monthly rent'].sort_values()"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7KuF5E6nqyFE"
      },
      "source": [
        "logtemp = pd.get_dummies(logdf, drop_first=True)\r\n",
        "logcorr = logtemp.corr()['monthly rent'].sort_values()"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 499
        },
        "id": "y77OXlx-rPWF",
        "outputId": "1767d085-53fd-404c-99fb-452d562a33b5"
      },
      "source": [
        "plt.figure(figsize=(6,8))\r\n",
        "\r\n",
        "sns.scatterplot(y= logcorr.head(10).index.append(logcorr.tail(11)[:-1].index), \r\n",
        "                x= list(logcorr.head(10).values) + list(logcorr.tail(11)[:-1].values), label='log transformed')\r\n",
        "sns.scatterplot(y= logcorr.head(10).index.append(logcorr.tail(11)[:-1].index), \r\n",
        "                x= corr[logcorr.head(10).index.append(logcorr.tail(11)[:-1].index)].values, color='red', label='without log transformed')\r\n",
        "plt.title('Top 10 Positive and Negative correlations with monthly rent')\r\n",
        "plt.legend()\r\n",
        "plt.show()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w7Kfd1Zdev2L"
      },
      "source": [
        "## Preparing Training and Testing Datasets"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "53h668J7ucUE",
        "outputId": "bd285417-b477-41b7-d1a3-e851b3eb2989"
      },
      "source": [
        "x = temp.drop('monthly rent', axis = 1)\r\n",
        "y = temp['monthly rent']\r\n",
        "\r\n",
        "X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=42)\r\n",
        "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((10237, 356), (2560, 356), (10237,), (2560,))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E82FQjDgm7sB",
        "outputId": "78750efb-a0c7-4b81-cfcf-07d958676e15"
      },
      "source": [
        "logx = logtemp.drop('monthly rent', axis = 1)\r\n",
        "logy = logtemp['monthly rent']\r\n",
        "\r\n",
        "logx_train, logx_test, logy_train, logy_test = train_test_split(logx, logy, test_size=0.2, random_state=42)\r\n",
        "logx_train.shape, logx_test.shape, logy_train.shape, logy_test.shape"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((10237, 356), (2560, 356), (10237,), (2560,))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZoEQbj8-ev2W"
      },
      "source": [
        "## Scaling Datasets to convert them into same range."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e5cnl78Xe11c"
      },
      "source": [
        "linear = Pipeline([('scaler', StandardScaler()),\r\n",
        "                   ('model', LinearRegression())])"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8ku82TKofl0P",
        "outputId": "20621097-ad0c-4579-c499-5afff0ad7548"
      },
      "source": [
        "linear.fit(X_train, y_train)\r\n",
        "print('r2 score:',r2_score(y_test, linear.predict(X_test)))\r\n",
        "print('rmse:',mean_squared_error(y_test, linear.predict(X_test))**.5)"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "r2 score: 0.8850249399852936\n",
            "rmse: 12192.737654367309\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QxueZ-bvgmJB",
        "outputId": "6d574d0d-d872-4d83-f703-7b0cfb169cac"
      },
      "source": [
        "linear.fit(logx_train,logy_train)\r\n",
        "print('r2 score:',r2_score(logy_test, linear.predict(logx_test)))\r\n",
        "print('rmse:',mean_squared_error(logy_test, linear.predict(logx_test))**.5)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "r2 score: 0.9275648684792374\n",
            "rmse: 0.19852773815105582\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XZ3jsbHKhJCa"
      },
      "source": [
        "**Clearly log transformed help to improve score**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kfQoIcNCvMiF"
      },
      "source": [
        "## Linear models"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "id": "sDMZIt2LhISL",
        "outputId": "784d68f6-2224-453f-f174-c8566abb0128"
      },
      "source": [
        "Final_Scores = pd.DataFrame( index = ['With Training Data r2 score','With Test Data r2 score',\r\n",
        "                                       'With Training Data rmse', 'With Test Data rmse'])\r\n",
        "\r\n",
        "models = [LinearRegression(), Ridge(), Lasso()]\r\n",
        "names = ['LinearRegression', 'Ridge', 'Lasso']\r\n",
        "\r\n",
        "for name,model in zip(names,models):\r\n",
        "\r\n",
        "  pipeline = Pipeline([('scaler', StandardScaler()),\r\n",
        "                    (name, model)])\r\n",
        "  \r\n",
        "  pipeline.fit(logx_train,logy_train)\r\n",
        "\r\n",
        "  Final_Scores[name] = [round(r2_score(logy_train, pipeline.predict(logx_train)), 4), \r\n",
        "                         round(r2_score(logy_test, pipeline.predict(logx_test)), 4),\r\n",
        "                         round(mean_squared_error(logy_train, pipeline.predict(logx_train))**.5, 4),\r\n",
