applied_machine_learning_UK_car_data_final_project - ricky.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hlYctAC8GlGo",
        "outputId": "cb796809-3e5c-4165-eec0-40703438f1f4"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/dask/dataframe/__init__.py:42: FutureWarning: \n",
            "Dask dataframe query planning is disabled because dask-expr is not installed.\n",
            "\n",
            "You can install it with `pip install dask[dataframe]` or `conda install dask`.\n",
            "This will raise in a future version.\n",
            "\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        }
      ],
      "source": [
        "# Data manipulation and numerical operations\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "# Data visualization\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "# Machine learning and statistical models\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
        "from sklearn.linear_model import LinearRegression, LogisticRegression\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
        "from sklearn.cluster import KMeans\n",
        "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report\n",
        "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report, mean_absolute_percentage_error\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.model_selection import RandomizedSearchCV\n",
        "import xgboost as xgb\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import mean_absolute_percentage_error\n",
        "#!pip install catboost scikit-learn\n",
        "#from catboost import CatBoostRegressor\n",
        "\n",
        "#scale all numerical features and y data\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "\n",
        "# Deep learning\n",
        "from sklearn.cluster import KMeans\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Dropout, Conv2D, MaxPooling2D, Flatten\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.callbacks import EarlyStopping\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "from google.colab import drive\n",
        "from google.colab import files\n",
        "import pandas as pd\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from tabulate import tabulate\n",
        "import re\n",
        "import xgboost as xgb\n",
        "from sklearn.model_selection import GridSearchCV, train_test_split\n",
        "from sklearn.metrics import mean_absolute_percentage_error\n",
        "import lightgbm as lgb\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import mean_absolute_percentage_error, r2_score\n",
        "import joblib"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# # Please download csv dataset from kaggle: https://www.kaggle.com/datasets/deepcontractor/car-price-prediction-challenge and put it in GoogleDrive\n",
        "\n",
        "# # Mount Google Drive\n",
        "# drive.mount('/content/drive')\n",
        "\n",
        "# # Path to your file\n",
        "# file_path = '/content/drive/My Drive/car_price_prediction.csv'\n",
        "\n",
        "# # Load the data into a DataFrame\n",
        "# df = pd.read_csv(file_path)\n",
        "\n",
        "# df.head()"
      ],
      "metadata": {
        "id": "hxiCfKb0Gs2A"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# # Mount Google Drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "car_data_folder = '/content/drive/My Drive/cardata/'\n",
        "\n",
        "file_paths = {\n",
        "    'audi': car_data_folder + 'audi.csv',\n",
        "    'bmw': car_data_folder + 'bmw.csv',\n",
        "    #'cclass': car_data_folder + 'cclass.csv',\n",
        "    #'focus': car_data_folder + 'focus.csv',\n",
        "    'ford': car_data_folder + 'ford.csv',\n",
        "    'hyundi': car_data_folder + 'hyundi.csv',\n",
        "    'merc': car_data_folder + 'merc.csv',\n",
        "    'skoda': car_data_folder + 'skoda.csv',\n",
        "    'toyota': car_data_folder + 'toyota.csv',\n",
        "    'vauxhall': car_data_folder + 'vauxhall.csv',\n",
        "    'vw': car_data_folder + 'vw.csv'\n",
        "}\n",
        "\n",
        "car_data=[]\n",
        "\n",
        "for key,value in file_paths.items():\n",
        "  temporary_df=pd.read_csv(value)\n",
        "  temporary_df['brand']=str(key)\n",
        "  if key=='hyundi':\n",
        "    temporary_df.rename(columns={'tax(£)':'tax'},inplace=True)\n",
        "  car_data.append(temporary_df)\n",
        "\n",
        "# Concatenate all DataFrames in the list\n",
        "df = pd.concat(car_data, ignore_index=True)\n",
        "\n",
        "# Drop exact duplicate rows\n",
        "df.drop_duplicates(inplace=True)\n",
        "\n",
        "# Define the file path in your Google Drive\n",
        "#file_path = '/content/drive/My Drive/combinedcardatafinal.csv'\n",
        "\n",
