Olivier Bougeant

For 1), this is likely caused by the fact that you are using a sckit-learn version >= 0.22 which is not compatible with ELI5, as explained here: https://github.com/TeamHG-Memex/eli5/issues/361. The issue...

Hi @jonas-nothnagel, it looks like @icfly2 has submitted a PR for this fix. I'm not sure what needs to happen for his work to be merged in.

@lopuhin Sorry I did not actually check if it was still failing. Awesome that you guys fixed that.

At the moment, `shap` requires `numba` which is expected to be compatible with Python 3.9 in mid-December (https://github.com/numba/numba/issues/6345)

numba and llvmlite should now have support for Python 3.9

Same problem here with my own data. `f0` is a single element in a nested list (e.g. [[0.1234]]).

@megangrannan This is a problem with `scipy==1.7.1`. The problem goes away if you use `scipy==1.4.1`, which is not compatible with Python3.9 though.

@kheilman, downgrade to `scipy==1.4.1`.

Can we get an update on this issue? I also faced it while trying to use `cuml.UMAP` to speed up `BERTopic`. I've noticed that the problem gets much worse when...

@MaartenGr, as the author of Bertopic, you might be interested in this.