mlessentials/Lab07/Exercise7.09/Exercise7_09.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "2fCtd4kCt__z"
   },
   "source": [
    "# Lasso Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "jG6YpAbot__3"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression, Lasso\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import MinMaxScaler, PolynomialFeatures"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "sGKCHZb0t__8"
   },
   "outputs": [],
   "source": [
    "_df = pd.read_csv('../Dataset/ccpp.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 230
    },
    "colab_type": "code",
    "id": "MJzrSgbzuAAB",
    "outputId": "efbd59f2-3b49-44f6-b186-e1d275f13cc6"
   },
   "outputs": [],
   "source": [
    "_df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "WVkCs_VpuAAH"
   },
   "outputs": [],
   "source": [
    "X = _df.drop(['PE'], axis=1).values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "J4PHvpCzuAAM"
   },
   "outputs": [],
   "source": [
    "y = _df['PE'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "tFOurIZ-uAAW"
   },
   "outputs": [],
   "source": [
    "train_X, eval_X, train_y, eval_y = train_test_split(X, y, train_size=0.8, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "qChyCx_QuAAu"
   },
   "source": [
    "# Implement a LinearRegression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "mp6CUsPDuAA0"
   },
   "outputs": [],
   "source": [
    "lr_model_1 = LinearRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "f9w8TJiXuABY",
    "outputId": "f38d02e0-36e7-4aa5-dc4a-fb3adf505ee0"
   },
   "outputs": [],
   "source": [
    "lr_model_1.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "MEGafhCSuABo"
   },
   "outputs": [],
   "source": [
    "lr_model_1_preds = lr_model_1.predict(eval_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "WV0MFrBzuAB5",
    "outputId": "58a1231f-3e1c-452c-e799-307254d3e149"
   },
   "outputs": [],
   "source": [
    "print('lr_model_1 R2 Score: {}'.format(lr_model_1.score(eval_X, eval_y)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "oKUggTVmuACA",
    "outputId": "23656741-ee91-4b1c-a3f8-cfb93650bc34"
   },
   "outputs": [],
   "source": [
    "print('lr_model_1 MSE: {}'.format(mean_squared_error(eval_y, lr_model_1_preds)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "jqtZ38VhuACE"
   },
   "source": [
    "# Engineer cubic features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "jc3PXN3buACG"
   },
   "outputs": [],
   "source": [
    "steps = [\n",
    "    ('scaler', MinMaxScaler()),\n",
    "    ('poly', PolynomialFeatures(degree=3)),\n",
    "    ('lr', LinearRegression())\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "SNCYPewxuACJ"
   },
   "outputs": [],
   "source": [
    "lr_model_2 = Pipeline(steps)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 177
    },
    "colab_type": "code",
    "id": "0vr1LXSsuACM",
    "outputId": "27888045-5e27-4291-bc87-b96d761a4c20"
   },
   "outputs": [],
   "source": [
    "lr_model_2.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "0T__GpZEuACP",
    "outputId": "cbddcba9-37cd-4115-8675-2ce482e625ae"
   },
   "outputs": [],
   "source": [
    "print('lr_model_2 R2 Score: {}'.format(lr_model_2.score(eval_X, eval_y)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "hkcTjTXQuACT"
   },
   "outputs": [],
   "source": [
    "lr_model_2_preds = lr_model_2.predict(eval_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "ijQLoniluACX",
    "outputId": "7616aaf8-c9d3-4497-8810-ea16985493d8"
   },
   "outputs": [],
   "source": [
    "print('lr_model_2 MSE: {}'.format(mean_squared_error(eval_y, lr_model_2_preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 177
    },
    "colab_type": "code",
    "id": "2oCuH-VOuACa",
    "outputId": "948c8f4a-0909-4e74-f60f-40050f071014"
   },
   "outputs": [],
   "source": [
    "print(lr_model_2[-1].coef_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "4pBIh88SuACf",