        "# Save the DataFrame to the specified file path\n",
        "#df.to_csv(file_path, index=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BtxWXRpzHxay",
        "outputId": "6e92953a-22f0-4ffc-f185-799c322882d4"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 398
        },
        "id": "KEFg4T0-TmkQ",
        "outputId": "5d70cfb9-2f9c-44d3-f281-e42e1c439c7d"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "model           0\n",
              "year            0\n",
              "price           0\n",
              "transmission    0\n",
              "mileage         0\n",
              "fuelType        0\n",
              "tax             0\n",
              "mpg             0\n",
              "engineSize      0\n",
              "brand           0\n",
              "dtype: int64"
            ],
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>model</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>year</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>price</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>transmission</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mileage</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fuelType</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>tax</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mpg</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>engineSize</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>brand</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table><br><label><b>dtype:</b> int64</label>"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# List of variables to plot\n",
        "variables = ['year', 'price', 'transmission', 'mileage', 'fuelType', 'tax', 'mpg', 'engineSize', 'brand']\n",
        "\n",
        "# Create plots\n",
        "for var in variables:\n",
        "    plt.figure(figsize=(12, 5))\n",
        "\n",
        "    # Histogram\n",
        "    plt.subplot(1, 2, 1)\n",
        "    if df[var].dtype in ['int64', 'float64']:\n",
        "        sns.histplot(df[var], kde=True)\n",
        "    else:\n",
        "        sns.countplot(y=df[var])\n",
        "    plt.title(f'Histogram of {var}')\n",
        "\n",
        "    # Boxplot\n",
        "    plt.subplot(1, 2, 2)\n",
        "    if df[var].dtype in ['int64', 'float64']:\n",
        "        sns.boxplot(x=df[var])\n",
        "    else:\n",
        "        sns.boxplot(y=df[var])\n",
        "    plt.title(f'Boxplot of {var}')\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "Sw0egj8VQceR",
        "outputId": "ea8fc2b0-1b15-4490-c979-034091fbdbde"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Split labels\n",
        "column_names_list = list(df.columns)\n",
        "column_names_list.remove('price')\n",
        "X_train=df[column_names_list]\n",
        "Y_train=df['price']\n",
        "\n",
        "# Split the data into training and validation sets 60/20/20\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_train, Y_train, test_size=0.2, random_state=42)\n",
        "X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.25, random_state=42)"
      ],
      "metadata": {
        "id": "b70W-IKmSVhB"
      },
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Baseline Linear Regression model - using only numerical features\n",
        "model = LinearRegression()\n",
        "\n",
        "#define categorical and numerical features\n",
        "categorical_features = df.select_dtypes(include=['object', 'category']).columns.tolist()\n",
        "numerical_features = X_train.select_dtypes(include=['number']).columns.tolist()\n",
        "\n",
        "# Train the model\n",
        "model.fit(X_train[numerical_features], y_train)\n",
        "\n",
        "# Make predictions\n",
        "y_pred = model.predict(X_test[numerical_features])\n",
        "\n",
        "# Evaluate the model using MAPE and R^2\n",
        "mape = mean_absolute_percentage_error(y_test, y_pred)\n",
        "r2 = r2_score(y_test, y_pred)\n",
        "\n",
        "print(f'MAPE: {mape:.4f}')\n",
        "print(f'R^2: {r2:.4f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z9zTg-_vQgi_",
        "outputId": "0d7711f3-31b5-4552-9de1-ea8e94826862"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MAPE: 29.6399\n",
            "R^2: -20130230.9306\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Initialize the OneHotEncoder\n",
        "encoder = OneHotEncoder(sparse=False, handle_unknown='ignore')\n",
        "\n",
        "#Feature engineering for model 3\n",
        "categorical_features = df.select_dtypes(include=['object', 'category']).columns.tolist()\n",
        "numerical_features=df.select_dtypes(include=['number']).columns.tolist()\n",
        "\n",
        "# Fit and transform the categorical features of the training set\n",
