Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[benchmarks] Fix moco initialization device. #7598

Merged
merged 1 commit into from
Jul 2, 2024
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion benchmarks/torchbench_model.py
Original file line number Diff line number Diff line change
Expand Up @@ -344,7 +344,8 @@ def should_initialize_on_xla(self):
# 1. Models don't expect 'tpu' as their device.
# 2. 'moco' initializes a ProcessGroup, i.e. the backend depends on
# the given device
return self.is_accelerator_tpu() or self.model_name == "moco"
return self.is_accelerator_tpu() or (self.model_name == "moco" and
self.benchmark_experiment.xla)

def is_inference(self):
return self.benchmark_experiment.test == "eval"
Expand Down
Loading