mirror of
https://github.com/hpcaitech/ColossalAI.git
synced 2025-09-02 17:46:42 +00:00
[booster] add low level zero plugin (#3594)
* [booster] add low level zero plugin * [booster] fix gemini plugin test * [booster] fix precision * [booster] add low level zero plugin test * [test] fix booster plugin test oom * [test] fix booster plugin test oom * [test] fix googlenet and inception output trans * [test] fix diffuser clip vision model * [test] fix torchaudio_wav2vec2_base * [test] fix low level zero plugin test
This commit is contained in:
@@ -18,6 +18,7 @@ data_vae_fn = lambda: dict(sample=torch.randn(2, 3, 32, 32))
|
||||
data_unet_fn = lambda: dict(sample=torch.randn(2, 3, 32, 32), timestep=3)
|
||||
|
||||
identity_output = lambda x: x
|
||||
clip_vision_model_output = lambda x: dict(pooler_output=x[1])
|
||||
|
||||
|
||||
def data_clip_model():
|
||||
@@ -65,7 +66,7 @@ model_zoo.register(name='diffusers_clip_text_model',
|
||||
model_zoo.register(name='diffusers_clip_vision_model',
|
||||
model_fn=partial(transformers.CLIPVisionModel, config=transformers.CLIPVisionConfig()),
|
||||
data_gen_fn=data_clip_vision,
|
||||
output_transform_fn=identity_output)
|
||||
output_transform_fn=clip_vision_model_output)
|
||||
|
||||
model_zoo.register(name='diffusers_unet2d_model',
|
||||
model_fn=diffusers.UNet2DModel,
|
||||
|
Reference in New Issue
Block a user