stft: move towards always returning complex #72882
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
For `stft` this makes all cases where `return_complex` default to `False` into an error and adds a warning when `return_complex=False` is passed explicitly. For `istft` this raises an error if the input is not a complex tensor. [ghstack-poisoned]
CI Flow Status⚛️ CI FlowRuleset - Version:
|
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 4e08615 (more details on the Dr. CI page):
🕵️ 12 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages
|
For `stft` this makes all cases where `return_complex` default to `False` into an error and adds a warning when `return_complex=False` is passed explicitly. For `istft` this raises an error if the input is not a complex tensor. [ghstack-poisoned]
For `stft` this makes all cases where `return_complex` default to `False` into an error and adds a warning when `return_complex=False` is passed explicitly. For `istft` this raises an error if the input is not a complex tensor. [ghstack-poisoned]
peterbell10
commented
Feb 17, 2022
| length=length, return_complex=return_complex) | ||
| )SCRIPT"}, | ||
| {"stft_0_10", R"SCRIPT( | ||
| def stft_0_10(self: Tensor, n_fft: int, hop_length: Optional[int]=None, win_length: Optional[int]=None, |
There was a problem hiding this comment.
I'm getting a major issue with these upgraders. stft and istft have python definitions in torch/functional.py that wrap the ATen operators. So the TorchScript models use prim::CallFunction instead of aten::stft and the upgrader is never called.
There was a problem hiding this comment.
Is there a reason why these functions have to be defined in python? can we move it to C++? It seems like they just handle torch function or call into torch.{stft/istft} otherwise.
There was a problem hiding this comment.
@tugsbayasgalan should know what to do
There was a problem hiding this comment.
We currently iterate over all nodes in the graph and replace the aten operators with upgraders when necessary. I am wondering if we can be little smarter in that pass that would dig deeper when we encounter prim::CallFunction. I am not sure how feasible that is @eellison @gmagogsfm. Alternatively, fixing the bug @anjali411 mentioned to unblock the moving to C++ might be the way to go. What do you guys think?
…lex" For `stft` this makes all cases where `return_complex` default to `False` into an error and adds a warning when `return_complex=False` is passed explicitly. For `istft` this raises an error if the input is not a complex tensor. [ghstack-poisoned]
|
@peterbell10 please let me know when this PR is ready for review |
|
Sure thing. The actual code shouldn't need changed much, but I'm still waiting on #73434 which is blocked by an FC period that ends this week. |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
|
/easycla As part of the transition to the PyTorch Foundation, this project now requires contributions be covered under the new CLA. See #85559 for additional details. This comment will trigger a new check of this PR. If you are already covered, you will simply see a new "EasyCLA" check that passes. If you are not covered, a bot will leave a new comment with a link to sign. |
|
|
@peterbell10 we should merge this PR right? |
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.
Stack from ghstack (oldest at bottom):
For
stftthis makes all cases wherereturn_complexdefault toFalseinto an error and adds a warning whenreturn_complex=Falseis passed explicitly.
For
istftthis raises an error if the input is not a complex tensor.