ColossalAI/colossalai/shardformer/policies
flybird11111 20190b49a5
[shardformer] to fix whisper test failed due to significant accuracy differences. (#4710)
* [shardformer] fix whisper test failed

* [shardformer] fix whisper test failed

* [shardformer] fix whisper test failed

* [shardformer] fix whisper test failed
2023-09-14 21:34:20 +08:00
..
__init__.py [shardformer] init shardformer code structure (#3731) 2023-07-04 16:05:01 +08:00
auto_policy.py [Feature] The first PR to Add TP inference engine, kv-cache manager and related kernels for our inference system (#4577) 2023-09-12 01:22:56 +08:00
base_policy.py [shardformer] Add overlap support for gpt2 (#4535) 2023-08-29 18:30:50 +08:00
bert.py [shardformer/fix overlap bug] fix overlap bug, add overlap as an option in shardco… (#4516) 2023-08-28 17:16:40 +08:00
blip2.py [shardformer] chatglm support sequence parallel (#4482) 2023-08-22 23:59:31 +08:00
bloom.py [shardformer/fix overlap bug] fix overlap bug, add overlap as an option in shardco… (#4516) 2023-08-28 17:16:40 +08:00
chatglm2.py [shardformer] Pytree fix (#4533) 2023-09-04 17:52:23 +08:00
gpt2.py [shardformer] Add overlap support for gpt2 (#4535) 2023-08-29 18:30:50 +08:00
llama.py [shardformer] update llama2/opt finetune example and fix llama2 policy (#4645) 2023-09-09 22:45:36 +08:00
opt.py [shardformer] fix opt test hanging (#4521) 2023-08-30 14:50:34 +08:00
sam.py [shardformer] chatglm support sequence parallel (#4482) 2023-08-22 23:59:31 +08:00
t5.py [shardformer] fix opt test hanging (#4521) 2023-08-30 14:50:34 +08:00
vit.py [shardformer] vit/llama/t5 ignore the sequence parallelism flag and some fix. (#4498) 2023-08-24 15:50:02 +08:00
whisper.py [shardformer] to fix whisper test failed due to significant accuracy differences. (#4710) 2023-09-14 21:34:20 +08:00