[fix] vec * mat in matmul in onnx converter #11174
Merged
Conversation
This file contains 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
masahi
reviewed
Apr 28, 2022
masahi
approved these changes
Apr 28, 2022
|
@masahi CI finally passed! I think we can also close the issue after the merge. :-) |
shtinsa
pushed a commit
to Deelvin/tvm
that referenced
this pull request
May 17, 2022
* fix: vec * mat in matmul in onnx converter * fix: pylint * fix: vec-mat matmul * fix test * fix test
SebastianBoblest
pushed a commit
to SebastianBoblest/tvm
that referenced
this pull request
May 27, 2022
* fix: vec * mat in matmul in onnx converter * fix: pylint * fix: vec-mat matmul * fix test * fix test
juda
pushed a commit
to juda/tvm
that referenced
this pull request
Jun 21, 2022
* fix: vec * mat in matmul in onnx converter * fix: pylint * fix: vec-mat matmul * fix test * fix test
junrushao
pushed a commit
that referenced
this pull request
Nov 21, 2022
…n.matmul check (#13448) This PR brings 2 bug fixes: 1. ONNX converter for matmul: ONNX matmul follows NumPy [rules](https://numpy.org/doc/stable/reference/generated/numpy.matmul.html): > If the first argument is 1-D, it is promoted to a matrix by prepending a 1 to its dimensions. After matrix multiplication the prepended 1 is removed. > If the second argument is 1-D, it is promoted to a matrix by appending a 1 to its dimensions. After matrix multiplication the appended 1 is removed. The (my) previous fix #11174 did not consider the second rule (append 1 dimension for the rhs vector). 2. Relay's `nn.matmul` takes 2-D matrices and the checker was removed in a recent PR #13287. This PR puts the checker back to prevent process crashes (make it a readable TVMError) for readability (and also for that the CI in ise-uiuc/nnsmith#64 won't be terminated while using TVM-10).
xinetzone
pushed a commit
to daobook/tvm
that referenced
this pull request
Nov 25, 2022
…n.matmul check (apache#13448) This PR brings 2 bug fixes: 1. ONNX converter for matmul: ONNX matmul follows NumPy [rules](https://numpy.org/doc/stable/reference/generated/numpy.matmul.html): > If the first argument is 1-D, it is promoted to a matrix by prepending a 1 to its dimensions. After matrix multiplication the prepended 1 is removed. > If the second argument is 1-D, it is promoted to a matrix by appending a 1 to its dimensions. After matrix multiplication the appended 1 is removed. The (my) previous fix apache#11174 did not consider the second rule (append 1 dimension for the rhs vector). 2. Relay's `nn.matmul` takes 2-D matrices and the checker was removed in a recent PR apache#13287. This PR puts the checker back to prevent process crashes (make it a readable TVMError) for readability (and also for that the CI in ise-uiuc/nnsmith#64 won't be terminated while using TVM-10).
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.
Should fix #10651 for ONNX converter.
cc: @masahi