        "                         round(mean_squared_error(logy_test, pipeline.predict(logx_test))**.5, 4)]\r\n",
        "else:\r\n",
        "  display(Final_Scores)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "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>LinearRegression</th>\n",
              "      <th>Ridge</th>\n",
              "      <th>Lasso</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>With Training Data r2 score</th>\n",
              "      <td>0.9210</td>\n",
              "      <td>0.9209</td>\n",
              "      <td>0.0000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Test Data r2 score</th>\n",
              "      <td>0.9276</td>\n",
              "      <td>0.9276</td>\n",
              "      <td>-0.0001</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Training Data rmse</th>\n",
              "      <td>0.2071</td>\n",
              "      <td>0.2071</td>\n",
              "      <td>0.7365</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Test Data rmse</th>\n",
              "      <td>0.1985</td>\n",
              "      <td>0.1985</td>\n",
              "      <td>0.7377</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                             LinearRegression   Ridge   Lasso\n",
              "With Training Data r2 score            0.9210  0.9209  0.0000\n",
              "With Test Data r2 score                0.9276  0.9276 -0.0001\n",
              "With Training Data rmse                0.2071  0.2071  0.7365\n",
              "With Test Data rmse                    0.1985  0.1985  0.7377"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-MzT8w5dvWWf",
        "outputId": "dffcb46e-3e41-41b4-b2bc-b2d72b4cf85f"
      },
      "source": [
        "param_grid = {'alpha':[.0001,.001, .01, .1, 1, 10, 100]}\r\n",
        "\r\n",
        "model = GridSearchCV(Ridge(), param_grid, cv=5)\r\n",
        "model.fit(logx_train,logy_train)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5, error_score=nan,\n",
              "             estimator=Ridge(alpha=1.0, copy_X=True, fit_intercept=True,\n",
              "                             max_iter=None, normalize=False, random_state=None,\n",
              "                             solver='auto', tol=0.001),\n",
              "             iid='deprecated', n_jobs=None,\n",
              "             param_grid={'alpha': [0.0001, 0.001, 0.01, 0.1, 1, 10, 100]},\n",
              "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
              "             scoring=None, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-iVsaRNzw1Tb",
        "outputId": "12744975-6efe-4950-b29e-779b81fc50a6"
      },
      "source": [
        "model.best_params_, model.best_score_"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "({'alpha': 0.001}, 0.9170490887456196)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RTdpls0kHvBA",
        "outputId": "2a2103b3-83d8-40b8-af59-fc9bd6d8323e"
      },
      "source": [
        "print('r2 score:',r2_score(logy_test, model.predict(logx_test)))\r\n",
        "print('rmse:',mean_squared_error(logy_test, model.predict(logx_test))**.5)"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "r2 score: 0.9275667338831004\n",
            "rmse: 0.19852518181695236\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BjSCZpF8H8xB"
      },
      "source": [
        "# Tree based models using One Hot Encoding"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 190
        },
        "id": "7cYgsdXZICjT",
        "outputId": "84fc1b23-645e-4459-c7f4-968edb8d0115"
      },
      "source": [
        "models = [DecisionTreeRegressor(), RandomForestRegressor(), XGBRegressor(), GradientBoostingRegressor()]\r\n",
        "names = ['DecisionTreeRegressorOHE', 'RandomForestRegressorOHE', 'XGBRegressorOHE', 'GradientBoostingRegressorOHE']\r\n",
        "\r\n",
        "for name,model in zip(names,models):\r\n",
        "\r\n",
        "  model.fit(logx_train,logy_train)\r\n",
        "\r\n",
        "  Final_Scores[name] = [round(r2_score(logy_train, pipeline.predict(logx_train)), 4), \r\n",
        "                         round(r2_score(logy_test, pipeline.predict(logx_test)), 4),\r\n",
        "                         round(mean_squared_error(logy_train, pipeline.predict(logx_train))**.5, 4),\r\n",
        "                         round(mean_squared_error(logy_test, pipeline.predict(logx_test))**.5, 4)]\r\n",
        "else:\r\n",
        "  display(Final_Scores)"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[18:58:23] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "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>LinearRegression</th>\n",
              "      <th>Ridge</th>\n",
              "      <th>Lasso</th>\n",
              "      <th>DecisionTreeRegressorOHE</th>\n",
              "      <th>RandomForestRegressorOHE</th>\n",
              "      <th>XGBRegressorOHE</th>\n",
              "      <th>GradientBoostingRegressorOHE</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>With Training Data r2 score</th>\n",
