Passing explicit pretrained_backbone #74372

datumbox · 2022-03-17T13:03:27Z

Summary: In preparation to the multi-weight support porting, we pass explicitly the pretrained_blackbone value. We use the default value True for most cases, except for when the use-case is clearly a test and thus should avoid downloading the weights of the backbone.

Test Plan: running project unit-tests

Differential Revision: D34961147

pytorch-bot · 2022-03-17T13:03:32Z

CI Flow Status

⚛️ CI Flow

Ruleset - Version: v1
Ruleset - File: https://github.com/datumbox/pytorch/blob/e078f066764b58465f8c9bc081cb0785b05f7302/.github/generated-ciflow-ruleset.json
PR ciflow labels: ciflow/default
Add ciflow labels to this PR to trigger more builds:

Workflows	Labels (bold enabled)	Status
Triggered Workflows
deploy-linux-xenial-cuda11.3-py3.7-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
linux-binary-conda	`ciflow/binaries`, `ciflow/binaries_conda`, `ciflow/default`	✅ triggered
linux-binary-libtorch-cxx11-abi	`ciflow/all`, `ciflow/binaries`, `ciflow/binaries_libtorch`, `ciflow/default`, `ciflow/trunk`	✅ triggered
linux-binary-libtorch-pre-cxx11	`ciflow/all`, `ciflow/binaries`, `ciflow/binaries_libtorch`, `ciflow/default`, `ciflow/trunk`	✅ triggered
linux-binary-manywheel	`ciflow/all`, `ciflow/binaries`, `ciflow/binaries_wheel`, `ciflow/default`, `ciflow/trunk`	✅ triggered
linux-bionic-py3.7-clang9	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/noarch`, `ciflow/trunk`	✅ triggered
linux-bionic-rocm4.5-py3.7	`ciflow/all`, `ciflow/default`, `ciflow/linux`, `ciflow/rocm`, `ciflow/trunk`	✅ triggered
linux-docs	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/docs`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
linux-vulkan-bionic-py3.7-clang9	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`, `ciflow/vulkan`	✅ triggered
linux-xenial-cuda11.3-py3.7-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
linux-xenial-cuda11.3-py3.7-gcc7-bazel-test	`ciflow/all`, `ciflow/bazel`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
linux-xenial-py3-clang5-mobile-build	`ciflow/all`, `ciflow/default`, `ciflow/linux`, `ciflow/mobile`, `ciflow/trunk`	✅ triggered
linux-xenial-py3-clang5-mobile-custom-build-static	`ciflow/all`, `ciflow/default`, `ciflow/linux`, `ciflow/mobile`, `ciflow/trunk`	✅ triggered
linux-xenial-py3.7-clang7-asan	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/sanitizers`, `ciflow/trunk`	✅ triggered
linux-xenial-py3.7-clang7-onnx	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/onnx`, `ciflow/trunk`	✅ triggered
linux-xenial-py3.7-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
linux-xenial-py3.7-gcc5.4-mobile-lightweight-dispatch-build	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/libtorch`, `ciflow/linux`, `ciflow/mobile`, `ciflow/trunk`	✅ triggered
linux-xenial-py3.7-gcc7	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
linux-xenial-py3.7-gcc7-no-ops	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
macos-arm64-binary-conda	`ciflow/binaries`, `ciflow/binaries_conda`, `ciflow/default`	✅ triggered
macos-arm64-binary-wheel	`ciflow/binaries`, `ciflow/binaries_wheel`, `ciflow/default`	✅ triggered
macos-binary-conda	`ciflow/binaries`, `ciflow/binaries_conda`, `ciflow/default`	✅ triggered
macos-binary-libtorch-cxx11-abi	`ciflow/binaries`, `ciflow/binaries_libtorch`, `ciflow/default`	✅ triggered
macos-binary-libtorch-pre-cxx11	`ciflow/binaries`, `ciflow/binaries_libtorch`, `ciflow/default`	✅ triggered
macos-binary-wheel	`ciflow/binaries`, `ciflow/binaries_wheel`, `ciflow/default`	✅ triggered
