# howork_scikit_learn.py# 1. 数据准备importpandasaspdfromsklearn.model_selectionimporttrain_test_split datapd.read_csv(dataset.csv)Xdata.drop(target,axis1)# 特征ydata[target]# 标签# 2. 划分训练集和测试集# X_train, X_test, y_train, y_test train_test_split(# X, y, test_size0.2, random_state42# )y_binary(yy.mean()).astype(int)# 根据均值分割X_train,X_test,y_train,y_testtrain_test_split(X,y_binary,test_size0.2,random_state42)# 3. 特征工程标准化fromsklearn.preprocessingimportStandardScaler scalerStandardScaler()X_train_scaledscaler.fit_transform(X_train)X_test_scaledscaler.transform(X_test)# 4. 选择模型并训练fromsklearn.ensembleimportRandomForestClassifier modelRandomForestClassifier(n_estimators100,random_state42)model.fit(X_train_scaled,y_train)# 5. 预测与评估fromsklearn.metricsimportaccuracy_score predictionsmodel.predict(X_test_scaled)accuracyaccuracy_score(y_test,predictions)print(f模型准确率:{accuracy:.2f})# 6. 模型保存importjoblib joblib.dump(model,trained_model.pkl)运行结果(ai_env)$ python3 howork_scikit_learn.py 模型准确率: 0.91(ai_env)$
自适应惯量控制技术解析)
