Python机器学习（三十三）Sklearn速查

　　Scikit-learn是一个开源Python库，它使用统一的接口实现了一系列机器学习、预处理、交叉验证和可视化算法。 一个基本例子from sklearn import neighbors, datasets, preprocessing from sklearn.model_selection import train_test_split from sklearn.metrics import accuracy_score iris = datasets.load_iris() X, y = iris.data[:, :2], iris.target X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=33) scaler = preprocessing.StandardScaler().fit(X_train) X_train = scaler.transform(X_train) X_test = scaler.transform(X_test) knn = neighbors.KNeighborsClassifier(n_neighbors=5) knn.fit(X_train, y_train) y_pred = knn.predict(X_test) accuracy_score(y_test, y_pred)加载数据
　　数据类型可以是NumPy数组、SciPy稀疏矩阵，或者其他可转换为数组的类型，如panda DataFrame等。 import numpy as np X = np.random.random((10,5)) y = np.array([＂M＂,＂M＂,＂F＂,＂F＂,＂M＂,＂F＂,＂M＂,＂M＂,＂F＂,＂F＂,＂F＂]) X[X < 0.7] = 0预处理数据标准化/Standardizationfrom sklearn.preprocessing import StandardScaler scaler = StandardScaler().fit(X_train) standardized_X = scaler.transform(X_train) standardized_X_test = scaler.transform(X_test)归一化/Normalizationfrom sklearn.preprocessing import Normalizer scaler = Normalizer().fit(X_train) normalized_X = scaler.transform(X_train) normalized_X_test = scaler.transform(X_test)二值化/Binarizationfrom sklearn.preprocessing import Binarizer binarizer = Binarizer(threshold=0.0).fit(X) binary_X = binarizer.transform(X)类别特征编码from sklearn.preprocessing import LabelEncoder enc = LabelEncoder() y = enc.fit_transform(y)缺失值估算>>>from sklearn.preprocessing import Imputer >>>imp = Imputer(missing_values=0, strategy=＂mean＂, axis=0) >>>imp.fit_transform(X_train)生成多项式特征from sklearn.preprocessing import PolynomialFeatures poly = PolynomialFeatures(5) oly.fit_transform(X)训练与测试数据分组from sklearn.model_selection import train_test_split X_train, X_test, y_train, y_test = train_test_split(X,y,random_state=0)创建模型有监督学习模型线性回归from sklearn.linear_model import LinearRegression lr = LinearRegression(normalize=True)支持向量机(SVM)from sklearn.svm import SVC svc = SVC(kernel=＂linear＂)朴素贝叶斯from sklearn.naive_bayes import GaussianNB gnb = GaussianNB()KNNfrom sklearn.naive_bayes import GaussianNB gnb = GaussianNB()无监督学习模型主成分分析(PCA)from sklearn.decomposition import PCA pca = PCA(n_components=0.95)k均值/K Meansfrom sklearn.cluster import KMeans k_means = KMeans(n_clusters=3, random_state=0)模型拟合有监督学习lr.fit(X, y) knn.fit(X_train, y_train) svc.fit(X_train, y_train)无监督学习k_means.fit(X_train) pca_model = pca.fit_transform(X_train)
　　回到顶部 模型预测有监督学习y_pred = svc.predict(np.random.random((2,5))) y_pred = lr.predict(X_test) y_pred = knn.predict_proba(X_test))无监督学习y_pred = k_means.predict(X_test)评估模型性能分类指标准确度knn.score(X_test, y_test) from sklearn.metrics import accuracy_score accuracy_score(y_test, y_pred)分类报告from sklearn.metrics import classification_report print(classification_report(y_test, y_pred)))混淆矩阵from sklearn.metrics import confusion_matrix print(confusion_matrix(y_test, y_pred)))回归指标平均绝对误差from sklearn.metrics import mean_absolute_error y_true = [3, -0.5, 2]) mean_absolute_error(y_true, y_pred))均方差from sklearn.metrics import mean_squared_error mean_squared_error(y_test, y_pred))R^2Rfrom sklearn.metrics import r2_score r2_score(y_true, y_pred))聚类指标调整兰德系数from sklearn.metrics import adjusted_rand_score adjusted_rand_score(y_true, y_pred))同质性/Homogeneityfrom sklearn.metrics import homogeneity_score homogeneity_score(y_true, y_pred))调和平均指标/V-measurefrom sklearn.metrics import v_measure_score metrics.v_measure_score(y_true, y_pred))交叉验证print(cross_val_score(knn, X_train, y_train, cv=4)) print(cross_val_score(lr, X, y, cv=2))模型调优网格搜索from sklearn.grid_search import GridSearchCV  params = {＂n_neighbors＂: np.arange(1,3), ＂metric＂: [＂euclidean＂, ＂cityblock＂]}  grid = GridSearchCV(estimator=knn,param_grid=params)  grid.fit(X_train, y_train)  print(grid.best_score_)  print(grid.best_estimator_.n_neighbors)随机参数优化from sklearn.grid_search import RandomizedSearchCV  params = {＂n_neighbors＂: range(1,5), ＂weights＂: [＂uniform＂, ＂distance＂]}  rsearch = RandomizedSearchCV(estimator=knn,    param_distributions=params,    cv=4,    n_iter=8,    random_state=5)  rsearch.fit(X_train, y_train)  print(rsearch.best_score_)