Coerce bool to int in unpack #287
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
While trying to use torch.compile with `backend=inductor` for a model that uses unpack as
```
unpack(x, ps, "b d *")[0]
```
I received the error
```
torch._dynamo.exc.TorchRuntimeError: Failed running call_function <function unpack at 0x7f5c30806200>(*(FakeTensor(..., device='cuda:0', size=(1, 128, (s1//2))), [((s1//2),)], 'b d *'), **{}):
unsupported operand type(s) for +: 'int' and 'SymBool'
```
tracing back to this line. Coercing the boolean to an int so `sum` does not try to sum a boolean and an int fixed my issue locally.
|
Did you report this to torchdynamo team? Using boolean in arithmetics is perfectly legal in python |
|
reported to torch as well |
|
Thanks for reporting it to pytorch. The issue came up specifically when using I will add that information to the issue you reported and thank you for merging the PR. |
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Outdated suggestions cannot be applied.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.
While trying to use torch.compile with
backend=inductorfor a model that uses unpack asI received the error
tracing back to the line changed in this PR. Coercing the boolean to an int so
sumdoes not try to sum a boolean and an int fixed my issue locally.