Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Dynamic shape support needed for flatten using ints in tosa dialect #3004

Open
Abhishek-TyRnT opened this issue Mar 10, 2024 · 0 comments
Open

Dynamic shape support needed for flatten using ints in tosa dialect #3004

Abhishek-TyRnT opened this issue Mar 10, 2024 · 0 comments

Comments

@Abhishek-TyRnT
Copy link
Contributor

flattenusingints op currently only supports static shapes.

torch-mlir/lib/Conversion/TorchToTosa/TorchToTosa.cpp

Lines 2488 to 2491 in a3fe130

if (!selfType || !selfType.hasStaticShape())
return rewriter.notifyMatchFailure(
op,
"Only ranked tensor types with static shapes are currently supported");

Due to this tests like argmax fail. Sending a PR for fix
@Abhishek-TyRnT Abhishek-TyRnT mentioned this issue Mar 10, 2024
Merged
ramiro050 pushed a commit that referenced this issue Mar 19, 2024
@Abhishek-TyRnT
Dynamic size support for flatten (#3005)
df02692 
Added support for dynamic shapes in `flattenusingints` op in tosa
dialect. Due to this some Argmax tests pass
This PR fixes this issue #3004

The following tests pass after this PR
 ```
1. "ArgmaxIntModule_basic"
2. "ArgmaxIntModule_multiple_maxs"
3. "ArgmaxModule_basic"
```
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@Abhishek-TyRnT