Dynamic Shape Simple NN Model Test #4074
Open
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
vanbasten23
reviewed
Oct 6, 2022
| import torch_xla.core.xla_model as xm | ||
| import numpy | ||
|
|
||
| pd = torch._C._EnablePythonDispatcher() |
There was a problem hiding this comment.
I wonder what this line does
There was a problem hiding this comment.
it enables us to run python implementations of CompositeAutogradImplicit ops.
CompositeAutogradImplicit means we don't have an explicit backward formula for an op instead an op is composed of a bunch of ops that do have backward formulas and combines this formulas is equivalent to differentiating the op explicitly.
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Outdated suggestions cannot be applied.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.
Testing dynamic shape functionality using a simple model. This PR is meant to identify the gaps in dynamic shape op support. We will keep track of the bugs we address as part of the investigation in this PR.
Blockers
tensor.shapeis not yet implemented for dynamic tensors #4070torch.nn.functional#4075