ColossalAI/tests/test_shardformer
Insu Jang 00525f7772
[shardformer] fix pipeline forward error if custom layer distribution is used (#5189)
* Use self.[distribute_layers|get_stage_index] to exploit custom layer distribution

* Change static methods for t5 layer distribution to member functions

* Change static methods for whisper layer distribution to member functions

* Replace whisper policy usage with self one

* Fix test case to use non-static layer distribution methods

* fix: fix typo

---------

Co-authored-by: Wenhao Chen <cwher@outlook.com>
2024-03-27 13:57:00 +08:00
..
test_hybrid_parallel_grad_clip_norm [gemini] gemini support extra-dp (#5043) 2023-11-16 21:03:04 +08:00
test_layer [shardformer] fix pipeline forward error if custom layer distribution is used (#5189) 2024-03-27 13:57:00 +08:00
test_model [shardformer] update colo attention to support custom mask (#5510) 2024-03-27 11:19:32 +08:00
__init__.py [shardformer] adapted T5 and LLaMa test to use kit (#4049) 2023-07-04 16:05:01 +08:00
test_flash_attention.py [shardformer] update colo attention to support custom mask (#5510) 2024-03-27 11:19:32 +08:00
test_shard_utils.py [misc] update pre-commit and run all files (#4752) 2023-09-19 14:20:26 +08:00
test_with_torch_ddp.py [ci] fixed ddp test (#5254) 2024-01-11 17:16:32 +08:00