[BUG]: not all models compared in `compare_models` about pycaret HOT 3 OPEN

I got the similar issue. the test code on git hub does not work .compare_model() return empty set.

from pycaret.

If I set preprocess=True and omit the pca* args, then I see again all the models being compared. Also seems that including categorical data was problematic - works as expected if I transform manually from categorical to numerical. Would be nice to get a warning why the models were left out.

from pycaret.

I also saw that mlp was not included in the output. It appears that it was silently dropped, because my dataset had missing values.

An error is correctly thrown if I try to call it directly

mlp = exp.create_model('mlp')
ValueError: Input X contains NaN.
MLPRegressor does not accept missing values encoded as NaN natively.

from pycaret.