Add command line flag to specify device for inference #25
+23
−18
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
There are some merge conflicts now. Before I attempt to resolve them, please let me know if this PR is something you're interested in. I'm happy to make any amendments, but if it's an unwanted change I'm also happy to abandon the effort. Please let me know either way. |
yoavhacohen
requested changes
Dec 20, 2024
There was a problem hiding this comment.
Thank you for the contribution! This PR looks excellent, and we’re excited to support the --device argument for CUDA, TPU, MPS, and CPU.
Could you please rebase it on the latest main branch?
If you’re able to add TPU support via PyTorch XLA, that would be fantastic. Otherwise, we can include it in a follow-up PR.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds a new command line flag
--devicethat allows users to specify the device (e.g.cuda,cpu, ormps) to use for inference. The flag defaults tocudabut will automatically fall back tocpuif CUDA is not available. This change also updates the code to use the specified device for loading models and running the pipeline.torch.manual_seedsets the seed for generating random numbers on all devices. I removed the redundant call totorch.cuda.manual_seed.This code was generated while working on pytorch/pytorch#141471. Note that LTX-Video does not currently work when using MPS as the backend in latest PyTorch. This is a regression as it worked in PyTorch v2.4.1. Please follow pytorch/pytorch#141471 if you'd like updates on progress towards a fix.