Pages

Categories

Some important updates

8/15/24, by artyom ; 0 comments

Lots of passed since latest updates... So

  1. Now it works with pytorch 2.4 - in fact it is requirement. Either 1.13 or >=2.4
  2. I created a much easier interface to use - all you need is to import pytorch_ocl module and you'll get all the goodies on Linux and Windows.
  3. With python module you can use torch.ocl.synchonize() and torch.ocl.empty_cache() as with CUDA
  4. I ordered Intel Arc GPU (a380) - so hopefully I'm going to be able to check/optimise for a new platform
  5. Implemented other things needed like manual_seed_all - as required for backed.

Known issues: Currently there is an issue with loading back saved state dictionary if it was saved from ocl device. It crashes for some reason (outside of ocl backend).

Workaround: When you save/restore the model move it to CPU and than back to ocl device.

Add Comment:

 
 the email would not displayed
 

You can write your messages using Markdown syntax.

You must enable JavaScript in order to post comments.