[ray] fix: use get_device_name() for automatic device detection in RayWorkerGroup instead of by parameter passing

[jianjunzhong]

…ad of by parameter passing

What does this PR do?

Use get_device_name() for automatic device detection in RayWorkerGroup instead of by parameter passing

Checklist Before Starting

• Search for similar PRs. Paste at least one query link here: ...
• Format the PR title as [{modules}] {type}: {description} (This will be checked by the CI)
  • {modules} include fsdp, megatron, veomni, sglang, vllm, rollout, trainer, ci, training_utils, recipe, hardware, deployment, ray, worker, single_controller, misc, perf, model, algo, env, tool, ckpt, doc, data, cfg, reward
  • If this PR involves multiple modules, separate them with , like [megatron, fsdp, doc]
  • {type} is in feat, fix, refactor, chore, test
  • If this PR breaks any API (CLI arguments, config, function signature, etc.), add [BREAKING] to the beginning of the title.
  • Example: [BREAKING][fsdp, megatron] feat: dynamic batching

Test

For changes that can not be tested by CI (e.g., algorithm implementation, new model support), validate by experiment(s) and show results like training curve plots, evaluation results, etc.

API and Usage Example

Demonstrate how the API changes if any, and provide usage example(s) if possible.

# Add code snippet or script demonstrating how to use this

Design & Code Changes

Demonstrate the high-level design if this PR is complex, and list the specific changes.

Checklist Before Submitting

Important

Please check all the following items before requesting a review, otherwise the reviewer might deprioritize this PR for review.

• Read the Contribute Guide.
• Apply pre-commit checks: pre-commit install && pre-commit run --all-files --show-diff-on-failure --color=always
• Add / Update the documentation.
• Add unit or end-to-end test(s) to the CI workflow to cover all the code. If not feasible, explain why: ...
• Once your PR is ready for CI, send a message in the ci-request channel in the verl Slack workspace. (If not accessible, please try the Feishu group (飞书群).)
• If your PR is related to the recipe submodule, please also update the reference to the submodule commit via git submodule update --remote or cd recipe && git pull origin main.

[gemini-code-assist]

Code Review

This pull request aims to improve automatic device detection in RayWorkerGroup by using get_device_name(). While this is a good improvement for the default behavior, the current implementation removes the ability to manually override the device via the device_name parameter, which is a breaking change and a loss of flexibility. My review includes a suggestion to retain this capability while still using automatic detection as the default.

[gemini-code-assist]

The change from kwargs.get("device_name", "cuda") to get_device_name() improves the default device detection. However, it removes the ability for users to override the device by passing the device_name parameter to the RayWorkerGroup constructor. This is a potentially breaking change and a loss of flexibility. For instance, a user might want to run a CPU-only job on a machine with GPUs for testing purposes.

Several tests are also passing device_name as a keyword argument (e.g., in tests/single_controller/test_worker_group_basics.py), which will be ignored with this change.

To maintain flexibility while improving the default, I suggest using get_device_name() as the default value, but still allowing it to be overridden.

Suggested change

	self.device_name = get_device_name()
	self.device_name = kwargs.get("device_name", get_device_name())