Skip to content

Conversation

AshNicolus
Copy link

Remove TensorFlow dependency from multimodal/image.py

Replace tf.image decoding/resizing with Pillow + NumPy + JAX.

  • Convert float inputs (0..1 or 0..255) to uint8 for processing.
  • Resize with PIL.BILINEAR, convert back to JAX array.
  • Normalize and clip outputs to [-1, 1].
  • Resolve TODO(eyvinec) to remove TF dependency.

AshNicolus and others added 2 commits October 17, 2025 22:31
Replace tf.image decoding/resizing with Pillow + NumPy + JAX.
- Convert float inputs (0..1 or 0..255) to uint8 for processing.
- Resize with PIL.BILINEAR, convert back to JAX array.
- Normalize and clip outputs to [-1, 1].
- Resolve TODO(eyvinec) to remove TF dependency.
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.

1 participant