        "df_cat = pd.DataFrame(encoder.fit_transform(df[categorical_features]), columns=encoder.get_feature_names_out(categorical_features))\n",
        "\n",
        "# Reset index to avoid index misalignment issues\n",
        "df_cat.reset_index(drop=True, inplace=True)\n",
        "df.reset_index(drop=True, inplace=True)\n",
        "\n",
        "# Concatenate the transformed categorical features and the original numerical features\n",
        "df = pd.concat([df_cat,df[numerical_features]], axis=1)\n",
        "\n",
        "print('df shape after onehot',df.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gbq67fbBSAvg",
        "outputId": "1282c2f7-1df6-4b7c-9e6c-085225e062e6"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/sklearn/preprocessing/_encoders.py:975: FutureWarning: `sparse` was renamed to `sparse_output` in version 1.2 and will be removed in 1.4. `sparse_output` is ignored unless you leave `sparse` to its default value.\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "df shape after onehot (97712, 219)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# KMeans - how many clusters?\n",
        "wcss = []\n",
        "for i in range(1, 11):\n",
        "    kmeans = KMeans(n_clusters=i, random_state=0)\n",
        "    kmeans.fit(df)\n",
        "    wcss.append(kmeans.inertia_)\n",
        "\n",
        "plt.plot(range(1, 11), wcss)\n",
        "plt.title('Elbow Method')\n",
        "plt.xlabel('Number of clusters')\n",
        "plt.ylabel('WCSS')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 849
        },
        "id": "b4FEGhDmTIhF",
        "outputId": "1171c08f-f89a-416e-9152-ab05692072bf"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#KMeans Clustering and Imputation using Cluster Average - optimal clusters = 10 from elbow method above\n",
        "\n",
        "#Before imputation\n",
        "print('Before:')\n",
        "print(df[numerical_features].isnull().sum())\n",
        "print((df['tax'] == 0).sum())\n",
        "print((df['mpg'] == 0).sum())\n",
        "print((df['engineSize'] == 0).sum())\n",
        "print('average of tax before imputation',df['tax'].mean())\n",
        "print('average of engineSize before imputation',df['engineSize'].mean())\n",
        "\n",
        "#impute zero values and nulls with cluster average\n",
        "# List of features including the target variable 'price' but without the two features that have a lot of 0s to impute\n",
        "features = df.drop(columns=['tax','engineSize']).columns.tolist()\n",
        "\n",
        "# Perform KMeans clustering with the optimal number of clusters 3\n",
        "n_clusters = 10\n",
        "#calculate clusters using KMeans then assign cluster labels to each row\n",
        "kmeans = KMeans(n_clusters=n_clusters, random_state=0)\n",
        "df['cluster_labels'] = kmeans.fit_predict(df[features])\n",
        "\n",
        "# Function to impute cluster averages\n",
        "def impute_cluster_average(df, column, cluster_column):\n",
        "    #list of unique cluster values in column\n",
        "    clusters = df[cluster_column].unique()\n",
        "    #for each cluster\n",
        "    for cluster in clusters:\n",
        "        #get all of the data in one cluster\n",
        "        cluster_data = df[df[cluster_column] == cluster]\n",
        "        #compute the cluster mean that is non-zero\n",
        "        cluster_mean = cluster_data[column][cluster_data[column] != 0].mean()\n",
        "\n",
        "        # Handle NaN values: Replace NaN cluster mean with global mean or another default value\n",
        "        if pd.isna(cluster_mean):\n",
        "            #if the cluster mean is null use the overall average of the column\n",
        "            cluster_mean = df[column][df[column] != 0].mean()\n",
        "        #replace 0s in the column within the cluster with the cluster mean\n",
        "        df.loc[(df[cluster_column] == cluster) & (df[column] == 0), column] = cluster_mean\n",
        "    return df\n",
        "\n",
        "# Impute cluster averages for the 'column_to_impute'\n",
        "df = impute_cluster_average(df, 'tax', 'cluster_labels')\n",
        "df = impute_cluster_average(df, 'engineSize', 'cluster_labels')\n",
        "\n",
        "print('average of tax after imputation',df['tax'].mean())\n",
        "print('average of engineSize after imputation',df['engineSize'].mean())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ggKuKf90WcHR",
        "outputId": "40615988-679e-4c57-9e00-f0ab25949af2"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Before:\n",
            "year          0\n",
            "price         0\n",
            "mileage       0\n",
            "tax           0\n",
            "mpg           0\n",
            "engineSize    0\n",
            "dtype: int64\n",
            "6259\n",
            "0\n",
            "268\n",
