XGBoost鸢尾花.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f2a80df3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:18:12.612372Z",
     "start_time": "2023-04-10T02:18:11.666105Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(150, 4)"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import xgboost as xgb\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "# 读入数据集\n",
    "iris = load_iris()\n",
    "X = iris.data\n",
    "y = iris.target\n",
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "faff9333",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:18:12.627570Z",
     "start_time": "2023-04-10T02:18:12.613546Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((105, 4), (105,))"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 划分数据集\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_train.shape,y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e9231ca9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:18:13.227702Z",
     "start_time": "2023-04-10T02:18:12.629569Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', callbacks=None,\n",
       "              colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,\n",
       "              early_stopping_rounds=None, enable_categorical=False,\n",
       "              eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',\n",
       "              importance_type=None, interaction_constraints='',\n",
       "              learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,\n",
       "              max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,\n",
       "              missing=nan, monotone_constraints='()', n_estimators=100,\n",
       "              n_jobs=0, num_parallel_tree=1, objective='multi:softprob',\n",
       "              predictor='auto', random_state=0, reg_alpha=0, ...)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 训练XGBoost模型\n",
    "model = xgb.XGBClassifier()\n",
    "model.fit(X_train, y_train)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27c3f9a9",
   "metadata": {},
   "source": [
    "np.argsort() 是 NumPy 库中的一个函数，用于返回排序后数组中元素的索引，而不是元素本身\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "arr = np.array([3, 1, 4, 2])\n",
    "\n",
    "idx = np.argsort(arr)\n",
    "\n",
    "print(idx)\n",
    "\n",
    "输出为： [1 3 0 2]。\n",
    "\n",
    "这表示原始数组排序后，元素1的索引为1，元素2的索引为3，元素3的索引为0，元素4的索引为2。\n",
    "[::-1] 表示将排序后的数组倒序排列，即按照从大到小的顺序排列"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f6d8250a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:18:13.242955Z",
     "start_time": "2023-04-10T02:18:13.229256Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 3 0 2]\n",
      "[2 0 3 1]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "arr = np.array([3, 1, 4, 2])\n",
    "\n",
    "idx = np.argsort(arr)\n",
    "print(idx)\n",
    "print(idx[::-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2b2cd61f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:18:13.258450Z",
     "start_time": "2023-04-10T02:18:13.244932Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1. feature 2 (0.738762)\n",
      "2. feature 3 (0.220724)\n",
      "3. feature 1 (0.030385)\n",
      "4. feature 0 (0.010128)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 特征重要性排序\n",
    "importance = model.feature_importances_  # array([0.01012843, 0.0303854 , 0.73876214, 0.220724  ], dtype=float32)\n",
    "indices = np.argsort(importance)[::-1]  # array([2, 3, 1, 0], dtype=int64)  [::-1]表示倒叙\n",
    "\n",
    "# 打印每个特征的重要性得分\n",
    "for f in range(X.shape[1]):\n",
    "    print(\"%d. feature %d (%f)\" % (f + 1, indices[f], importance[indices[f]]))\n",
    "# X.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "10a29b9e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:18:13.935936Z",
     "start_time": "2023-04-10T02:18:13.260687Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 取前两个重要特征\n",
    "X_selected = X[:, indices[:2]]\n",
    "\n",
    "# 训练XGBoost模型并预测\n",
    "model.fit(X_selected, y)\n",
    "y_pred = model.predict(X_selected)\n",
    "\n",
    "# 可视化\n",
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.scatter(X_selected[:, 0], X_selected[:, 1], c=y_pred, cmap='viridis')\n",
    "plt.xlabel('Feature 3 (Petal length)')\n",
    "plt.ylabel('Feature 2 (Petal width)')\n",
    "plt.title('XGBoost Feature Selection and Dimension Reduction', fontsize=16)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3edaaa26",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:20:14.333535Z",
     "start_time": "2023-04-10T02:20:14.324627Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 3], dtype=int64)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# indices\n",
    "indices[:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ec5369cb",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:18:13.966881Z",
     "start_time": "2023-04-10T02:18:13.952882Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.01012843, 0.0303854 , 0.73876214, 0.220724  ], dtype=float32)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ace2026d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T02:30:01.748444Z",
     "start_time": "2023-04-10T02:30:01.732198Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 3, 1, 0], dtype=int64)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.argsort(importance)[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "34e9771c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T05:53:27.437036Z",
     "start_time": "2023-04-10T05:53:27.418983Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True, False,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred == y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "e4426df6",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-04-10T05:53:30.306813Z",
     "start_time": "2023-04-10T05:53:29.380515Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "E:\\anaconda\\envs\\pytorch\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1037: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  \"KMeans is known to have a memory leak on Windows \"\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "sse = []\n",
    "k_range = range(1, 11)\n",
    "for k in k_range:\n",
    "    kmeans = KMeans(n_clusters=k)\n",
    "    kmeans.fit(X)\n",
    "    sse.append(kmeans.inertia_)\n",
    "\n",
    "plt.plot(k_range, sse)\n",
    "plt.xlabel('Number of clusters')\n",
    "plt.ylabel('SSE')\n",
    "plt.title('Elbow Method for Optimal k')\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pytorch",
   "language": "python",
   "name": "pytorch"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.15"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}