[<Ray component: Core|RLlib|etc...>] ray train

[Hxinyue]

What happened + What you expected to happen

When ray train is performed, the train_loop_per_worker function has a return value. How can I get the return value of the train_loop_per_worker function after running the trainer.fit()

Versions / Dependencies

ray==2.11.0
python==3.10

Reproduction script

def main(config):
alg_id = config["alg_id"]
return {"success": 1}

trainer = TorchTrainer(
main,
scaling_config=ScalingConfig(use_gpu=use_gpu, num_workers=num_workers,
resources_per_worker={"CPU": cpu, "GPU": gpu}),
train_loop_config=config,
...
)
trainer.fit()

Issue Severity

High: It blocks me from completing my task.