Skip to content

⚡️ Speed up method ClassifierHead.forward by 7% #5

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

Open
codeflash-ai wants to merge 1 commit into
base: main
Choose a base branch
from
Open

⚡️ Speed up method ClassifierHead.forward by 7% #5

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
17 changes: 13 additions & 4 deletions doctr/models/classification/vit/pytorch.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,8 @@ def __init__(

def forward(self, x: torch.Tensor) -> torch.Tensor:
# (batch_size, num_classes) cls token
return self.head(x[:, 0])
x0 = x.select(1, 0)
return self.head(x0)


class VisionTransformer(nn.Sequential):
Expand Down Expand Up @@ -90,7 +91,9 @@ def __init__(
) -> None:
_layers: list[nn.Module] = [
PatchEmbedding(input_shape, d_model, patch_size),
EncoderBlock(num_layers, num_heads, d_model, d_model * ffd_ratio, dropout, nn.GELU()),
EncoderBlock(
num_layers, num_heads, d_model, d_model * ffd_ratio, dropout, nn.GELU()
),
]
if include_top:
_layers.append(ClassifierHead(d_model, num_classes))
Expand All @@ -114,7 +117,9 @@ def _vit(
ignore_keys: list[str] | None = None,
**kwargs: Any,
) -> VisionTransformer:
kwargs["num_classes"] = kwargs.get("num_classes", len(default_cfgs[arch]["classes"]))
kwargs["num_classes"] = kwargs.get(
"num_classes", len(default_cfgs[arch]["classes"])
)
kwargs["input_shape"] = kwargs.get("input_shape", default_cfgs[arch]["input_shape"])
kwargs["classes"] = kwargs.get("classes", default_cfgs[arch]["classes"])

Expand All @@ -130,7 +135,11 @@ def _vit(
if pretrained:
# The number of classes is not the same as the number of classes in the pretrained model =>
# remove the last layer weights
_ignore_keys = ignore_keys if kwargs["num_classes"] != len(default_cfgs[arch]["classes"]) else None
_ignore_keys = (
ignore_keys
if kwargs["num_classes"] != len(default_cfgs[arch]["classes"])
else None
)
model.from_pretrained(default_cfgs[arch]["url"], ignore_keys=_ignore_keys)

return model
Expand Down