fix: 424 allow any np.intX as training target

dreamquark-ai / tabnet

PyTorch implementation of TabNet paper : https://arxiv.org/pdf/1908.07442.pdf

https://dreamquark-ai.github.io/tabnet/

MIT License

2.55k stars 470 forks source link

Closed Optimox closed 1 year ago

Optimox commented 1 year ago

What kind of change does this PR introduce? Bug fix to avoid save load errors when targets are np.int8, np.int16, np.int32

Does this PR introduce a breaking change?

No this is just a small bugfix

What needs to be documented once your changes are merged?

I don't think there is a need for documentation here.

Closing issues

closes #424