3_3_XGBoost.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* XGBoost 버전 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost\n",
    "\n",
    "print(xgboost.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 파이썬 래퍼 XGBoost 적용 – 위스콘신 유방암 예측"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost as xgb\n",
    "from xgboost import plot_importance\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.model_selection import train_test_split\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "dataset = load_breast_cancer()\n",
    "X_features= dataset.data\n",
    "y_label = dataset.target\n",
    "\n",
    "cancer_df = pd.DataFrame(data=X_features, columns=dataset.feature_names)\n",
    "cancer_df['target']= y_label\n",
    "cancer_df.head(3)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(dataset.target_names)\n",
    "print(cancer_df['target'].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# cancer_df에서 feature용 DataFrame과 Label용 Series 객체 추출\n",
    "# 맨 마지막 칼럼이 Label임. Feature용 DataFrame은 cancer_df의 첫번째 칼럼에서 맨 마지막 두번째 칼럼까지를 :-1 슬라이싱으로 추출.\n",
    "X_features = cancer_df.iloc[:, :-1]\n",
    "y_label = cancer_df.iloc[:, -1]\n",
    "\n",
    "# 전체 데이터 중 80%는 학습용 데이터, 20%는 테스트용 데이터 추출\n",
    "X_train, X_test, y_train, y_test=train_test_split(X_features, y_label,\n",
    "                                         test_size=0.2, random_state=156 )\n",
    "\n",
    "# 위에서 만든 X_train, y_train을 다시 쪼개서 90%는 학습과 10%는 검증용 데이터로 분리\n",
    "X_tr, X_val, y_tr, y_val= train_test_split(X_train, y_train, test_size=0.1, random_state=156 )\n",
    "print(X_train.shape , X_test.shape)\n",
    "print(X_tr.shape, X_val.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 만약 구버전 XGBoost에서 DataFrame으로 DMatrix 생성이 안될 경우 X_train.values로 넘파이 변환.\n",
    "# 학습, 검증, 테스트용 DMatrix를 생성.\n",
    "dtr = xgb.DMatrix(data=X_tr, label=y_tr)\n",
    "dval = xgb.DMatrix(data=X_val, label=y_val)\n",
    "dtest = xgb.DMatrix(data=X_test , label=y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "params = { 'max_depth':3,\n",
    "          'eta': 0.05,\n",
    "          'objective':'binary:logistic',\n",
    "          'eval_metric':'logloss'\n",
    "         }\n",
    "num_rounds = 400"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 학습 데이터 셋은 'train' 또는 평가 데이터 셋은 'eval' 로 명기합니다. \n",
    "eval_list = [(dtr,'train'),(dval,'eval')] # 또는 eval_list = [(dval,'eval')] 만 명기해도 무방. \n",
    "\n",
    "# 하이퍼 파라미터와 early stopping 파라미터를 train( ) 함수의 파라미터로 전달\n",
    "xgb_model = xgb.train(params = params , dtrain=dtr , num_boost_round=num_rounds , \\\n",
    "                      early_stopping_rounds=50, evals=eval_list )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_probs = xgb_model.predict(dtest)\n",
    "print('predict( ) 수행 결과값을 10개만 표시, 예측 확률 값으로 표시됨')\n",
    "print(np.round(pred_probs[:10],3))\n",
    "\n",
    "# 예측 확률이 0.5 보다 크면 1 , 그렇지 않으면 0 으로 예측값 결정하여 List 객체인 preds에 저장 \n",
    "preds = [ 1 if x > 0.5 else 0 for x in pred_probs ]\n",
    "print('예측값 10개만 표시:',preds[:10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix, accuracy_score\n",
    "from sklearn.metrics import precision_score, recall_score\n",
    "from sklearn.metrics import f1_score, roc_auc_score\n",
    "\n",
    "def get_clf_eval(y_test, pred=None, pred_proba=None):\n",
    "    confusion = confusion_matrix( y_test, pred)\n",
    "    accuracy = accuracy_score(y_test , pred)\n",
    "    precision = precision_score(y_test , pred)\n",
    "    recall = recall_score(y_test , pred)\n",
    "    f1 = f1_score(y_test,pred)\n",
    "    # ROC-AUC 추가 \n",
    "    roc_auc = roc_auc_score(y_test, pred_proba)\n",
    "    print('오차 행렬')\n",
    "    print(confusion)\n",
    "    # ROC-AUC print 추가\n",
    "    print('정확도: {0:.4f}, 정밀도: {1:.4f}, 재현율: {2:.4f},\\\n",
    "    F1: {3:.4f}, AUC:{4:.4f}'.format(accuracy, precision, recall, f1, roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "get_clf_eval(y_test , preds, pred_probs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10, 12))\n",
    "plot_importance(xgb_model, ax=ax)\n",
    "plt.savefig('p239_xgb_feature_importance.tif', format='tif', dpi=300, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 사이킷런 래퍼 XGBoost의 개요 및 적용"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 사이킷런 래퍼 XGBoost 클래스인 XGBClassifier 임포트\n",
    "from xgboost import XGBClassifier\n",
    "\n",
    "xgb_wrapper = XGBClassifier(n_estimators=400, learning_rate=0.1, max_depth=3)\n",
    "xgb_wrapper.fit(X_train, y_train)\n",
    "w_preds = xgb_wrapper.predict(X_test)\n",
    "w_pred_proba = xgb_wrapper.predict_proba(X_test)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "get_clf_eval(y_test , w_preds, w_pred_proba)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from xgboost import XGBClassifier\n",
    "\n",
    "xgb_wrapper = XGBClassifier(n_estimators=400, learning_rate=0.1, max_depth=3)\n",
    "evals = [(X_test, y_test)]\n",
    "xgb_wrapper.fit(X_train, y_train, early_stopping_rounds=100, eval_metric=\"logloss\", \n",
    "                eval_set=evals, verbose=True)\n",
    "\n",
    "ws100_preds = xgb_wrapper.predict(X_test)\n",
    "ws100_pred_proba = xgb_wrapper.predict_proba(X_test)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "get_clf_eval(y_test , ws100_preds, ws100_pred_proba)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# early_stopping_rounds를 10으로 설정하고 재 학습. \n",
    "xgb_wrapper.fit(X_train, y_train, early_stopping_rounds=10, \n",
    "                eval_metric=\"logloss\", eval_set=evals,verbose=True)\n",
    "\n",
    "ws10_preds = xgb_wrapper.predict(X_test)\n",
    "ws10_pred_proba = xgb_wrapper.predict_proba(X_test)[:, 1]\n",
    "get_clf_eval(y_test , ws10_preds, ws10_pred_proba)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from xgboost import plot_importance\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10, 12))\n",
    "# 사이킷런 래퍼 클래스를 입력해도 무방. \n",
    "plot_importance(xgb_wrapper, ax=ax)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}