problems using featurewiz in a pipeline

[sdaza]

Hi, I'm trying to use featurewiz in a pipeline, but it doesn't seem to be working.

from featurewiz import FeatureWiz
from sklearn.pipeline import Pipeline
from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor
pipeline = Pipeline([('fw', FeatureWiz()),('model', RandomForestClassifier())])

# %% 
pipeline.fit(X_train, y_train)

I got this error:

ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 2 dimensions. The detected shape was (2, 209334) + inhomogeneous part.

The training data is just two columns (numeric), and the target binary (0/1).

Thanks!

[AutoViML]

Hi @sdaza
Thanks for bringing this problem to my attention 👍

Can you upload a zip file with just 20 rows from your dataset so I can test the above and find out where this issue is happening?

Also can you try out the featurewiz-polars library below for your transformer since it is better suited for large dataset?
https://github.com/AutoViML/featurewiz_polars