Richard Stotz

Hi, thank you for providing additional details. I looked into the tree implementation and it seems like it's just the tree root that does not have a setter method -...

Tl;Dr: You need to manually fix the input signature: ```python def copy_model_sig(m): """ Copy the model signature to a new model. """ spec = m.save_spec()[0][0] return lambda insp: spec ......

Hi, Unfortunately, TF-DF cannot serialize all its methods within the Keras model format. This means that some functions, including `predict_get_leaves` are not available in saved models, and this includes all...

Hi @Realvincentyuan, looking forward to your PRs! I don't remember a specific reason for not not including and id for non-leaf nodes - IIRC, there just wasn't a use case...

Hi, a subdirectory of `contrib` is probably best for this. Let's discuss the other question on the PR directly.

Hi, sorry for not getting back to you for a while. The article looks good! I wonder if the best strategy would be to just create an "External examples" file...

Hi, thank you for reporting this. The error message given by TF-DF is unfortunately not very clear, but I believe that the Random Forest model has not been trained. Note...

I don't think the Keras tuner or the TF-DF tuner support tuning CNN and RF at the same time - @achoum can you please confirm? [YDF](https://ydf.readthedocs.io/) runs on Windows and...

Yes, we are planning to make TF-DF compatible with Keras 3, but there is no timeline yet, since it might be a non-trivial change.

Hi, looks like we have to update this class - thank you for notifying us. Very likely, you only need to add ``` import os # Keep using Keras 2...