ColossalAI/colossalai/shardformer/modeling
FoolPlayer 879301d0da [shardformer] support Blip2 (#4243)
* support base blip2

* add support for downstream blip2 model

* update readme

* add forward injection

* skip not compatible models test

* fix test for gemini and low_level_zero_pugin
2023-08-15 23:25:14 +08:00
..
__init__.py [shardformer] added development protocol for standardization (#4149) 2023-07-04 16:05:01 +08:00
bert.py [pipeline] fix return_dict/fix pure_pipeline_test (#4331) 2023-08-15 23:25:14 +08:00
blip2.py [shardformer] support Blip2 (#4243) 2023-08-15 23:25:14 +08:00
bloom.py [pipeline] fix return_dict/fix pure_pipeline_test (#4331) 2023-08-15 23:25:14 +08:00
gpt2.py [pipeline] fix return_dict/fix pure_pipeline_test (#4331) 2023-08-15 23:25:14 +08:00
llama.py [plugin] add 3d parallel plugin (#4295) 2023-08-15 23:25:14 +08:00
sam.py [shardformer] support Blip2 (#4243) 2023-08-15 23:25:14 +08:00
t5.py [pipeline] add pipeline support for all T5 models (#4310) 2023-08-15 23:25:14 +08:00
vit.py [shardformer] support pipeline base vit model (#4284) 2023-08-15 23:25:14 +08:00