..
__init__.py
[shardformer] init shardformer code structure ( #3731 )
2023-07-04 16:05:01 +08:00
auto_policy.py
[shardformer] support ep for deepseek v3 ( #6185 )
2025-02-11 16:10:25 +08:00
base_policy.py
[fp8] Merge feature/fp8_comm to main branch of Colossalai ( #6016 )
2024-08-22 09:21:34 +08:00
bert.py
[Zerobubble] merge main. ( #6142 )
2024-11-19 19:00:36 +08:00
blip2.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
bloom.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
chatglm2.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
command.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
deepseek.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
deepseek_v3.py
[shardformer] support pipeline for deepseek v3 and optimize lora save ( #6188 )
2025-02-14 14:48:54 +08:00
falcon.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
gpt2.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
gptj.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
llama.py
[Zerobubble] merge main. ( #6142 )
2024-11-19 19:00:36 +08:00
mistral.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
mixtral.py
[Zerobubble] merge main. ( #6142 )
2024-11-19 19:00:36 +08:00
opt.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
qwen2.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
sam.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
t5.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
vit.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00
whisper.py
[Sharderformer] Support zbv in Sharderformer Policy ( #6150 )
2025-01-02 10:22:26 +08:00