- Medium: It contributes to significant difficulty to complete my task, but I can work around it.
Howdy! I’m attempting to run a set of tests that relies on repeated rollouts. Unfortunately I’m experiencing an error on the second run, on
ValueError: Attempting to start raylet with 1 GPUs, but CUDA_VISIBLE_DEVICES contains .
Everything runs correctly the first time, and I do not expect to use a GPU for any of these rollouts.
The problem line of code is specifically:
ray.init( num_cpus=1, num_gpus=1, local_mide=True, log_to_driver=True)
Anyone know what is causing this?