              "      <td>0.9210</td>\n",
              "      <td>0.9209</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.0000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Test Data r2 score</th>\n",
              "      <td>0.9276</td>\n",
              "      <td>0.9276</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>-0.0001</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Training Data rmse</th>\n",
              "      <td>0.2071</td>\n",
              "      <td>0.2071</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.7365</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Test Data rmse</th>\n",
              "      <td>0.1985</td>\n",
              "      <td>0.1985</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.7377</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                             LinearRegression  ...  GradientBoostingRegressorOHE\n",
              "With Training Data r2 score            0.9210  ...                        0.0000\n",
              "With Test Data r2 score                0.9276  ...                       -0.0001\n",
              "With Training Data rmse                0.2071  ...                        0.7365\n",
              "With Test Data rmse                    0.1985  ...                        0.7377\n",
              "\n",
              "[4 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w7G0gKO1I0KK"
      },
      "source": [
        "As expected they can't work with OHE"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oyYBWK-5z9u3"
      },
      "source": [
        "## LabelEncoder"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 210
        },
        "id": "J9uW5cvu0chY",
        "outputId": "ab22cb79-6157-4004-ff7b-eb16b5dde39f"
      },
      "source": [
        "df1 = logdf.copy()\r\n",
        "df1.head(3)"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>society</th>\n",
              "      <th>type</th>\n",
              "      <th>location</th>\n",
              "      <th>bedrooms</th>\n",
              "      <th>built-up area</th>\n",
              "      <th>furnishing</th>\n",
              "      <th>age</th>\n",
              "      <th>floor</th>\n",
              "      <th>total floors</th>\n",
              "      <th>monthly rent</th>\n",
              "      <th>area-per-bedroom</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>other society</td>\n",
              "      <td>Residential Apartment</td>\n",
              "      <td>Bandra (West)</td>\n",
              "      <td>2</td>\n",
              "      <td>800.0</td>\n",
              "      <td>Furnished</td>\n",
              "      <td>5 to 10 years old</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>11.350407</td>\n",
              "      <td>400.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>other society</td>\n",
              "      <td>Residential Apartment</td>\n",
              "      <td>Khar West</td>\n",
              "      <td>2</td>\n",
              "      <td>1050.0</td>\n",
              "      <td>Furnished</td>\n",
              "      <td>1 to 5 years old</td>\n",
              "      <td>8</td>\n",
              "      <td>12</td>\n",
              "      <td>11.608236</td>\n",
              "      <td>525.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Gajra Bhoomi Symphony</td>\n",
              "      <td>Residential Apartment</td>\n",
              "      <td>Sector-20 Koparkhairane</td>\n",
              "      <td>2</td>\n",
              "      <td>1050.0</td>\n",
              "      <td>Unfurnished</td>\n",
              "      <td>1 to 5 years old</td>\n",
              "      <td>14</td>\n",
              "      <td>17</td>\n",
              "      <td>10.373491</td>\n",
              "      <td>525.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                 society                   type  ... monthly rent  area-per-bedroom\n",
              "0          other society  Residential Apartment  ...    11.350407             400.0\n",
              "1          other society  Residential Apartment  ...    11.608236             525.0\n",
              "2  Gajra Bhoomi Symphony  Residential Apartment  ...    10.373491             525.0\n",
              "\n",
              "[3 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_bKKKSE5pXhg"
      },
      "source": [
        "society_encoder = LabelEncoder().fit(df1['society'])\r\n",
        "df1['society'] = society_encoder.transform(df1['society'])\r\n",
        "\r\n",
        "type_encoder = LabelEncoder().fit(df1['type'])\r\n",
        "df1['type'] = type_encoder.transform(df1['type'])\r\n",
        "\r\n",
        "furnishing_encoder = LabelEncoder().fit(df1['furnishing'])\r\n",
        "df1['furnishing'] = furnishing_encoder.transform(df1['furnishing'])\r\n",
        "\r\n",
        "location_encoder = LabelEncoder().fit(df1['location'])\r\n",
