错误：“管道的最后一步应实现fit或字符串‘passthrough’

Question

我已经创建了一个管道，该管道应该适合我的数据和评分本身的不同模型，但我收到了这个错误

代码如下：

    `models = {"Ridge": Ridge(), 
         "SVR_linear": SVR(kernel="linear"),
         "SVR_rbf": SVR(kernel="rbf"),
         "RandomForestRegressor": RandomForestRegressor()}
reg_results = {}`

`for model_name, model in models.items(): 
    
    # Model preprocessor with pipeline
    model_pipeline = Pipeline(steps=[("preprocessor", preprocessor), 
                                     ("model", models)])
    
    # Fit models
    print(f"Fitting {model_name}...")
    model_pipeline.fit(car_X_train, car_y_train)
    
    # Score model and append model results dict with values
    print(f"Scoring {model_name}...")
    reg_results[model_name] = model.pipeline.score(car_X_test, car_y_test)`

Answer 1

model_pipeline = Pipeline(steps=[("preprocessor", preprocessor), 
                                     ("model", model)])

将“模型”替换为“模型”

• 因为“模型”是你的模型字典
• 您应该在这个管道步骤中使用字典值（来自for循环的“model”）