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

fixes ImageNet, SimpleRNN examples #499

Closed
wants to merge 6 commits into from
Closed

Conversation

DhruvDh
Copy link
Contributor

@DhruvDh DhruvDh commented Feb 24, 2024

Fixes #498

Running the resnet model per the example from the ImageNet README requires adding Flux and Metalhead per the Blotz README. I added those and replicated the unsafe_free! for AMDGPU.

I am new to Julia so sorry if something isn't quite right.

Also added the fix to #498 mentioned in the conversation.

- QOL improvements for AMDGPU
Copy link
Member

@avik-pal avik-pal left a comment

Choose a reason for hiding this comment

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

Remove the AMDGPU.allowscalar and we should be good.

examples/GravitationalWaveForm/main.jl Outdated Show resolved Hide resolved
examples/Basics/main.jl Outdated Show resolved Hide resolved
examples/ImageNet/main.jl Outdated Show resolved Hide resolved
Copy link

codecov bot commented Feb 24, 2024

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 87.53%. Comparing base (043de1b) to head (3988840).

Additional details and impacted files
@@            Coverage Diff             @@
##             main     #499      +/-   ##
==========================================
+ Coverage   87.40%   87.53%   +0.12%     
==========================================
  Files          30       30              
  Lines        1643     1644       +1     
==========================================
+ Hits         1436     1439       +3     
+ Misses        207      205       -2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

- use 32-bit float in SimpleRNN for Metal compatibility
@DhruvDh DhruvDh changed the title fixes ImageNet Example fixes ImageNet, SimpleRNN examples Feb 25, 2024
@avik-pal
Copy link
Member

Our GPU CIs are down atm and I don't think are going to be online before the next week, so we will have to wait till then to merge

This reverts commit 9740ade.
examples/ImageNet/main.jl Outdated Show resolved Hide resolved
examples/Basics/main.jl Outdated Show resolved Hide resolved
@avik-pal
Copy link
Member

Thanks for the fixes, merged it via the other PR

@avik-pal avik-pal closed this Feb 27, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

SimpleRNN example is broken on AMDGPU
2 participants