        "df1['location'] = location_encoder.transform(df1['location'])\r\n",
        "\r\n",
        "age_encoder = LabelEncoder().fit(df1['age'])\r\n",
        "df1['age'] = age_encoder.transform(df1['age'])"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "yXqSMxjlpdnX",
        "outputId": "d98e35cc-6dc3-48ce-8ae4-28c0c531c556"
      },
      "source": [
        "df1"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>society</th>\n",
              "      <th>type</th>\n",
              "      <th>location</th>\n",
              "      <th>bedrooms</th>\n",
              "      <th>built-up area</th>\n",
              "      <th>furnishing</th>\n",
              "      <th>age</th>\n",
              "      <th>floor</th>\n",
              "      <th>total floors</th>\n",
              "      <th>monthly rent</th>\n",
              "      <th>area-per-bedroom</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>176</td>\n",
              "      <td>1</td>\n",
              "      <td>8</td>\n",
              "      <td>2</td>\n",
              "      <td>800.0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>11.350407</td>\n",
              "      <td>400.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>176</td>\n",
              "      <td>1</td>\n",
              "      <td>63</td>\n",
              "      <td>2</td>\n",
              "      <td>1050.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>8</td>\n",
              "      <td>12</td>\n",
              "      <td>11.608236</td>\n",
              "      <td>525.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>36</td>\n",
              "      <td>1</td>\n",
              "      <td>128</td>\n",
              "      <td>2</td>\n",
              "      <td>1050.0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>14</td>\n",
              "      <td>17</td>\n",
              "      <td>10.373491</td>\n",
              "      <td>525.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>20</td>\n",
              "      <td>1</td>\n",
              "      <td>88</td>\n",
              "      <td>3</td>\n",
              "      <td>2275.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>40</td>\n",
              "      <td>41</td>\n",
              "      <td>11.918391</td>\n",
              "      <td>758.333333</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>8</td>\n",
              "      <td>1</td>\n",
              "      <td>150</td>\n",
              "      <td>1</td>\n",
              "      <td>550.0</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>7</td>\n",
              "      <td>10.203592</td>\n",
              "      <td>550.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12792</th>\n",
              "      <td>176</td>\n",
              "      <td>1</td>\n",
              "      <td>94</td>\n",
              "      <td>3</td>\n",
              "      <td>2529.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>12</td>\n",
              "      <td>32</td>\n",
              "      <td>12.072541</td>\n",
              "      <td>843.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12793</th>\n",
              "      <td>150</td>\n",
              "      <td>1</td>\n",
              "      <td>83</td>\n",
              "      <td>3</td>\n",
              "      <td>1382.0</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>27</td>\n",
              "      <td>40</td>\n",
              "      <td>10.778956</td>\n",
              "      <td>460.666667</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12794</th>\n",
              "      <td>176</td>\n",
              "      <td>1</td>\n",
              "      <td>165</td>\n",
              "      <td>3</td>\n",
              "      <td>1750.0</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>9</td>\n",
              "      <td>16</td>\n",
              "      <td>12.206073</td>\n",
              "      <td>583.333333</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12795</th>\n",
              "      <td>176</td>\n",
              "      <td>1</td>\n",
              "      <td>109</td>\n",
              "      <td>2</td>\n",
              "      <td>950.0</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>8</td>\n",
              "      <td>10</td>\n",
              "      <td>10.146434</td>\n",
              "      <td>475.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12796</th>\n",
              "      <td>176</td>\n",
              "      <td>1</td>\n",
              "      <td>8</td>\n",
              "      <td>3</td>\n",
              "      <td>2500.0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>5</td>\n",
              "      <td>10</td>\n",
              "      <td>11.695247</td>\n",
              "      <td>833.333333</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>12797 rows × 11 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       society  type  location  ...  total floors  monthly rent  area-per-bedroom\n",