            "average of tax before imputation 120.14240830194858\n",
            "average of engineSize before imputation 1.6649132143441951\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/sklearn/cluster/_kmeans.py:1416: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
            "  super()._check_params_vs_input(X, default_n_init=10)\n",
            "<ipython-input-34-254f353fcbfd>:38: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '120.15709438230489' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n",
            "  df.loc[(df[cluster_column] == cluster) & (df[column] == 0), column] = cluster_mean\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "average of tax after imputation 127.45725832073938\n",
            "average of engineSize after imputation 1.6694732588882264\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Split labels\n",
        "column_names_list = list(df.columns)\n",
        "column_names_list.remove('price')\n",
        "X_train=df[column_names_list]\n",
        "Y_train=df['price']\n",
        "\n",
        "# Split the data into training and validation sets 60/20/20\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_train, Y_train, test_size=0.2, random_state=42)\n",
        "X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.25, random_state=42)\n",
        "\n",
        "# Verify the shape of the concatenated dataframes\n",
        "print(f\"X_train shape: {X_train.shape}\")\n",
        "print(f\"X_val shape: {X_val.shape}\")\n",
        "print(f\"X_test shape: {X_test.shape}\")\n",
        "\n",
        "# Print the shapes of the target dataframes\n",
        "print(f\"y_train shape: {y_train.shape}\")\n",
        "print(f\"y_val shape: {y_val.shape}\")\n",
        "print(f\"y_test shape: {y_test.shape}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tlUVGFW8TQEZ",
        "outputId": "e6215fe9-c3d8-4e98-8a50-aee73b89060d"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_train shape: (58626, 219)\n",
            "X_val shape: (19543, 219)\n",
            "X_test shape: (19543, 219)\n",
            "y_train shape: (58626,)\n",
            "y_val shape: (19543,)\n",
            "y_test shape: (19543,)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Do not run this cell\n",
        "#Do not run this cell\n",
        "#Do not run this cell\n",
        "#Do not run this cell\n",
        "\n",
        "#We will be scaling numerical features including one-hot encoded categorical features\n",
        "\n",
        "#scale all numerical features and y data\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "#inititalize scaler\n",
        "scaler=StandardScaler()\n",
        "\n",
        "#store categorical columns and numerical as a list\n",
        "categorical_features = X_train.select_dtypes(include=['object', 'category']).columns.tolist()\n",
        "numerical_features=X_train.select_dtypes(include=['number']).columns.tolist()\n",
        "\n",
        "# Fit and transform the numerical features of the training set\n",
        "X_train = pd.DataFrame(scaler.fit_transform(X_train[numerical_features]), columns=numerical_features)\n",
        "X_test = pd.DataFrame(scaler.transform(X_test[numerical_features]), columns=numerical_features)\n",
        "X_val = pd.DataFrame(scaler.transform(X_val[numerical_features]), columns=numerical_features)\n",
        "\n",
        "# Verify the shape of the concatenated dataframes\n",
        "print(f\"X_train shape: {X_train.shape}\")\n",
        "print(f\"X_val shape: {X_val.shape}\")\n",
        "print(f\"X_test shape: {X_test.shape}\")\n",
        "\n",
        "#scale y\n",
        "y_train=scaler.fit_transform(np.array(y_train).reshape(-1,1))\n",
        "y_test=scaler.fit_transform(np.array(y_test).reshape(-1,1))\n",
        "y_val=scaler.fit_transform(np.array(y_val).reshape(-1,1))\n",
        "\n",
        "# Print the shapes of the target dataframes\n",
        "print(f\"y_train shape: {y_train.shape}\")\n",
        "print(f\"y_val shape: {y_val.shape}\")\n",
        "print(f\"y_test shape: {y_test.shape}\")\n",
        "\n",
        "#shapes should match before because we are label encoding instead of one-hot encoding"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m8Jp8aW7YOrt",
        "outputId": "c215d3b4-21e0-42eb-fd98-6423325d66e1"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_train shape: (58626, 218)\n",
            "X_val shape: (19543, 218)\n",
            "X_test shape: (19543, 218)\n",
            "y_train shape: (58626, 1)\n",
            "y_val shape: (19543, 1)\n",
            "y_test shape: (19543, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Linear Regression model post data processing\n",
        "model = LinearRegression()\n",
        "\n",
        "# Train the model\n",
        "model.fit(X_train, y_train)\n",
