Machine learning

Current implementation of Go binding can not specify options.

GPUOptions struct is in internal package. And go generate doesn't work for protobuf directory. So we can't specify GPUOptions for NewSession.

📚 The doc issue

The documentation of multilabel_soft_margin_loss misses reduction parameter.

Suggest a potential alternative/fix

The right signature should be

torch.nn.functional.multilabel_soft_margin_loss(input, target, weight=None, size_average=None, reduction='mean') → 

Describe the issue linked to the documentation

The homepage has a link name "Maling list: [email protected]" which points to https://mail.python.org/mailman/listinfo/scikit-learn

(Would prefer it indicate the sign-up page for mailing list instead of an email address for posting to mailing list.)

Suggest a potential alternative/fix

Update this text "Mailing list: sciki

It might be useful to have rationalize(x) return x if x is already rational, to allow generic code that takes both reals and rationals as inputs.

Currently many more Python projects like dask and optuna are using Python type hints. With the Python package of xgboost gaining more and more features, we should also adopt mypy as a safe guard against some type errors and for better code documentation.