              "0          176     1         8  ...             6     11.350407        400.000000\n",
              "1          176     1        63  ...            12     11.608236        525.000000\n",
              "2           36     1       128  ...            17     10.373491        525.000000\n",
              "3           20     1        88  ...            41     11.918391        758.333333\n",
              "4            8     1       150  ...             7     10.203592        550.000000\n",
              "...        ...   ...       ...  ...           ...           ...               ...\n",
              "12792      176     1        94  ...            32     12.072541        843.000000\n",
              "12793      150     1        83  ...            40     10.778956        460.666667\n",
              "12794      176     1       165  ...            16     12.206073        583.333333\n",
              "12795      176     1       109  ...            10     10.146434        475.000000\n",
              "12796      176     1         8  ...            10     11.695247        833.333333\n",
              "\n",
              "[12797 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kBHjRjQ8ev2u"
      },
      "source": [
        "## Tree based models"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PijnObFv2WF8",
        "outputId": "e43227fa-28e5-4fd8-8a41-431fcf041ee5"
      },
      "source": [
        "x_train, x_test, y_train, y_test = train_test_split(df1.drop('monthly rent', axis = 1), df1['monthly rent'], test_size=0.2, random_state=42)\r\n",
        "x_train.shape, x_test.shape, y_train.shape, y_test.shape "
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((10237, 10), (2560, 10), (10237,), (2560,))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 414
        },
        "id": "CbR-06qMev2u",
        "outputId": "08408edb-c68a-40f8-b828-74628742ae87"
      },
      "source": [
        "models = [DecisionTreeRegressor(), RandomForestRegressor(), XGBRegressor(), GradientBoostingRegressor()]\n",
        "names = ['DecisionTreeRegressorLE', 'RandomForestRegressorLE', 'XGBRegressorLE', 'GradientBoostingRegressorLE']\n",
        "\n",
        "for name,model in zip(names,models):\n",
        "\n",
        "  model.fit(x_train,y_train)\n",
        "\n",
        "  Final_Scores[name] = [round(r2_score(y_train, model.predict(x_train)), 4), \n",
        "                         round(r2_score(y_test, model.predict(x_test)), 4),\n",
        "                         round(mean_squared_error(y_train, model.predict(x_train))**.5, 4),\n",
        "                         round(mean_squared_error(y_test, model.predict(x_test))**.5, 4)]\n",
        "else:\n",
        "  display(Final_Scores)"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[18:58:40] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "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>LinearRegression</th>\n",
              "      <th>Ridge</th>\n",
              "      <th>Lasso</th>\n",
              "      <th>DecisionTreeRegressorOHE</th>\n",
              "      <th>RandomForestRegressorOHE</th>\n",
              "      <th>XGBRegressorOHE</th>\n",
              "      <th>GradientBoostingRegressorOHE</th>\n",
              "      <th>DecisionTreeRegressorLE</th>\n",
              "      <th>RandomForestRegressorLE</th>\n",
              "      <th>XGBRegressorLE</th>\n",
              "      <th>GradientBoostingRegressorLE</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>With Training Data r2 score</th>\n",
              "      <td>0.9210</td>\n",
              "      <td>0.9209</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>0.9988</td>\n",
              "      <td>0.9779</td>\n",
              "      <td>0.8047</td>\n",
              "      <td>0.8064</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Test Data r2 score</th>\n",
              "      <td>0.9276</td>\n",
              "      <td>0.9276</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>-0.0001</td>\n",
              "      <td>0.7221</td>\n",
              "      <td>0.8536</td>\n",
              "      <td>0.7959</td>\n",
              "      <td>0.7986</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Training Data rmse</th>\n",
              "      <td>0.2071</td>\n",
              "      <td>0.2071</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.7365</td>\n",
              "      <td>0.0251</td>\n",
              "      <td>0.1095</td>\n",
              "      <td>0.3255</td>\n",
              "      <td>0.3241</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>With Test Data rmse</th>\n",
              "      <td>0.1985</td>\n",
              "      <td>0.1985</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.7377</td>\n",
              "      <td>0.3888</td>\n",
              "      <td>0.2822</td>\n",
              "      <td>0.3332</td>\n",
              "      <td>0.3310</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                             LinearRegression  ...  GradientBoostingRegressorLE\n",
