Skip to content

Conversation

@robin-ede
Copy link
Contributor

Summary

Fixed 3 typos in README.md

Changes

  • Line 140: weighgt -> weight
  • Line 304: hyrid -> hybrid
  • Line 347: dytpe -> dtype

Context

Line 140: Adan optimizer caution, and Lamb decoupled weighgt decay options...
Line 304: 3 more MobileNetV4 hyrid weights with different MQA weight init scheme...
Line 347: Add flag for transforms, return non-normalized torch.Tensor with original dytpe...

@rwightman rwightman merged commit 59fdbaa into huggingface:main Jul 9, 2025
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.

2 participants