[Relay] [Training] Add numerical gradient check. #3630
Merged
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
|
It looks good. Though it has been years that I didn’t check correctness via numerically...why not just compare with PyTorch? |
junrushao
approved these changes
Jul 26, 2019
|
@junrushao1994 the add_grads pr was originally in comparison of PyTorch, but after discussing with @tqchen we agree that it is bad to depend on pytorch (which might also has errors in gradient) for correctness. |
vinx13
approved these changes
Jul 26, 2019
|
Numeral stability would cause trouble when it comes together with randomized inputs, which results in unnecessary flaky tests. Just be careful then :-) |
wweic
pushed a commit
to wweic/tvm
that referenced
this pull request
Aug 9, 2019
* add check_grad * finish * what does the fox say? * lint lint lint lint lint lint lint lint lint
wweic
pushed a commit
to neo-ai/tvm
that referenced
this pull request
Sep 6, 2019
* add check_grad * finish * what does the fox say? * lint lint lint lint lint lint lint lint lint
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
Thanks for contributing to TVM! Please refer to guideline https://docs.tvm.ai/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews from Reviewers.
@vinx13 @junrushao1994 @jroesch @tqchen @srkreddy1238 can you guys give some review?