Group argument in fit method - sklearn

Question

I'm trying to run a cross validation with grouped data in Sklearn when I found that the fit method

fit(X, y=None, groups=None, **fit_params)

has a groups argument defined as:

groups : array-like, with shape (n_samples,), optional Group labels for the samples used while splitting the dataset into train/test set.

Is this the same thing as spliting my data with other iterators for grouped data (GroupKFold, LeaveOneGroupOut)? If not, what is the proper way to run GridSeachCV with grouped data?

Answer 1

Yes, they are same.

Please refer to the documentation of GridsearchCV fit():

groups : array-like, with shape (n_samples,), optional

Group labels for the samples used while splitting the dataset into train/test set.