Skip to content

Conversation

@mrwyattii
Copy link
Contributor

@mrwyattii mrwyattii commented Aug 2, 2022

Re-enabling the AMD CI runner, with small updates:

  • AMD CI runner now installs its own pytorch / ROCm (previously, using the system-installed)
  • Torch version being tested is now 1.12 (previously 1.10)
  • Unit tests being run are restricted to those that were refactored with updated DistributedTest in Fix for distributed tests on pytorch>=1.12 #2141
  • Remove the use of pytest-forked due to a CUDA init error (this can be revisited if we find --forked to be necessary)

@mrwyattii mrwyattii changed the title use torch-provided rocm in AMD runner Update for AMD CI workflow Aug 2, 2022
@mrwyattii mrwyattii merged commit d1cd18e into deepspeedai:master Aug 3, 2022
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