GitLab

* [CLI] add CLI launcher

* Revert "[CLI] add CLI launcher"

This reverts commit df7e6506d4500af6a9220ef7fe4d3c7b1daebd4c.

* [context]support arbitary module materialization.

* [test]add numerical check for lazy init context.

* [CLI] add CLI launcher

* Revert "[CLI] add CLI launcher"

This reverts commit df7e6506d4500af6a9220ef7fe4d3c7b1daebd4c.

* [context]use meta tensor to init model lazily.

* polish

* make module with device kwargs bypass the normal init.

* change unit test to adapt updated context.

* add save/load state dict for ColoDDP

* add unit test

* refactor unit test folder

* polish unit test

* rename unit test

* add set_params_to_ignore for ColoDDP

* polish code

* fix zero hook v2

* add unit test

* polish docstr

* [pipeline] refactor the pipeline module

* polish code

* [context] support lazy init of module

* polish code

* [test] skip tests when not enough GPUs are detected

* polish code

* polish code

* [test] ignore 8 gpu test

* polish code

* polish workflow

* polish workflow

* refactor StatefulTensor, tensor utilities

* add unitest for GeminiMemoryManager

* [CLI] add CLI launcher

* Revert "[CLI] add CLI launcher"

This reverts commit df7e6506d4500af6a9220ef7fe4d3c7b1daebd4c.

* [pipeline]add module lazy init feature to support large model initization.

* [pipeline]add to_layer_list and partition method to support arbitrary non-pp model

* refactor the module structure

* polish

* [pipelinable]add unit test for pipelinable

* polish

* polish

* Fix CodeFactor issues.

* [test] refactored with the new rerun decorator

* polish test case

* remove hybrid adam in test_moe_zero_optim

* fix activation checkpointing and its unitest

* [zero] polish sharded param name

* polish code

* polish

* polish code

* polish

* polsih

* polish

* Added tensor detector

* Added the - states

* Allowed change include_cpu when detect()

* refactored test with component func

* fixed bug