pytorch-linux-xenial-py3-clang5-android-ndk-r19c-gradle-custom-build-single	`ciflow/all`, `ciflow/android`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
pytorch-linux-xenial-py3-clang5-android-ndk-r19c-gradle-custom-build-single-full-jit	`ciflow/all`, `ciflow/android`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/trunk`	✅ triggered
win-vs2019-cpu-py3	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/trunk`, `ciflow/win`	✅ triggered
win-vs2019-cuda11.3-py3	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/trunk`, `ciflow/win`	✅ triggered
windows-binary-conda	`ciflow/binaries`, `ciflow/binaries_conda`, `ciflow/default`	✅ triggered
windows-binary-libtorch-debug	`ciflow/all`, `ciflow/binaries`, `ciflow/binaries_libtorch`, `ciflow/default`, `ciflow/trunk`	✅ triggered
windows-binary-libtorch-release	`ciflow/all`, `ciflow/binaries`, `ciflow/binaries_libtorch`, `ciflow/default`, `ciflow/trunk`	✅ triggered
windows-binary-wheel	`ciflow/all`, `ciflow/binaries`, `ciflow/binaries_wheel`, `ciflow/default`, `ciflow/trunk`	✅ triggered
Skipped Workflows
caffe2-linux-xenial-py3.7-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`, `ciflow/trunk`	🚫 skipped
docker-builds	`ciflow/all`, `ciflow/trunk`	🚫 skipped
ios-12-5-1-arm64	`ciflow/all`, `ciflow/ios`, `ciflow/macos`, `ciflow/scheduled`	🚫 skipped
ios-12-5-1-arm64-coreml	`ciflow/all`, `ciflow/ios`, `ciflow/macos`, `ciflow/scheduled`	🚫 skipped
ios-12-5-1-arm64-custom-ops	`ciflow/all`, `ciflow/ios`, `ciflow/macos`, `ciflow/scheduled`	🚫 skipped
ios-12-5-1-arm64-metal	`ciflow/all`, `ciflow/ios`, `ciflow/macos`, `ciflow/scheduled`	🚫 skipped
ios-12-5-1-x86-64	`ciflow/all`, `ciflow/ios`, `ciflow/macos`, `ciflow/trunk`	🚫 skipped
ios-12-5-1-x86-64-coreml	`ciflow/all`, `ciflow/ios`, `ciflow/macos`, `ciflow/trunk`	🚫 skipped
libtorch-linux-xenial-cuda10.2-py3.7-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`, `ciflow/trunk`	🚫 skipped
libtorch-linux-xenial-cuda11.3-py3.7-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`, `ciflow/trunk`	🚫 skipped
linux-bionic-cuda10.2-py3.9-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/slow`, `ciflow/trunk`	🚫 skipped
linux-bionic-rocm4.5-py3.7-distributed	`ciflow/all`, `ciflow/linux`, `ciflow/rocm`, `ciflow/trunk`	🚫 skipped
linux-docs-push	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
linux-xenial-cuda11.3-py3.7-gcc7-no-ops	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/trunk`	🚫 skipped
macos-10-15-py3-arm64	`ciflow/all`, `ciflow/macos`, `ciflow/trunk`	🚫 skipped
macos-10-15-py3-lite-interpreter-x86-64	`ciflow/all`, `ciflow/macos`, `ciflow/trunk`	🚫 skipped
macos-11-py3-x86-64	`ciflow/all`, `ciflow/macos`, `ciflow/trunk`	🚫 skipped
parallelnative-linux-xenial-py3.7-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`, `ciflow/trunk`	🚫 skipped
periodic-libtorch-linux-bionic-cuda11.5-py3.7-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-linux-bionic-cuda11.5-py3.7-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-linux-xenial-cuda10.2-py3-gcc7-slow-gradcheck	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`, `ciflow/slow`, `ciflow/slow-gradcheck`	🚫 skipped
periodic-linux-xenial-cuda11.3-py3.7-gcc7-debug	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-win-vs2019-cuda11.5-py3	`ciflow/all`, `ciflow/cuda`, `ciflow/scheduled`, `ciflow/win`	🚫 skipped
pytorch-linux-xenial-py3-clang5-android-ndk-r19c-build	`ciflow/all`, `ciflow/android`, `ciflow/cpu`, `ciflow/linux`, `ciflow/trunk`	🚫 skipped
pytorch-xla-linux-bionic-py3.7-clang8	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`, `ciflow/trunk`, `ciflow/xla`	🚫 skipped

