InvalidParameterError: The 'scoring' parameter of GridSearchCV must be a str among

Question

When I try to fit the training dataset to the GridsearchCV, it states that the scoring must be a str among:

{'jaccard_samples', 'precision_macro', 'balanced_accuracy', ...<long list of allowed classes, see edit history>...}, 
a callable, an instance of 'list', an instance of 'tuple', an instance of 'dict' or None. 
Got {'precision', 'recall', 'accuracy', 'f1'} instead."

Even though my scoring is a list of: "{'precision','f1','recall','accuracy'}".

I run the code with the Jupyter kernel extension in VS code. I tried changing it to only go for the scoring of 'Recall' and it worked, however that I would like to have multiple scoring variables.

Answer 1

The allowed types for scoring in GridSearchCV are

str, callable, list, tuple or dict,

You are passing a set which is not a valid type.

Use scoring=['precision','f1','recall','accuracy'] with square brackets not curly ones.

---

To more comprehensive for multiple scoring types you can pass (see linked doc) :

• a list or tuple of unique strings;
• a callable returning a dictionary where the keys are the metric names and the values are the metric scores;
• a dictionary with metric names as keys and callables a values.

---