              "With Training Data r2 score            0.9210  ...                       0.8064\n",
              "With Test Data r2 score                0.9276  ...                       0.7986\n",
              "With Training Data rmse                0.2071  ...                       0.3241\n",
              "With Test Data rmse                    0.1985  ...                       0.3310\n",
              "\n",
              "[4 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CXlHam57z7RM"
      },
      "source": [
        "##Decisiontree gridsearch"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vrW3Qi36yCsg",
        "outputId": "f084b76f-722d-4cea-89c9-b1d252ad94b7"
      },
      "source": [
        "param_grid = {'max_depth':[3,4,5,6,7,10,15,20],\r\n",
        "              'min_samples_leaf':[2,3,4,5,6]}\r\n",
        "\r\n",
        "model = GridSearchCV(DecisionTreeRegressor(), param_grid, cv=5)\r\n",
        "model.fit(x_train,y_train)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=5, error_score=nan,\n",
              "             estimator=DecisionTreeRegressor(ccp_alpha=0.0, criterion='mse',\n",
              "                                             max_depth=None, max_features=None,\n",
              "                                             max_leaf_nodes=None,\n",
              "                                             min_impurity_decrease=0.0,\n",
              "                                             min_impurity_split=None,\n",
              "                                             min_samples_leaf=1,\n",
              "                                             min_samples_split=2,\n",
              "                                             min_weight_fraction_leaf=0.0,\n",
              "                                             presort='deprecated',\n",
              "                                             random_state=None,\n",
              "                                             splitter='best'),\n",
              "             iid='deprecated', n_jobs=None,\n",
              "             param_grid={'max_depth': [3, 4, 5, 6, 7, 10, 15, 20],\n",
              "                         'min_samples_leaf': [2, 3, 4, 5, 6]},\n",
              "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
              "             scoring=None, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CwtTNuM1yCpw",
        "outputId": "39637b22-7067-4d2a-d3b9-8f222b857a1b"
      },
      "source": [
        "model.best_params_, model.best_score_"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "({'max_depth': 15, 'min_samples_leaf': 6}, 0.7482444408521531)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cGAanbqjyCm4",
        "outputId": "17d63dcb-23d3-4107-9586-ae1940a55e45"
      },
      "source": [
        "round(mean_squared_error(y_train, model.predict(x_train))**.5, 4)"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.2478"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CjTse2v6zMxp",
        "outputId": "fa1f07e0-f01a-4e12-d72d-4a7b82d8dafe"
      },
      "source": [
        "round(mean_squared_error(y_test, model.predict(x_test))**.5, 4)"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.3509"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_ZSqSCc5z3wr"
      },
      "source": [
        "##RandomForest gridsearch"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LrssRdkHz3DO",
        "outputId": "51a8eaf9-4c5d-418a-8b15-62a749efb5b3"
      },
      "source": [
        "param_grid = {'n_estimators':[50,100,150,200],\r\n",
        "              'max_depth':[3,4,5,6,7,10,15,20],\r\n",
        "              'min_samples_leaf':[2,3,4,5,6]}\r\n",
        "\r\n",
        "model = RandomizedSearchCV(RandomForestRegressor(), param_grid, cv=5)\r\n",
        "model.fit(x_train,y_train)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomizedSearchCV(cv=5, error_score=nan,\n",
              "                   estimator=RandomForestRegressor(bootstrap=True,\n",
              "                                                   ccp_alpha=0.0,\n",
              "                                                   criterion='mse',\n",
              "                                                   max_depth=None,\n",
              "                                                   max_features='auto',\n",
              "                                                   max_leaf_nodes=None,\n",
              "                                                   max_samples=None,\n",
              "                                                   min_impurity_decrease=0.0,\n",
              "                                                   min_impurity_split=None,\n",
              "                                                   min_samples_leaf=1,\n",
              "                                                   min_samples_split=2,\n",