facebook-github-bot · 2022-03-17T13:03:32Z

🔗 Helpful links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/74372
📄 Preview docs built from this PR
📄 Preview C++ docs built from this PR
↩️ [fb-only] Re-run with SSH instructions
🔧 Opt-in to CIFlow to control what jobs run on your PRs

💊 CI failures summary and remediations

As of commit 013a21e (more details on the Dr. CI page):

1/1 failures introduced in this PR

1 failure not recognized by patterns:

Job	Step	Action
^{pytorch-xla-linux-bionic-py3.7-clang8 / test (xla, 1, 1, linux.2xlarge)}	^Test	🔁 rerun

This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

facebook-github-bot · 2022-03-17T13:03:59Z

This pull request was exported from Phabricator. Differential Revision: D34961147

facebook-github-bot · 2022-03-18T09:34:03Z

This pull request was exported from Phabricator. Differential Revision: D34961147

datumbox · 2022-03-19T08:15:34Z

@BowenBao @shubhambhokare1 I believe the failing test is flaky and unrelated to the changes. The specific test is not impacted by the PR:

pytorch/test/onnx/test_pytorch_onnx_onnxruntime.py

Lines 10536 to 10548 in 57c7bf7

    
           def test_quantized_linear(self): 
        
               model = torch.nn.quantized.Linear(4, 8) 
        
               # Set non-zero bias. 
        
               bias = torch.arange(8).to(torch.float) 
        
               model.set_weight_bias(model.weight(), bias) 
        
               input = torch.randn(4, 4) 
        
               input_tensor = torch.quantize_per_tensor(input, 0.5, 128, torch.quint8) 
        
               # Currently, we need convert the model to ScriptModule before export. 
        
               # The reason is that PackedParams contains int (not tensor). 
        
               # Then it fails when the exporter calls _trace_and_get_graph_from_model(). 
        
               # TODO: https://msdata.visualstudio.com/Vienna/_workitems/edit/1547858 
        
               self.run_test(torch.jit.trace(model, input_tensor), (input_tensor,)) 
        
               self.run_test(torch.jit.script(model), (input_tensor,))

Edit: Confirmed that the test is flaky by rerunning the test and now it passes.

Summary: Pull Request resolved: pytorch#74372 In preparation to the multi-weight support porting, we pass explicitly the pretrained_blackbone value. We use the default value `True` for most cases, except for when the use-case is clearly a test and thus should avoid downloading the weights of the backbone. Test Plan: running project unit-tests Reviewed By: jdsgomes Differential Revision: D34961147 fbshipit-source-id: bb5a9e062e24d1f7c76f1b56072272beb93ccd6d

facebook-github-bot · 2022-03-21T12:11:18Z

This pull request was exported from Phabricator. Differential Revision: D34961147

datumbox · 2022-03-21T14:11:37Z

The XLA test which fails is unrelated. The master branch currently fails as well.

datumbox · 2022-03-22T09:59:50Z

@BowenBao @shubhambhokare1 could you please review? :)

BowenBao · 2022-03-22T17:24:05Z

@datumbox Could point me to a link where explains the relation of pretrained and pretrained_backbone? Will there be changes to behavior in coming versions?

datumbox · 2022-03-22T17:37:00Z

@BowenBao Both parameters existed for multiple releases. On detection/segmentation models, pretrained=True means load fully pretrained weights. pretrained_backbone=True means load partially pretrained weights on the backbone and only takes effect if pretrained=False.

This PR is part of a larger update to explicitly define the value of pretrained_backbone in every use and try to turn it off where possible (especially on tests). Many users don't currently realise that the default value is True meaning that they download weights by default. We are considering changing this behaviour on the future, and that's why we would like at least within pytorch to be explicit about the parameters.