        "\n",
        "# Make predictions\n",
        "y_pred = model.predict(X_test)\n",
        "\n",
        "# Evaluate the model using MAPE and R^2\n",
        "mape = mean_absolute_percentage_error(y_test, y_pred)\n",
        "r2 = r2_score(y_test, y_pred)\n",
        "\n",
        "print(f'MAPE: {mape:.4f}')\n",
        "print(f'R^2: {r2:.4f}')\n",
        "\n",
        "#MAPE to beat is 28%"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4r4oh4eZYQqu",
        "outputId": "57459d42-f5ce-46f7-f744-fc93695735d8"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MAPE: 93.9466\n",
            "R^2: -203960052.2815\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#random forest\n",
        "\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report, mean_absolute_percentage_error\n",
        "\n",
        "# Initialize the model\n",
        "rf_regressor = RandomForestRegressor(n_estimators=100, random_state=42)\n",
        "\n",
        "# Train the model\n",
        "rf_regressor.fit(X_train, y_train)\n",
        "\n",
        "# Predict on the validation set\n",
        "y_val_pred = rf_regressor.predict(X_val)\n",
        "\n",
        "# Predict on the validation set\n",
        "y_test_pred = rf_regressor.predict(X_test)\n",
        "\n",
        "# Calculate MAPE on the validation set\n",
        "val_mape = mean_absolute_percentage_error(y_val, y_val_pred)\n",
        "print(f'Validation MAPE: {val_mape:.2%}')\n",
        "# Calculate MAPE on the test set\n",
        "test_mape = mean_absolute_percentage_error(y_test, y_test_pred)\n",
        "print(f'Test MAPE: {test_mape:.2%}')\n",
        "\n",
        "\n",
        "#7.38%\n",
        "\n",
        "#7.01% one-hot encode, KMeans cluster average impute"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BqNoJJMfcMf_",
        "outputId": "1cd0b739-faf2-4ee1-e59d-19f46f4ed479"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Validation MAPE: 7.01%\n",
            "Test MAPE: 7.07%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#XGBoost\n",
        "\n",
        "import xgboost as xgb\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import mean_absolute_percentage_error\n",
        "\n",
        "# Convert to DMatrix\n",
        "dtrain = xgb.DMatrix(X_train, label=y_train)\n",
        "dtest = xgb.DMatrix(X_test, label=y_test)\n",
        "\n",
        "# Set parameters\n",
        "params = {\n",
        "    'booster': 'gbtree',\n",
        "    'objective': 'reg:squarederror',  # For regression\n",
        "    'eta': 0.1,\n",
        "    'max_depth': 20,\n",
        "    'subsample': 0.8,\n",
        "    'colsample_bytree': 0.8\n",
        "}\n",
        "\n",
        "# Train the model\n",
        "bst = xgb.train(params, dtrain, num_boost_round=100)\n",
        "\n",
        "# Make predictions\n",
        "y_pred = bst.predict(dtest)\n",
        "\n",
        "# Evaluate the model using MAPE\n",
        "mape = mean_absolute_percentage_error(y_test, y_pred)\n",
        "print(f'MAPE: {mape:.4f}')\n",
        "\n",
        "#6.93%\n",
        "\n",
        "#6.5%"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mqI6-IZQcM_F",
        "outputId": "6a669722-fbd5-43d7-e683-95407c0f4c24"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MAPE: 0.0651\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#XGBoost optimized\n",
        "\n",
        "import xgboost as xgb\n",
        "from sklearn.model_selection import GridSearchCV, train_test_split\n",
        "from sklearn.metrics import mean_absolute_percentage_error\n",
        "\n",
        "# Set up the parameter grid\n",
        "param_grid = {\n",
        "    'learning_rate': [0.01, 0.05, 0.1],\n",
        "    'max_depth': [15, 20, 25],\n",
        "    'min_child_weight': [1, 3, 5],\n",
        "    'subsample': [0.6, 0.8, 1.0],\n",
        "    'colsample_bytree': [0.6, 0.8, 1.0],\n",
        "    'n_estimators': [100, 200, 300]\n",
        "}\n",
        "\n",
        "# Initialize the XGBoost model\n",
        "xgb_model = xgb.XGBRegressor(objective='reg:squarederror', n_jobs=-1)\n",
        "\n",
        "# Initialize GridSearchCV\n",
        "grid_search = GridSearchCV(estimator=xgb_model, param_grid=param_grid, scoring='neg_mean_absolute_percentage_error', cv=3, verbose=1)\n",
        "\n",
        "# Fit GridSearchCV\n",
        "grid_search.fit(X_train, y_train)\n",
        "\n",
        "# Get the best parameters and score\n",
        "best_params = grid_search.best_params_\n",
        "best_score = -grid_search.best_score_\n",
        "\n",
        "print(f\"Best parameters found: {best_params}\")\n",
        "print(f\"Best MAPE: {best_score:.4f}\")\n",
        "\n",
        "# Train the final model with the best parameters\n",
        "final_model = xgb.XGBRegressor(objective='reg:squarederror', **best_params)\n",