    "outputId": "42d39731-1dbc-4385-8c3c-a83cb935e369"
   },
   "outputs": [],
   "source": [
    "print(len(lr_model_2[-1].coef_))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "sRnkygHkuACn"
   },
   "source": [
    "# Engineer polynomial features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Fgc0hLuiuACn"
   },
   "outputs": [],
   "source": [
    "steps = [\n",
    "    ('scaler', MinMaxScaler()),\n",
    "    ('poly', PolynomialFeatures(degree=10)),\n",
    "    ('lr', LinearRegression())\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ENi3637QuACr"
   },
   "outputs": [],
   "source": [
    "lr_model_3 = Pipeline(steps)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "colab_type": "code",
    "id": "6Fa3hJ4duACx",
    "outputId": "9faa5eb7-1ed0-40b0-b99d-c0ba085b4d5f"
   },
   "outputs": [],
   "source": [
    "lr_model_3.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "colab_type": "code",
    "id": "dvVQIDtNuAC1",
    "outputId": "3448cfd7-6ad9-4b58-ccb8-3cfb0e04aea7"
   },
   "outputs": [],
   "source": [
    "print('lr_model_3 R2 Score: {}'.format(lr_model_3.score(eval_X, eval_y)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "g_J5KhfCuAC7"
   },
   "outputs": [],
   "source": [
    "lr_model_3_preds = lr_model_3.predict(eval_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "colab_type": "code",
    "id": "phBeRc0FuADC",
    "outputId": "4de4e58b-e326-4639-d0c0-08a40456ee04"
   },
   "outputs": [],
   "source": [
    "print('lr_model_3 MSE: {}'.format(mean_squared_error(eval_y, lr_model_3_preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "colab_type": "code",
    "id": "Bx8QmaJduADL",
    "outputId": "1832ef7c-f3a3-42f7-aba9-d4a4d640135b"
   },
   "outputs": [],
   "source": [
    "print(len(lr_model_3[-1].coef_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "colab_type": "code",
    "id": "omV4-ydJuADT",
    "outputId": "4ffe3e60-4949-4a96-8ad8-042bb5551ea0"
   },
   "outputs": [],
   "source": [
    "print(lr_model_3[-1].coef_[:35])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "R3KTPZd5uADY"
   },
   "source": [
    "# Implement Lasso on the same pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "wvtRmkOLuADc"
   },
   "outputs": [],
   "source": [
    "steps = [\n",
    "    ('scaler', MinMaxScaler()),\n",
    "    ('poly', PolynomialFeatures(degree=10)),\n",
    "    ('lr', Lasso(alpha=0.01))\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "VKkWpShFuADi"
   },
   "outputs": [],
   "source": [
    "lasso_model = Pipeline(steps)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 212
    },
    "colab_type": "code",
    "id": "CFUKWqGSuAD9",
    "outputId": "304b16fe-a245-4278-8219-4d0f466e21ea"
   },
   "outputs": [],
   "source": [
    "lasso_model.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "_bOVRGGeuAEh",
    "outputId": "6ff11eaa-45ec-4dab-b807-dbdc27d57093"
   },
   "outputs": [],
   "source": [
    "print('lasso_model R2 Score: {}'.format(lasso_model.score(eval_X, eval_y)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "B0sCTyzxuAE1"
   },
   "outputs": [],
   "source": [
    "lasso_preds = lasso_model.predict(eval_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "terkLrI6uAE7",
    "outputId": "5804a278-f495-4a16-a703-2621378d370d"
   },
   "outputs": [],
   "source": [
    "print('lasso_model MSE: {}'.format(mean_squared_error(eval_y, lasso_preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "2y24Ljr_uAFG",
    "outputId": "759e6c34-cee9-4f1e-9791-408aad4e4b20"
   },
   "outputs": [],
   "source": [
    "print(len(lasso_model[-1].coef_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 141
    },
    "colab_type": "code",
    "id": "9fPTFs6CuAFV",
    "outputId": "d15394eb-8ff7-4f36-af46-b61e12f49b97"
   },
   "outputs": [],
   "source": [
    "print(lasso_model[-1].coef_[:35])"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "Exercise7.09.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}