              "                                                   min_weight_fraction_leaf=0.0,\n",
              "                                                   n_estimators=100,\n",
              "                                                   n_jobs=None, oob_score=False,\n",
              "                                                   random_state=None, verbose=0,\n",
              "                                                   warm_start=False),\n",
              "                   iid='deprecated', n_iter=10, n_jobs=None,\n",
              "                   param_distributions={'max_depth': [3, 4, 5, 6, 7, 10, 15,\n",
              "                                                      20],\n",
              "                                        'min_samples_leaf': [2, 3, 4, 5, 6],\n",
              "                                        'n_estimators': [50, 100, 150, 200]},\n",
              "                   pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
              "                   return_train_score=False, scoring=None, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "14lG17ja0i0I",
        "outputId": "0fd43299-0649-450f-ece2-0ec0042fe384"
      },
      "source": [
        "model.best_params_, model.best_score_"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "({'max_depth': 15, 'min_samples_leaf': 6, 'n_estimators': 100},\n",
              " 0.8157228142491405)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RDWlSmnT0pmu",
        "outputId": "493bbcff-4889-4d1e-9686-43d355e59735"
      },
      "source": [
        "round(mean_squared_error(y_train, model.predict(x_train))**.5, 4)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.2368"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "70hIMyNB0pj4",
        "outputId": "47ee35fe-0abb-4e1b-f4d2-a9331e77f792"
      },
      "source": [
        "round(mean_squared_error(y_test, model.predict(x_test))**.5, 4)"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.3084"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uBmWbL8TrNyj"
      },
      "source": [
        "##XGBRegressor gridsearch"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1PZeOoYHrOzX",
        "outputId": "71b8ae62-a68e-40b8-9f66-19f5124c3e22"
      },
      "source": [
        "parameters = {'objective':['reg:linear'],\r\n",
        "              'learning_rate': [.03, 0.05, .07],\r\n",
        "              'max_depth': [5, 6, 7, 8],\r\n",
        "              'n_estimators': [200,300,400,500]}\r\n",
        "\r\n",
        "model = RandomizedSearchCV(XGBRegressor(), parameters, cv = 5)\r\n",
        "model.fit(x_train,y_train)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[19:09:38] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:09:41] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:09:45] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:09:49] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:09:53] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:09:56] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:09:58] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:09:59] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:01] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:02] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:04] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:08] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:11] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:15] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:19] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:22] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:26] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:29] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:33] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:37] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:40] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:43] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:45] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:47] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:50] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:52] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:55] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:10:59] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:02] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:05] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:08] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:11] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:14] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:17] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:20] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:24] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:25] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:27] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:29] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:31] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:32] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:35] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:39] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:42] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:45] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:48] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:50] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:52] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:54] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:56] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[19:11:58] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomizedSearchCV(cv=5, error_score=nan,\n",
