2023/machine_learning/sklearn_pipeline.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "hide_input": false,
    "tags": []
   },
   "outputs": [],
   "source": [
    "from preamble import *\n",
    "%matplotlib inline\n",
    "import matplotlib as mpl\n",
    "mpl.rcParams['figure.dpi'] = 80\n",
    "mpl.rcParams['legend.numpoints'] = 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 整个机器学习流程大概是这个样子"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.972027972027972"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "# The breast cancer dataset is a classic and very easy binary classification\n",
    "# dataset.\n",
    "\n",
    "# =================   ==============\n",
    "# Classes                          2\n",
    "# Samples per class    212(M),357(B)\n",
    "# Samples total                  569\n",
    "# Dimensionality                  30\n",
    "# Features            real, positive\n",
    "# =================   ==============\n",
    "\n",
    "# 切分训练集和测试集\n",
    "cancer = load_breast_cancer()\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    cancer.data, cancer.target, random_state=0)\n",
    "\n",
    "# 用scaler做特征预处理\n",
    "scaler = MinMaxScaler().fit(X_train)\n",
    "X_train_scaled = scaler.transform(X_train)\n",
    "\n",
    "svm = SVC()\n",
    "# 应用SVM进行建模\n",
    "svm.fit(X_train_scaled, y_train)\n",
    "# 对test数据做预处理，并使用模型预测打分\n",
    "X_test_scaled = scaler.transform(X_test)\n",
    "svm.score(X_test_scaled, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 回顾网格搜索"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best cross-validation accuracy: 0.9812311901504789\n",
      "test set score:  0.972027972027972\n",
      "best parameters:  {'C': 1, 'gamma': 1}\n",
      "CPU times: user 1.34 s, sys: 542 µs, total: 1.34 s\n",
      "Wall time: 1.34 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "param_grid = {'C': [0.001, 0.01, 0.1, 1, 10, 100],\n",
    "              'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}\n",
    "grid = GridSearchCV(SVC(), param_grid=param_grid, cv=5)\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "print(\"best cross-validation accuracy:\", grid.best_score_)\n",
    "print(\"test set score: \", grid.score(X_test_scaled, y_test))\n",
    "print(\"best parameters: \", grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "hide_input": false,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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",
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_improper_processing()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 搭建流程/Pipeline！！\n",
    "\n",
    "\n",
    "- 使用 `Pipeline` 类来创建一个包含数据预处理和模型训练两个步骤的机器学习流程。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.pipeline import Pipeline\n",
    "pipe = Pipeline([(\"scaler\", MinMaxScaler()), (\"svm\", SVC())])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;scaler&#x27;, MinMaxScaler()), (&#x27;svm&#x27;, SVC())])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;scaler&#x27;, MinMaxScaler()), (&#x27;svm&#x27;, SVC())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">MinMaxScaler</label><div class=\"sk-toggleable__content\"><pre>MinMaxScaler()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SVC</label><div class=\"sk-toggleable__content\"><pre>SVC()</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('scaler', MinMaxScaler()), ('svm', SVC())])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.972027972027972"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Pipeline的优点是可以将多个步骤组合成一个整体，使得代码更加简洁，同时也可以防止数据泄露，因为它确保了数据预处理和模型训练始终在同一数据集上进行。\n",
    "- 此外，Pipeline还可以与GridSearchCV等工具一起使用，进行超参数搜索。\n",
    "- Pipeline通常在我们需要进行多个连续的步骤，例如数据预处理、特征选择、模型训练等，时使用。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 串起来流程和网格搜索"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [],
   "source": [
    "param_grid = {'svm__C': [0.001, 0.01, 0.1, 1, 10, 100],\n",
    "              'svm__gamma': [0.001, 0.01, 0.1, 1, 10, 100]}\n",
    "\n",
    "# 这里非常重要！一定要写明是名字为”svm“的超参数，中间的两个下划线是固定的写法"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best cross-validation accuracy: 0.9812311901504789\n",
      "test set score:  0.972027972027972\n",
      "best parameters:  {'svm__C': 1, 'svm__gamma': 1}\n"
     ]
