Daniel Herman

> Moving data in and out of polars via numpy can be zero-copy (it depends on the data type, null data and dimensions) Does this imply that potentially LigthGBM could...

So, if I understand it correctly as of now there is now way how to pass data from polars to LightGBM without copying the data in memory. For the project...

> Potentially, a viable temporary alternative might also be to pass data via files (?) Is it possible directly in python? I could then output data into temp file and...

No I don't have any update on it

+1 Because scikit-learn=0.24.0 is really outdated.

> > +1 Because scikit-learn=0.24.0 is really outdated. > > Just another reason to update auto-sklearn to the latest scikit-learn versions right? Exactly, not sure whether auto-sklearn is slowly dying...

This would be a great enhancement.

Most likely this will be solved in https://github.com/dwgoon/jpegio/pull/27 could you check it now please? @mohaoran93

I think there should be a different way than `.profile()`, the last I tried it on huge `DataFrame`, my jupyter lab ran out of RAM. I think just getting total...

It works when I use `transformers==4.16.0` ```python import sys if 'google.colab' in sys.modules: print('Running in Colab.') !pip3 install transformers==4.16.0 timm==0.4.12 fairscale==0.4.4 ```