mlessentials/Lab06/Activity6.01/Activity6_01.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "uGkVk0AaLY6l"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "dRMP8E1FLpWI"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('../Dataset/bank-additional-full.csv', sep=';')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 459
    },
    "colab_type": "code",
    "id": "r7XYKYARLqt4",
    "outputId": "e66dec91-0d8e-45db-fa6f-008de8cdb3a8"
   },
   "outputs": [],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 224
    },
    "colab_type": "code",
    "id": "MtOHZhE7LuIu",
    "outputId": "892b49b5-9a15-44e3-b0c0-51ccb5760383"
   },
   "outputs": [],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "kA7IlrPiMOGF",
    "outputId": "aed6bf89-aa0f-417a-da20-ad7b21a15923"
   },
   "outputs": [],
   "source": [
    "cat_cols = ['job', 'marital', 'education', 'default', 'housing', 'loan', 'contact', 'month', 'day_of_week', 'poutcome']\n",
    "_df = pd.get_dummies(df, columns=cat_cols, prefix=cat_cols, drop_first=True)\n",
    "_df.info()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 241
    },
    "colab_type": "code",
    "id": "sEJc8NZeMTZm",
    "outputId": "2f6b481c-ef34-445c-effd-379ecfc21515"
   },
   "outputs": [],
   "source": [
    "_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "q8WPkabYMVXy"
   },
   "outputs": [],
   "source": [
    "X = _df.drop(['y'], axis=1)\n",
    "X = X.values\n",
    "y = df['y'].apply(lambda x: 0 if x == 'no' else 1)\n",
    "y = y.values\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "9PX4ava1MXiu"
   },
   "outputs": [],
   "source": [
    "train_X, eval_X, train_y, eval_y = train_test_split(X, y, test_size=0.3, random_state=0)\n",
    "val_X, test_X, val_y, test_y = train_test_split(eval_X, eval_y, random_state=0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "lqdsD7YzMZ-2"
   },
   "outputs": [],
   "source": [
    "lr_model = LogisticRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 258
    },
    "colab_type": "code",
    "id": "-rdzBySdMcH2",
    "outputId": "9c4611af-6eeb-4822-cc08-70da5cc69684"
   },
   "outputs": [],
   "source": [
    "lr_model.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "YKNwsA8FMdw2"
   },
   "outputs": [],
   "source": [
    "lr_preds = lr_model.predict(val_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 170
    },
    "colab_type": "code",
    "id": "7L_lyv5uMgkU",
    "outputId": "ef706117-e639-4b05-9c94-e77b06cb72f3"
   },
   "outputs": [],
   "source": [
    "lr_report = classification_report(val_y, lr_preds)\n",
    "print(lr_report)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "EusQF-6MMidI"
   },
   "outputs": [],
   "source": [
    "dt_model = DecisionTreeClassifier(max_depth= 6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 119
    },
    "colab_type": "code",
    "id": "HvYPMNDGMkIf",
    "outputId": "96850069-5b13-45b7-f772-80c734f1120b"
   },
   "outputs": [],
   "source": [
    "dt_model.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "aSBt8RvDMm_m"
   },
   "outputs": [],
   "source": [
    "dt_preds = dt_model.predict(val_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 170
    },
    "colab_type": "code",
    "id": "9CNLWLklMo5_",
    "outputId": "95017766-faae-4c17-81a0-0398adb19027"
   },
   "outputs": [],
   "source": [
    "dt_report = classification_report(val_y, dt_preds)\n",
    "print(dt_report)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 153
    },
    "colab_type": "code",
    "id": "Q60nt2b0Mq0X",
    "outputId": "e6408ad0-722c-4d53-a801-75f985784c9b"
   },
   "outputs": [],
   "source": [
    "rf_model = RandomForestClassifier(n_estimators=1000)\n",
    "rf_model.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "T37wZC1lMy7-"
   },
   "outputs": [],
   "source": [
    "rf_preds = rf_model.predict(val_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 170
    },
    "colab_type": "code",
    "id": "rfyAqubjNF5h",
    "outputId": "3454b27f-91be-4157-8e7b-9dfd20a77137"
   },
   "outputs": [],
   "source": [
    "rf_report = classification_report(val_y, rf_preds)\n",
    "print(rf_report)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "HkSlaECTNH1X",
    "outputId": "d2a068a4-25b1-4767-b782-df1133d03cba"
   },
   "outputs": [],
   "source": [
    "print('Linear Score: {}, DecisionTree Score: {}, RandomForest Score: {}'.format(lr_model.score(val_X, val_y), dt_model.score(val_X, val_y), rf_model.score(val_X, val_y)))"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "Activity6.01.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
}