GitLab

* [workflow] added cuda extension build test before release

* polish code

* [workflow] added test-pypi check before release

* polish code

* [workflow] added notification if scheduled build fails

* polish code

* polish code

* [workflow] added discussion stats to community report

* polish code

* [workflow] adjust the GPU memory threshold for scheduled unit test

* polish code

* [workflow] fixed example check workflow

* polish yaml

* [workflow] added contributor and user-engagement report

* polish code

* polish code

* update readme

* update

* update

bug/fix diffusion ckpt problem

Co-authored-by: github-actions <github-actions@github.com>

* [kernel] fixed repeated loading of kernels

* polish code

* polish code

* [doc] added pull  request template

* polish code

* polish code

Support multi outputs chunk search. Previously we only support single output chunk search. It is more flexible and improve performance by a large margin. For transformer, we reduce memory by 40% than previous search strategy.

1. rewrite search strategy to support multi outputs chunk search
2. fix many, many bugs
3. update tests

fix README

* [workflow] only report coverage for changed files

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file

* polish file