mlessentials/Lab09/Activity9.01/Activity9_1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "6lf7dlK6Ykw8"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score\n",
    "import altair as alt\n",
    "from sklearn.inspection import plot_partial_dependence\n",
    "from mlxtend.evaluate import feature_importance_permutation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_5m6hmonYhPH"
   },
   "outputs": [],
   "source": [
    "file_url = '../Dataset/KDDCup99.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "vdt3sXysYrRj"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(file_url)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 224
    },
    "colab_type": "code",
    "id": "TanrIa0OY5CL",
    "outputId": "485b874f-83bb-472a-8578-ed931d592740"
   },
   "outputs": [],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "MMGyKdDxZDmG"
   },
   "outputs": [],
   "source": [
    "y = df.pop('label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "YR3muhtlMpSN"
   },
   "outputs": [],
   "source": [
    "df = pd.get_dummies(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "FTQJZEYfZDrt"
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(df, y, test_size=0.3, random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 170
    },
    "colab_type": "code",
    "id": "R3FoM7q_YqDS",
    "outputId": "8ea7c58f-cfdf-4d6b-9b77-d1c3b7c86632"
   },
   "outputs": [],
   "source": [
    "rf_model = RandomForestClassifier(random_state=168)\n",
    "rf_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "7OrrC55GNKyH"
   },
   "outputs": [],
   "source": [
    "train_preds = rf_model.predict(X_train)\n",
    "test_preds = rf_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "id": "kf31T5toNK_X",
    "outputId": "3243de5a-304e-4533-da6f-c27dc929bf81"
   },
   "outputs": [],
   "source": [
    "train_acc = accuracy_score(y_train, train_preds)\n",
    "test_acc = accuracy_score(y_test, test_preds)\n",
    "print(train_acc)\n",
    "print(test_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 527
    },
    "colab_type": "code",
    "id": "m0PbG5G1C4jl",
    "outputId": "01e44422-7535-4e09-e520-6055b4c813b2"
   },
   "outputs": [],
   "source": [
    "imp_vals, _ = feature_importance_permutation(predict_method=rf_model.predict, X=X_train.values, y=y_train.values, metric='accuracy', num_rounds=1, seed=2)\n",
    "imp_vals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QU8f-YtcfPDv"
   },
   "outputs": [],
   "source": [
    "perm_varimp_df = pd.DataFrame({'feature': X_train.columns, 'importance': imp_vals})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "id": "dxT-DTETDDJM",
    "outputId": "f6cfc9b8-f854-4808-9a7c-788a31a12a0e"
   },
   "outputs": [],
   "source": [
    "perm_varimp_df.sort_values('importance', ascending=False, inplace=True)\n",
    "perm_varimp_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 463
    },
    "colab_type": "code",
    "id": "hOvXZ892DJnG",
    "outputId": "e7a12985-07f0-45a5-ad63-9dd9a65b591e",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "alt.Chart(perm_varimp_df[:20]).mark_bar().encode(\n",
    "    x='importance',\n",
    "    y=alt.Y('feature:N', sort=alt.SortField(field='importance', order='descending'))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "colab_type": "code",
    "id": "REYq4IgnX_3z",
    "outputId": "6e472dbb-315e-45ad-cd0c-ee3ef4b4aefd"
   },
   "outputs": [],
   "source": [
    "feature_index = df.columns.get_loc(\"src_bytes\")\n",
    "plot_partial_dependence(rf_model, X_train, features=[feature_index], feature_names=X_train.columns,  target=\"normal\", response_method=\"predict_proba\", n_jobs=-1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 493
    },
    "colab_type": "code",
    "id": "teXyiRIIu4ru",
    "outputId": "b39c0aef-22fa-4929-e2d5-e6ba28681beb"
   },
   "outputs": [],
   "source": [
    "!pip install lime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "8j-mE2RcEcEe"
   },
   "outputs": [],
   "source": [
    "from lime.lime_tabular import LimeTabularExplainer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 408
    },
    "colab_type": "code",
    "id": "8UP9bJlLggn5",
    "outputId": "ff22695c-19b6-4747-b2cd-431e0308a581"
   },
   "outputs": [],
   "source": [
    "class_names = sorted(y.unique())\n",
    "class_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "K4F-4QHsvuuN"
   },
   "outputs": [],
   "source": [
    "lime_explainer = LimeTabularExplainer(X_train.values,\n",
    "      feature_names=X_train.columns,\n",
    "      class_names=class_names,\n",
    "      mode='classification')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "nDFatUx0N_3x",
    "outputId": "7771ebc5-0086-499e-ab02-99295c7e85fb"
   },
   "outputs": [],
   "source": [
    "exp = lime_explainer.explain_instance(X_test.iloc[99893,], rf_model.predict_proba, num_features=50, top_labels=1)\n",
    "exp.show_in_notebook()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Activity9_1",
   "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
}