Why am I receiving an error for return_train_score when trying to cross validate-CodePudding

I am trying to execute a for loop for the following dictionary, but I am receiving an error that states:

"cross_val_score() got an unexpected keyword argument 'return_train_score'"

I thought this was a valid parameter for cross_val_score so I'm a bit confused any help would be appreciated,

thanks

results_dict = {
    "n_neighbors": [],
    "mean_train_score": [],
    "mean_cv_score": []}

results_dict

for k in range(2,20, 2):
    knn = KNeighborsClassifier(n_neighbors=k)
    scores = cross_val_score(knn, X_train, y_train, cv=5, return_train_score=True)
    results_dict['n_neighbors'].append(k)
    results_dict['mean_train_score'].append(['train_score'])
    results_dict['mean_cv_score'].append(['test_score'])

results_dict

CodePudding user response：

The error message suggests that the return_train_score argument is not a valid parameter for the cross_val_score() function. You can confirm this by looking at the documentation for the function, which you can access by running help(cross_val_score) or by looking it up online.

Assuming that return_train_score is not a valid parameter for the cross_val_score() function, you can fix the error by simply removing that argument from the function call.

Note that you may need to adjust the code that appends the scores to the results_dict dictionary depending on how the cross_val_score() function returns its results. You can refer to the documentation for the function to learn how to access the train and test scores.

CodePudding user response：

Check documentation of the cross_val_score function here

The cross_val_score doesn't have any such parameter named return_train_score. So, I suggest removing this parameter will eliminate this error.