Skip to content

[shardformer] update vit model #5530

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Apr 3, 2024
Merged

[shardformer] update vit model #5530

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6a528b5
update vit model
wangbluo Mar 28, 2024
1126501
remove the output_hidden_states
wangbluo Apr 3, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 11 additions & 13 deletions colossalai/shardformer/modeling/vit.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,8 @@ def _encoder_forward(
end_idx: int,
hidden_states: torch.Tensor,
head_mask: Optional[torch.Tensor] = None,
output_attentions: bool = False,
output_hidden_states: bool = False,
return_dict: bool = True,
stage_manager: PipelineStageManager = None,
) -> Union[tuple, BaseModelOutput]:
Expand All @@ -23,20 +25,14 @@ def _encoder_forward(
layer_head_mask = head_mask[i] if head_mask is not None else None

if encoder.gradient_checkpointing and encoder.training:

def create_custom_forward(module):
def custom_forward(*inputs):
return module(*inputs, False)

return custom_forward

layer_outputs = torch.utils.checkpoint.checkpoint(
create_custom_forward(layer_module),
hidden_states,
layer_head_mask,
)
layer_outputs = encoder._gradient_checkpointing_func(
layer_module.__call__,
hidden_states,
layer_head_mask,
output_attentions,
)
else:
layer_outputs = layer_module(hidden_states, layer_head_mask, False)
layer_outputs = layer_module(hidden_states, layer_head_mask, output_attentions)

hidden_states = layer_outputs[0]
if not stage_manager.is_last_stage():
Expand Down Expand Up @@ -112,6 +108,8 @@ def pp_forward(
end_idx=stage_index[1],
hidden_states=hidden_states,
head_mask=head_mask,
output_attentions=output_attentions,
output_hidden_states=output_hidden_states,
return_dict=return_dict,
stage_manager=stage_manager,
)
Expand Down