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 feature, motivation and pitch

Usage of these two variables in tests is outdated. One should just write in the dtypes in question or use one of the ATen macro dispatch list functions (like get_all_fp_dtypes()).

Alternatives

No response

Additional context

No response

These examples take quite a long time to run, and they make our documentation CI fail quite frequently due to timeout. It'd be nice to speed the up a little bit.

To contributors: if you want to work on an example, first have a look at the example, and if you think you're comfortable working on it and have found a potential way to speed-up execution time while preserving the educational message

i was in need to define my own type <: AbstractSparseMatrixCSC (basically i needed the indexing behaviour of a SparseMatrixCSC, but with another storage type). The (not written in the docs) interface for an AbstractSparseMatrixCSC requires:

• getcolptr
• rowvals
• nonzeros
• size
  the problem is that an internal function _checkbuffers is required, mainly for the display machin

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.