Also note that TorchVision just released the Multi-weight support API which deprecates the pretrained/pretrained_backbone APIs in favour of weights/weights_backbone. You can read more details here: pytorch/vision#5618.

test/onnx/test_pytorch_onnx_onnxruntime.py

BowenBao

LGTM, thanks for explanation!

Summary: Pull Request resolved: #74372 In preparation to the multi-weight support porting, we pass explicitly the pretrained_blackbone value. We use the default value `True` for most cases, except for when the use-case is clearly a test and thus should avoid downloading the weights of the backbone. Test Plan: running project unit-tests Reviewed By: jdsgomes Differential Revision: D34961147 fbshipit-source-id: cf29e42545302716a7cd3f3eb0d69e44d5fb6c73

github-actions · 2022-03-22T18:37:23Z

Hey @datumbox.
You've committed this PR, but it does not have both a 'release notes: ...' and 'topics: ...' label. Please add one of each to the PR. The 'release notes: ...' label should represent the part of PyTorch that this PR changes (fx, autograd, distributed, etc) and the 'topics: ...' label should represent the kind of PR it is (not user facing, new feature, bug fix, perf improvement, etc). The list of valid labels can be found here for the 'release notes: ...' and here for the 'topics: ...'.
For changes that are 'topic: not user facing' there is no need for a release notes label.

Summary: Pull Request resolved: #74372 In preparation to the multi-weight support porting, we pass explicitly the pretrained_blackbone value. We use the default value `True` for most cases, except for when the use-case is clearly a test and thus should avoid downloading the weights of the backbone. Test Plan: running project unit-tests Reviewed By: jdsgomes Differential Revision: D34961147 fbshipit-source-id: cf29e42545302716a7cd3f3eb0d69e44d5fb6c73 (cherry picked from commit c4613b7)

datumbox requested review from BowenBao and shubhambhokare1 as code owners March 17, 2022 13:03

pytorch-bot bot added the ciflow/default label Mar 17, 2022

facebook-github-bot added the cla signed label Mar 17, 2022

facebook-github-bot added the fb-exported label Mar 17, 2022

datumbox force-pushed the export-D34961147 branch from e078f06 to 22c045d Compare March 18, 2022 09:34

datumbox force-pushed the export-D34961147 branch from 22c045d to 013a21e Compare March 21, 2022 12:11

BowenBao reviewed Mar 22, 2022

View reviewed changes

test/onnx/test_pytorch_onnx_onnxruntime.py Show resolved Hide resolved

BowenBao approved these changes Mar 22, 2022

View reviewed changes

pytorchmergebot closed this in ea56b9d Mar 22, 2022

datumbox added release notes: onnx torch.onnx related changes that should show up in the release notes topic: improvements topic category labels Mar 23, 2022

datumbox deleted the export-D34961147 branch March 23, 2022 08:16

WBobby mentioned this pull request Aug 17, 2022

Add ROCm5.2.3/AMDGPU support for PyTorch WBobby/pytorch#2

Closed

Passing explicit pretrained_backbone #74372

Passing explicit pretrained_backbone #74372

Uh oh!

Conversation

datumbox commented Mar 17, 2022

Uh oh!

pytorch-bot bot commented Mar 17, 2022

⚛️ CI Flow

Uh oh!

facebook-github-bot commented Mar 17, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful links

💊 CI failures summary and remediations

1 failure not recognized by patterns:

Uh oh!

facebook-github-bot commented Mar 17, 2022

Uh oh!

facebook-github-bot commented Mar 18, 2022

Uh oh!

datumbox commented Mar 19, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Mar 21, 2022

Uh oh!

datumbox commented Mar 21, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

datumbox commented Mar 22, 2022

Uh oh!

BowenBao commented Mar 22, 2022

Uh oh!

datumbox commented Mar 22, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

BowenBao left a comment

Choose a reason for hiding this comment

Uh oh!

github-actions bot commented Mar 22, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

facebook-github-bot commented Mar 17, 2022 •

edited

Loading

datumbox commented Mar 19, 2022 •

edited

Loading

datumbox commented Mar 21, 2022 •

edited

Loading

datumbox commented Mar 22, 2022 •

edited

Loading