Skip to content

Document ResNet architecture tweak #5977

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 13 commits into from
May 20, 2022
Merged

Document ResNet architecture tweak #5977

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
fd0eccb
To resolve issue #5964
puhuk May 9, 2022
1a563c2
Update resnet.py
puhuk May 10, 2022
94d9067
Merge branch 'main' into resnet_doc
puhuk May 10, 2022
a7f666d
Update resnet.py
puhuk May 10, 2022
737572a
Merge branch 'resnet_doc' of https://github.com/puhuk/vision into res…
puhuk May 10, 2022
2e5fed8
Update resnet.rst
puhuk May 10, 2022
a4c2e46
Merge branch 'main' into resnet_doc
puhuk May 10, 2022
d3a2ac7
Merge branch 'main' into resnet_doc
datumbox May 20, 2022
9216f5d
Fix stylings
datumbox May 20, 2022
b67a613
Add the same notes on model builders
datumbox May 20, 2022
f440e2d
Improve description
datumbox May 20, 2022
15fe780
Apply the change everywhere
datumbox May 20, 2022
ea44814
Remove trailing space
datumbox May 20, 2022
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
5 changes: 5 additions & 0 deletions docs/source/models/resnet.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,11 @@ ResNet
The ResNet model is based on the `Deep Residual Learning for Image Recognition
<https://arxiv.org/abs/1512.03385>`_ paper.

.. note::
The bottleneck of TorchVision places the stride for downsampling to the second 3x3
convolution while the original paper places it to the first 1x1 convolution.
This variant improves the accuracy and is known as `ResNet V1.5
<https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch>`_.

Model builders
--------------
Expand Down
18 changes: 18 additions & 0 deletions torchvision/models/resnet.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -699,6 +699,12 @@ def resnet34(*, weights: Optional[ResNet34_Weights] = None, progress: bool = Tru
def resnet50(*, weights: Optional[ResNet50_Weights] = None, progress: bool = True, **kwargs: Any) -> ResNet:
"""ResNet-50 from `Deep Residual Learning for Image Recognition <https://arxiv.org/pdf/1512.03385.pdf>`__.

.. note::
The bottleneck of TorchVision places the stride for downsampling to the second 3x3
convolution while the original paper places it to the first 1x1 convolution.
This variant improves the accuracy and is known as `ResNet V1.5
<https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch>`_.

Args:
weights (:class:`~torchvision.models.ResNet50_Weights`, optional): The
pretrained weights to use. See
Expand All @@ -724,6 +730,12 @@ def resnet50(*, weights: Optional[ResNet50_Weights] = None, progress: bool = Tru
def resnet101(*, weights: Optional[ResNet101_Weights] = None, progress: bool = True, **kwargs: Any) -> ResNet:
"""ResNet-101 from `Deep Residual Learning for Image Recognition <https://arxiv.org/pdf/1512.03385.pdf>`__.

.. note::
The bottleneck of TorchVision places the stride for downsampling to the second 3x3
convolution while the original paper places it to the first 1x1 convolution.
This variant improves the accuracy and is known as `ResNet V1.5
<https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch>`_.

Args:
weights (:class:`~torchvision.models.ResNet101_Weights`, optional): The
pretrained weights to use. See
Expand All @@ -749,6 +761,12 @@ def resnet101(*, weights: Optional[ResNet101_Weights] = None, progress: bool = T
def resnet152(*, weights: Optional[ResNet152_Weights] = None, progress: bool = True, **kwargs: Any) -> ResNet:
"""ResNet-152 from `Deep Residual Learning for Image Recognition <https://arxiv.org/pdf/1512.03385.pdf>`__.

.. note::
The bottleneck of TorchVision places the stride for downsampling to the second 3x3
convolution while the original paper places it to the first 1x1 convolution.
This variant improves the accuracy and is known as `ResNet V1.5
<https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch>`_.

Args:
weights (:class:`~torchvision.models.ResNet152_Weights`, optional): The
pretrained weights to use. See
Expand Down