    }
   ],
   "source": [
    "grid = GridSearchCV(pipe, param_grid=param_grid, cv=5)\n",
    "grid.fit(X_train, y_train)\n",
    "print(\"best cross-validation accuracy:\", grid.best_score_)\n",
    "print(\"test set score: \", grid.score(X_test, y_test))\n",
    "print(\"best parameters: \", grid.best_params_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 将Pipeline和GridSearchCV结合使用的优点是可以将数据预处理、模型训练和超参数搜索组合在一起，使得代码更加简洁，同时也可以防止数据泄露。\n",
    "- 此外，这种方法还可以处理更复杂的情况，例如当我们需要对流程中的多个步骤进行超参数搜索时。\n",
    "\n",
    "- 这种方法通常在我们需要同时进行数据预处理、模型训练和超参数搜索时使用，例如在比较不同的模型或者不同的超参数设置时。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "hide_input": false,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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",
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x640 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_proper_processing()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 特征选择也可以加到 pipeline 中\n",
    "\n",
    "特征选择是一种常见的数据预处理步骤，可以帮助我们减少数据的维度，提高模型的训练速度，同时也可能提高模型的性能。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "rnd = np.random.RandomState(seed=0)\n",
    "X = rnd.normal(size=(100, 10000))  # 1w个特征！\n",
    "y = rnd.normal(size=(100,))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(100, 500)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.feature_selection import SelectPercentile, f_regression\n",
    "\n",
    "# 使用SelectPercentile类进行特征选择。这个类会选择与目标变量最相关的前5%的特征。\n",
    "# 在这个例子中，我们使用了f_regression作为评分函数，它可以计算每个特征和目标变量的F值。\n",
    "select = SelectPercentile(score_func=f_regression, percentile=5).fit(X, y)\n",
    "X_selected = select.transform(X)\n",
    "print(X_selected.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9057953065239822"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.linear_model import Ridge\n",
    "# 使用Ridge类训练一个岭回归模型 （L2），并使用交叉验证评估其性能。\n",
    "\n",
    "np.mean(cross_val_score(Ridge(), X_selected, y, cv=5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.24655422384952824"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 使用Pipeline类创建一个包含特征选择和模型训练两个步骤的流程。\n",
    "\n",
    "pipe = Pipeline([(\"select\", SelectPercentile(score_func=f_regression, percentile=5)), (\"ridge\", Ridge())])\n",
    "np.mean(cross_val_score(pipe, X, y, cv=5))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 将特征选择加入到流程中的优点是可以将数据预处理和模型训练组合在一起，使得代码更加简洁，同时也可以防止数据泄露。\n",
    "- 此外，这种方法还可以处理更复杂的情况，例如当我们需要对流程中的多个步骤进行超参数搜索时。\n",
    "- 这种方法通常在我们需要进行特征选择时使用，例如当数据的维度很高，或者我们希望减少模型的复杂度时。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 使用更简单的函数 `make_pipeline`\n",
    "\n",
    "- make_pipeline函数是Pipeline类的一个简化版本，它会自动根据每个步骤的类名生成步骤的名称。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "# standard syntax\n",
    "pipe_long = Pipeline([(\"scaler\", MinMaxScaler()), (\"svm\", SVC(C=100))])\n",
    "# abbreviated syntax\n",
    "pipe_short = make_pipeline(MinMaxScaler(), SVC(C=100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('minmaxscaler', MinMaxScaler()), ('svc', SVC(C=100))]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 使用steps属性查看流程的步骤。\n",
    "# 这个属性返回一个列表，每个元素是一个元组，包含步骤的名称和对象。\n",
    "pipe_short.steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('standardscaler-1', StandardScaler()),\n",
       " ('pca', PCA(n_components=2)),\n",
       " ('standardscaler-2', StandardScaler())]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "pipe = make_pipeline(StandardScaler(), PCA(n_components=2), StandardScaler())\n",
    "pipe.steps"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 想要中间步骤的结果怎么办：取出来！\n",
    "\n",
    "- 使用named_steps属性访问流程中的步骤。\n",
    "- 这个属性返回一个字典，键是步骤的名称，值是步骤的对象。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2, 30)\n"
     ]
    }
   ],
   "source": [
    "# fit the pipeline defined above to the cancer dataset\n",
    "pipe.fit(cancer.data)\n",
    "# extract the first two principal components from the \"pca\" step\n",
    "# pca步骤的components_属性，这个属性包含了主成分分析的结果。\n",
    "components = pipe.named_steps[\"pca\"].components_\n",
    "print(components.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 想要网格搜索的中间结果怎么办：取！"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "pipe = make_pipeline(StandardScaler(), LogisticRegression(max_iter=1000))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "param_grid = {'logisticregression__C': [0.01, 0.1, 1, 10, 100]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[(&#x27;standardscaler&#x27;, StandardScaler()),\n",
