Add more model symbolic tracing tests from torchvision #55744
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
💊 CI failures summary and remediationsAs of commit 9ff2c82 (more details on the Dr. CI page):
This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions to the (internal) Dr. CI Users group. |
jamesr66a
approved these changes
Apr 14, 2021
test/test_fx.py
Outdated
| @@ -41,7 +41,7 @@ | |||
| from fx.named_tup import MyNamedTup | |||
|
|
|||
| try: | |||
| from torchvision.models import resnet18 | |||
| from torchvision import models | |||
There was a problem hiding this comment.
nit: can we name this something more unique like torchvision_models?
jamesr66a
reviewed
Apr 14, 2021
jamesr66a
reviewed
Apr 14, 2021
test/test_fx.py
Outdated
| qgraph_script = torch.jit.script(qgraph) | ||
| d = out_transform(qgraph(x)) | ||
| e = out_transform(qgraph_script(x)) | ||
| torch.testing.assert_allclose(a, d) |
There was a problem hiding this comment.
I think you're going to have to increase the tolerance (atol or rtol kwargs) if you're comparing between an unquantized and a quantized model. It might be a better idea to just remove the quantization tests, that stuff should be covered in test_quantize.py anyway
|
@suraj813 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
krshrimali
pushed a commit
to krshrimali/pytorch
that referenced
this pull request
May 19, 2021
Summary: Fixes pytorch#55398 Generates tests that calls `symbolic_trace` on torchvision models and verifies the parity of outputs from eager model, `fx.GraphModule`, `jit.ScriptModule`. Test errors: GoogleNet and Inception models throw a type mismatch when scripting the traced `fx.GraphModule`. ``` Return value was annotated as having type __torch__.torchvision.models.googlenet.GoogLeNetOutputs but is actually of type Tensor: dropout = self.dropout(flatten); flatten = None fc = self.fc(dropout); dropout = None return fc ~~~~~~~~~ <--- HERE ``` Relevant type-inconsistency https://github.com/pytorch/vision/blob/512ea299d4b2d2bbac3498a75a2d8c0190cfcb39/torchvision/models/googlenet.py#L200 ``` torch.jit.unused def eager_outputs(self, x: Tensor, aux2: Tensor, aux1: Optional[Tensor]) -> GoogLeNetOutputs: if self.training and self.aux_logits: return _GoogLeNetOutputs(x, aux2, aux1) else: return x # type: ignore[return-value] ``` Pull Request resolved: pytorch#55744 Reviewed By: albanD Differential Revision: D27920595 Pulled By: suraj813 fbshipit-source-id: 01f6f2aef7badbde29b5162a7787b5af9398090d
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #55398
Generates tests that calls
symbolic_traceon torchvision models and verifies the parity of outputs from eager model,fx.GraphModule,jit.ScriptModule.Test errors: GoogleNet and Inception models throw a type mismatch when scripting the traced
fx.GraphModule.Relevant type-inconsistency https://github.com/pytorch/vision/blob/512ea299d4b2d2bbac3498a75a2d8c0190cfcb39/torchvision/models/googlenet.py#L200