vision
vision copied to clipboard
pytorch
RandomOrder transform now accepts torch.Tensor and works well with torch.seed
Changed the RandomOrder transform to use torch.randperm and thus make it Scriptable
:link: Helpful Links
:test_tube: See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/7773
- :page_facing_up: Preview Python docs built from this PR
Note: Links to docs will display an error until the docs builds have been completed.
This comment was automatically generated by Dr. CI and updates every 15 minutes.
![pytorch-bot[bot] avatar](https://avatars.githubusercontent.com/in/40112?v=4 "Published by a pub.dev verified publisher"){kind=small}
This PR also lifts the restriction of RandomOrder to take only PIL.Image as input.
https://pytorch.org/vision/0.8/transforms.html#transforms-on-pil-image-only
JIT Scripting still doesn't work. Gives the following error RuntimeError: Expected integer literal for index, whilw JIT Scripting
It is due to https://github.com/pytorch/pytorch/issues/16123
But, other benefits of this PR:
- Removes restriction of inputs of RandomChoice of PIL.Image to also include torch.Tensors
- A step closer to make it JIT Scriptable
- Replaces the usage of the random module to native torch, thereby working well with torch seed