       "                                       (&#x27;logisticregression&#x27;,\n",
       "                                        LogisticRegression(max_iter=1000))]),\n",
       "             param_grid={&#x27;logisticregression__C&#x27;: [0.01, 0.1, 1, 10, 100]})</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[(&#x27;standardscaler&#x27;, StandardScaler()),\n",
       "                                       (&#x27;logisticregression&#x27;,\n",
       "                                        LogisticRegression(max_iter=1000))]),\n",
       "             param_grid={&#x27;logisticregression__C&#x27;: [0.01, 0.1, 1, 10, 100]})</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;standardscaler&#x27;, StandardScaler()),\n",
       "                (&#x27;logisticregression&#x27;, LogisticRegression(max_iter=1000))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" ><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(max_iter=1000)</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[('standardscaler', StandardScaler()),\n",
       "                                       ('logisticregression',\n",
       "                                        LogisticRegression(max_iter=1000))]),\n",
       "             param_grid={'logisticregression__C': [0.01, 0.1, 1, 10, 100]})"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    cancer.data, cancer.target, random_state=4)\n",
    "grid = GridSearchCV(pipe, param_grid, cv=5)\n",
    "grid.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 使用best_estimator_属性访问网格搜索的结果。\n",
    "- 这个属性返回一个流程，这个流程使用了最优的超参数。\n",
    "- 我们可以通过这个流程访问任何步骤的属性，例如在这个例子中，我们访问了logisticregression步骤的coef_属性，这个属性包含了逻辑回归模型的系数。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pipeline(steps=[('standardscaler', StandardScaler()),\n",
      "                ('logisticregression', LogisticRegression(C=1, max_iter=1000))])\n"
     ]
    }
   ],
   "source": [
    "print(grid.best_estimator_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression(C=1, max_iter=1000)\n"
     ]
    }
   ],
   "source": [
    "print(grid.best_estimator_.named_steps[\"logisticregression\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-0.436 -0.343 -0.408 -0.534 -0.15   0.61  -0.726 -0.785  0.039  0.275\n",
      "  -1.298  0.049 -0.673 -0.934 -0.139  0.45  -0.13  -0.101  0.434  0.716\n",
      "  -1.091 -1.095 -0.852 -1.064 -0.743  0.073 -0.823 -0.653 -0.644 -0.42 ]]\n"
     ]
    }
   ],
   "source": [
    "print(grid.best_estimator_.named_steps[\"logisticregression\"].coef_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 把各种预处理、参数调整、交叉验证放到一个流程里！"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用make_pipeline函数创建一个流程。这个流程包含三个步骤：\n",
    "- 首先，使用StandardScaler进行数据预处理，将数据标准化；\n",
    "- 然后，使用PolynomialFeatures生成多项式特征；\n",
    "- 最后，使用Ridge（岭回归）进行模型训练。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_url = \"http://lib.stat.cmu.edu/datasets/boston\"\n",
    "raw_df = pd.read_csv(data_url, sep=\"\\s+\", skiprows=22, header=None)\n",
    "data = np.hstack([raw_df.values[::2, :], raw_df.values[1::2, :2]])\n",
    "target = raw_df.values[1::2, 2]\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(data, target, random_state=0)\n",
    "\n",
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "pipe = make_pipeline(\n",
    "    StandardScaler(),\n",
    "    PolynomialFeatures(),\n",
    "    Ridge()) # L2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [],
   "source": [
    "param_grid = {'polynomialfeatures__degree': [1, 2, 3],\n",
    "              'ridge__alpha': [0.001, 0.01, 0.1, 1, 10, 100]}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用GridSearchCV类创建一个网格搜索对象。这个对象会在给定的超参数网格上进行搜索，找到最优的超参数。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {color: black;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[(&#x27;standardscaler&#x27;, StandardScaler()),\n",
       "                                       (&#x27;polynomialfeatures&#x27;,\n",
       "                                        PolynomialFeatures()),\n",
       "                                       (&#x27;ridge&#x27;, Ridge())]),\n",