        "final_model.fit(X_train, y_train)\n",
        "\n",
        "# Make predictions\n",
        "y_pred = final_model.predict(X_test)\n",
        "\n",
        "# Evaluate the model using MAPE\n",
        "mape = mean_absolute_percentage_error(y_test, y_pred)\n",
        "print(f'MAPE on test data: {mape:.4f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XPDwITEZgL_F",
        "outputId": "6cbd3f2e-9695-4d1a-adb7-135ce7440e21"
      },
      "execution_count": null,
      "outputs": [
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Fitting 3 folds for each of 729 candidates, totalling 2187 fits\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#try a model with dropouts and early stopping\n",
        "model_fnn_2 = Sequential([\n",
        "    Dense(128, input_dim=X_train.shape[1], activation='relu'),\n",
        "    Dropout(0.2),\n",
        "    Dense(64, activation='relu'),\n",
        "    Dropout(0.2),\n",
        "    Dense(32, activation='relu'),\n",
        "    Dense(1, activation='linear')\n",
        "])\n",
        "\n",
        "# Compile the model\n",
        "model_fnn_2.compile(optimizer='adam', loss='mean_absolute_percentage_error')\n",
        "\n",
        "early_stopping = EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)\n",
        "history = model_fnn_2.fit(X_train, y_train, epochs=100, batch_size=32, validation_split=0.2, callbacks=[early_stopping], verbose=True)\n",
        "\n",
        "# Make predictions\n",
        "y_pred = model_fnn_2.predict(X_test)\n",
        "\n",
        "# Evaluate the model using MAPE and R^2\n",
        "mape = mean_absolute_percentage_error(y_test, y_pred)\n",
        "r2 = r2_score(y_test, y_pred)\n",
        "\n",
        "print(f'MAPE: {mape:.4f}')\n",
        "print(f'R^2: {r2:.4f}')\n",
        "\n",
        "# 18%"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C4lFt8rhfF54",
        "outputId": "3588d366-a958-4d0c-8dfd-336e34858e51"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 5ms/step - loss: 47.7911 - val_loss: 34.8316\n",
            "Epoch 2/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3ms/step - loss: 36.0266 - val_loss: 34.2415\n",
            "Epoch 3/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3ms/step - loss: 35.1330 - val_loss: 33.9483\n",
            "Epoch 4/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 4ms/step - loss: 34.7220 - val_loss: 33.0662\n",
            "Epoch 5/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 2ms/step - loss: 32.4369 - val_loss: 27.3285\n",
            "Epoch 6/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - loss: 28.3152 - val_loss: 22.7137\n",
            "Epoch 7/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - loss: 25.4324 - val_loss: 21.0503\n",
            "Epoch 8/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - loss: 23.7736 - val_loss: 19.5011\n",
            "Epoch 9/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - loss: 23.0503 - val_loss: 18.8400\n",
            "Epoch 10/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3ms/step - loss: 22.2863 - val_loss: 20.9863\n",
            "Epoch 11/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3ms/step - loss: 22.1664 - val_loss: 18.8733\n",
            "Epoch 12/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - loss: 22.0289 - val_loss: 20.6363\n",
            "Epoch 13/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - loss: 21.4944 - val_loss: 19.0733\n",
            "Epoch 14/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - loss: 21.3820 - val_loss: 20.9414\n",
            "Epoch 15/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3ms/step - loss: 21.2078 - val_loss: 23.4940\n",
            "Epoch 16/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3ms/step - loss: 20.6982 - val_loss: 31.9636\n",
            "Epoch 17/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - loss: 20.3653 - val_loss: 19.5178\n",
            "Epoch 18/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 4ms/step - loss: 20.1280 - val_loss: 19.4353\n",
            "Epoch 19/100\n",
            "\u001b[1m1466/1466\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 2ms/step - loss: 20.4702 - val_loss: 19.1885\n",
            "\u001b[1m611/611\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step\n",
            "MAPE: 0.1876\n",
            "R^2: 0.6106\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Notes\n",
        "\n",
        "## Why does scaling reduce performance?\n",
        "\n",
        "## What should the order be -> clean, one-hot encode, impute, split, train, test, deploy?\n",
        "\n",
        "## how do we know we have a final model?\n",
        "\n"
      ],
      "metadata": {
        "id": "2Aj-6btufn40"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}