34j
34j
@weihua916 @zechengz @yiweny Would appreciate your review, thank you very much in advance.
No progress, sorry
- First, `torch_frame.DataSet` cannot properly handle `DataFrane` with indices not [0,1,...]. `dataset.py` L736 must be modified to `indices = np.where(self.df[self.split_col] == SPLIT_TO_NUM[split])[0]`. - ~~Second, though I am not aware of...
Almost done
@weihua916 Thanks for your patience, I've added tests.
@weihua916 Removed all tutorials and added `examples/sklearn_api.py` instead. The failing tests are probably due to a pandas version update and are not related to this PR.
@weihua916 Would you please reconsider merging this PR? This change makes it easy to try out `torch-frame` based neural networks on code that already uses `scikit-learn`. (which was demonstrated in...
@weihua916 @yiweny @akihironitta Any chance that this PR could be merged?
Thanks for your positive reply. In that case I would consider trying to make another package, but I am very busy right now, so please leave this as it is...
I have tried this and it seems to be very difficult. As a quick fix that isn't pretty, the following seems necessary: - Patch `skorch.utils.to_tensor_` to bypass `TensorFrame`. - Add...