       "             n_jobs=-1,\n",
       "             param_grid={&#x27;polynomialfeatures__degree&#x27;: [1, 2, 3],\n",
       "                         &#x27;ridge__alpha&#x27;: [0.001, 0.01, 0.1, 1, 10, 100]})</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" ><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[(&#x27;standardscaler&#x27;, StandardScaler()),\n",
       "                                       (&#x27;polynomialfeatures&#x27;,\n",
       "                                        PolynomialFeatures()),\n",
       "                                       (&#x27;ridge&#x27;, Ridge())]),\n",
       "             n_jobs=-1,\n",
       "             param_grid={&#x27;polynomialfeatures__degree&#x27;: [1, 2, 3],\n",
       "                         &#x27;ridge__alpha&#x27;: [0.001, 0.01, 0.1, 1, 10, 100]})</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-9\" type=\"checkbox\" ><label for=\"sk-estimator-id-9\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;standardscaler&#x27;, StandardScaler()),\n",
       "                (&#x27;polynomialfeatures&#x27;, PolynomialFeatures()),\n",
       "                (&#x27;ridge&#x27;, Ridge())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-10\" type=\"checkbox\" ><label for=\"sk-estimator-id-10\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-11\" type=\"checkbox\" ><label for=\"sk-estimator-id-11\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">PolynomialFeatures</label><div class=\"sk-toggleable__content\"><pre>PolynomialFeatures()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-12\" type=\"checkbox\" ><label for=\"sk-estimator-id-12\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Ridge</label><div class=\"sk-toggleable__content\"><pre>Ridge()</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[('standardscaler', StandardScaler()),\n",
       "                                       ('polynomialfeatures',\n",
       "                                        PolynomialFeatures()),\n",
       "                                       ('ridge', Ridge())]),\n",
       "             n_jobs=-1,\n",
       "             param_grid={'polynomialfeatures__degree': [1, 2, 3],\n",
       "                         'ridge__alpha': [0.001, 0.01, 0.1, 1, 10, 100]})"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid = GridSearchCV(pipe, param_grid=param_grid, cv=5, n_jobs=-1)\n",
    "grid.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7fd758c5ea10>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/pdf": "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",
      "image/png": "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",
      "text/plain": [
       "<Figure size 768x384 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 这个图显示了不同的超参数组合对应的交叉验证性能。\n",
    "plt.matshow(np.array([s for s in grid.cv_results_['mean_test_score']]).reshape(3, -1),\n",
    "            vmin=0, cmap=\"viridis\")\n",
    "plt.xlabel(\"ridge__alpha\")\n",
    "plt.ylabel(\"polynomialfeatures__degree\")\n",
    "plt.xticks(range(len(param_grid['ridge__alpha'])), param_grid['ridge__alpha'])\n",
    "plt.yticks(range(len(param_grid['polynomialfeatures__degree'])), param_grid['polynomialfeatures__degree'])\n",
    "\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'polynomialfeatures__degree': 2, 'ridge__alpha': 10}\n"
     ]
    }
   ],
   "source": [
    "print(grid.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7683045464100124"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6272671927504456"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 对比一下不做多项式特征生成的结果\n",
    "param_grid = {'ridge__alpha': [0.001, 0.01, 0.1, 1, 10, 100]}\n",
    "pipe = make_pipeline(StandardScaler(), Ridge())\n",
    "grid = GridSearchCV(pipe, param_grid, cv=5)\n",
    "grid.fit(X_train, y_train)\n",
    "grid.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 将Pipeline和GridSearchCV结合使用的优点是可以将数据预处理、模型训练和超参数搜索组合在一起，使得代码更加简洁，同时也可以防止数据泄露。\n",
    "- 此外，这种方法还可以处理更复杂的情况，例如当我们需要对流程中的多个步骤进行超参数搜索时。\n",
    "- 这种方法通常在我们需要同时进行数据预处理、模型训练和超参数搜索时使用，例如在比较不同的模型或者不同的超参数设置时。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ICTP-AP",
   "language": "python",
   "name": "ictp-ap"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}