ColossalAI/tests
Edenzzzz f5c84af0b0
[Feature] Zigzag Ring attention (#5905)
* halfway

* fix cross-PP-stage position id length diff bug

* fix typo

* fix typo

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* unified cross entropy func for all shardformer models

* remove redundant lines

* add basic ring attn; debug cross entropy

* fwd bwd logic complete

* fwd bwd logic complete; add experimental triton rescale

* precision tests passed

* precision tests passed

* fix typos and remove misc files

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* add sp_mode to benchmark; fix varlen interface

* update softmax_lse shape by new interface

* change tester name

* remove buffer clone; support packed seq layout

* add varlen tests

* fix typo

* all tests passed

* add dkv_group; fix mask

* remove debug statements

---------

Co-authored-by: Edenzzzz <wtan45@wisc.edu>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2024-08-16 13:56:38 +08:00
..
kit [Feature] Zigzag Ring attention (#5905) 2024-08-16 13:56:38 +08:00
test_analyzer [misc] update pre-commit and run all files (#4752) 2023-09-19 14:20:26 +08:00
test_auto_parallel [test] Fix/fix testcase (#5770) 2024-06-03 15:26:01 +08:00
test_autochunk [misc] update pre-commit and run all files (#4752) 2023-09-19 14:20:26 +08:00
test_booster [Feature] Zigzag Ring attention (#5905) 2024-08-16 13:56:38 +08:00
test_checkpoint_io [Feature] Zigzag Ring attention (#5905) 2024-08-16 13:56:38 +08:00
test_cluster [misc] refactor launch API and tensor constructor (#5666) 2024-04-29 10:40:11 +08:00
test_config [misc] update pre-commit and run all files (#4752) 2023-09-19 14:20:26 +08:00
test_device [misc] refactor launch API and tensor constructor (#5666) 2024-04-29 10:40:11 +08:00
test_fx [hotfix] fix testcase in test_fx/test_tracer (#5779) 2024-06-05 11:29:32 +08:00
test_infer [shardformer] upgrade transformers to 4.39.3 (#5815) 2024-06-14 10:59:33 +08:00
test_lazy [lazyinit] skip whisper test (#5653) 2024-04-26 14:03:12 +08:00
test_legacy [chore] solve moe ckpt test failure and some other arg pass failure 2024-08-01 10:06:59 +08:00
test_lora [Feature] Zigzag Ring attention (#5905) 2024-08-16 13:56:38 +08:00
test_moe [moe] refactor mesh assignment 2024-08-01 10:06:59 +08:00
test_optimizer [MoE/ZeRO] Moe refactor with zero refactor (#5821) 2024-06-28 14:00:08 +08:00
test_pipeline [Feature] Zigzag Ring attention (#5905) 2024-08-16 13:56:38 +08:00
test_shardformer [Feature] Zigzag Ring attention (#5905) 2024-08-16 13:56:38 +08:00
test_smoothquant [inference] Add smmoothquant for llama (#4904) 2023-10-16 11:28:44 +08:00
test_tensor [misc] refactor launch API and tensor constructor (#5666) 2024-04-29 10:40:11 +08:00
test_zero [zero] support all-gather overlap (#5898) 2024-07-11 18:59:59 +08:00
__init__.py [zero] Update sharded model v2 using sharded param v2 (#323) 2022-03-11 15:50:28 +08:00