Mikhail Korobov

Hi @igstan, No, unfortunately it is not possible currently. The following is missing: 1) there is no support for psd writing - currently psd-tools can only read psd files; 2)...

Hey @ihrankouski! 1) nothing changed; 2) there is a PR with a much more advanced layer composition: https://github.com/psd-tools/psd-tools/pull/49; it is not merged and requires some work; 3) as I recall,...

@qh582 looks good! Would you mind updating a few notes in docs, mentioning this new feature? * https://github.com/TeamHG-Memex/eli5/blob/master/README.rst * https://github.com/TeamHG-Memex/eli5/blob/master/docs/source/libraries/lightgbm.rst * https://github.com/TeamHG-Memex/eli5/blob/master/docs/source/overview.rst

Hm, I'm not sure bias makes sense for explain_weights + xgboost regressor; regressor predicts values regardless of mean; GBMs can handle shifts in data without any special handling, there is...

Thanks, looks like a great feature! Tests are failing in Python 2.7; I'm not sure what it takes to make it work. Does it at least work with n_jobs=1?

Hey @AshwinB-hat ! SHAP an interesting beast, because on a practical side, it combines both an algorithm which is LIME-like, and an algorithm which is treeinterpreter-like. They have somewhat different...

@DomHudson crfsuite does support continuous features

It should be possible to do so, but I haven't checked it in detail. In the meantime, you can also use processes.

@Ridinstar if you did call tagger.open it is likely a different issue. Do you have an example, how to reproduce it?

@Ridinstar mistake in formatting training data shouldn't cause a segfault, it is a python-crfsuite bug if it does :)