FYI - Scaler fails to load under sklearn >= 0.23

[daisukelab]

Hi,

This is for sharing a possible issue who prefers a newer sklearn.
it seems that a new member variable has been added to the MinMaxScaler.

model = return_loaded_model(AudioEncoder, 'coala/saved_models/dual_ae_c/audio_encoder_epoch_200.pt')
extract_audio_embedding_chunks(model, '/data/UrbanSound8K/audio/fold1/101415-3-0-3.wav')
# --> fails with this error: 'MinMaxScaler' object has no attribute 'clip' /data/UrbanSound8K/audio/fold1/101415-3-0-3.wav

https://github.com/scikit-learn/scikit-learn/blob/2beed5584/sklearn/preprocessing/_data.py#L248

I wanted to work with the newer sklearn, then I manually added the new member as follows.

>>> a = pickle.load(open('coala/scaler_top_1000.pkl', 'rb'))
>>> a
MinMaxScaler(copy=True, feature_range=(0, 1))
>>> a.clip =False
>>> f = open('coala/scaler_top_1000_plus_clip.pkl', 'wb')
>>> pickle.dump(a, f)
>>> f.close()

I guess you might want to update the scaler_top_1000.pkl in this repo.

P.S. Thank you for sharing your code!