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.

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

Squeeze last dimensions if possible for 5D (or bigger) reductions #408

Merged
merged 3 commits into from
Apr 3, 2023

Conversation

DenisVieriu97
Copy link
Collaborator

@DenisVieriu97 DenisVieriu97 commented Apr 3, 2023

Summary of changes:

  • Reduction ops optimization - squeeze all dimensions after 4th dim if they are all 1
  • Disable type inference only for 1D unary ops

Copy link

@razarmehr razarmehr left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good.

@kulinseth kulinseth merged commit 3353590 into master Apr 3, 2023
skotapati pushed a commit that referenced this pull request Apr 6, 2023
* Squeeze last dimensions if possible for 5D (or bigger) reductions

* Remove unnecessary code

* Disable type infernce only for 1D tensors in unary
skotapati pushed a commit that referenced this pull request Apr 7, 2023
* Squeeze last dimensions if possible for 5D (or bigger) reductions

* Remove unnecessary code

* Disable type infernce only for 1D tensors in unary
skotapati pushed a commit that referenced this pull request Apr 11, 2023
* Squeeze last dimensions if possible for 5D (or bigger) reductions

* Remove unnecessary code

* Disable type infernce only for 1D tensors in unary
DenisVieriu97 added a commit that referenced this pull request Apr 24, 2023
* Squeeze last dimensions if possible for 5D (or bigger) reductions

* Remove unnecessary code

* Disable type infernce only for 1D tensors in unary
@DenisVieriu97 DenisVieriu97 added Upstreamed Change has been upstreamed to PyTorch master and removed DV: In Progress labels Jun 16, 2023
jhavukainen pushed a commit that referenced this pull request Mar 15, 2024
* Squeeze last dimensions if possible for 5D (or bigger) reductions

* Remove unnecessary code

* Disable type infernce only for 1D tensors in unary
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Upstreamed Change has been upstreamed to PyTorch master
Projects
None yet
3 participants