              "                   estimator=XGBRegressor(base_score=0.5, booster='gbtree',\n",
              "                                          colsample_bylevel=1,\n",
              "                                          colsample_bynode=1,\n",
              "                                          colsample_bytree=1, gamma=0,\n",
              "                                          importance_type='gain',\n",
              "                                          learning_rate=0.1, max_delta_step=0,\n",
              "                                          max_depth=3, min_child_weight=1,\n",
              "                                          missing=None, n_estimators=100,\n",
              "                                          n_jobs=1, nthread=None,\n",
              "                                          objective='reg:linear',\n",
              "                                          random_state=0, reg_alpha=0,\n",
              "                                          reg_lambda=1, scale_pos_weight=1,\n",
              "                                          seed=None, silent=None, subsample=1,\n",
              "                                          verbosity=1),\n",
              "                   iid='deprecated', n_iter=10, n_jobs=None,\n",
              "                   param_distributions={'learning_rate': [0.03, 0.05, 0.07],\n",
              "                                        'max_depth': [5, 6, 7, 8],\n",
              "                                        'n_estimators': [200, 300, 400, 500],\n",
              "                                        'objective': ['reg:linear']},\n",
              "                   pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
              "                   return_train_score=False, scoring=None, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SqLgpPPbsOK1",
        "outputId": "2bd25ba7-0d33-48b1-8bf5-a0f98d7b7497"
      },
      "source": [
        "model.best_params_, model.best_score_"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "({'learning_rate': 0.07,\n",
              "  'max_depth': 6,\n",
              "  'n_estimators': 500,\n",
              "  'objective': 'reg:linear'},\n",
              " 0.9150054694964519)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O3qZFLmHsnsj",
        "outputId": "e9baf02e-ce32-4508-d5d1-1e11a906f77f"
      },
      "source": [
        "round(mean_squared_error(y_train, model.predict(x_train))**.5, 4)"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.1242"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iEKL1Rqssnp3",
        "outputId": "c4858cb7-77d1-4b62-ce6e-70ddc9d993c9"
      },
      "source": [
        "round(mean_squared_error(y_test, model.predict(x_test))**.5, 4)"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.1958"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "R2kSdWRXuYPe",
        "outputId": "ac4946a8-ec1d-402b-e1c2-ca1651487993"
      },
      "source": [
        "x_train.iloc[[0,]], y_train.iloc[0]"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(      society  type  location  ...  floor  total floors  area-per-bedroom\n",
              " 3781       93     1        29  ...     18            19             459.0\n",
              " \n",
              " [1 rows x 10 columns], 9.546812608597397)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UP-Iqf1IuT7a",
        "outputId": "32d99983-e537-4ef0-b4d6-7cadf84f5565"
      },
      "source": [
        "model.predict(x_train.iloc[[0,]]), y_train.iloc[0]"
      ],
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([9.554807], dtype=float32), 9.546812608597397)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-vdXZN0ovNyb",
        "outputId": "cb0aab7b-699f-4c44-8d39-fd275e0df9a0"
      },
      "source": [
        "np.exp(9.554807), np.exp(9.546812608597397)"
      ],
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(14112.370046235981, 14000.000000000025)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qI8mc0pf2ESb"
      },
      "source": [
        "Linear Regression and XGBoost with parameter tuning perform comparable."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BW9r2KUd2IRH"
      },
      "source": [
        "# Winner is **Linear Regression and XGBoost Regressor** with rmse of .19:)"
      ]
    }
  ]
}