“Bus Error and Resource Tracker Warning When Training PyTorch Model on GPU with MPS”

[pratheeshkumar99]

I am encountering issues while trying to train it on an Apple Mac M3 with a 12-core CPU and an 18-core GPU (18GB RAM) environment. Below are the details and issues I’m facing:

1.Device Selection Issue:

device = "cuda" if torch.cuda.is_available() else "mps" if torch.has_mps or torch.backends.mps.is_available() else "cpu" print("Using device:", device)

Although MPS is detected and selected, training crashes immediately after starting, with the following error:

/opt/anaconda3/envs/mynewenv/lib/python3.9/multiprocessing/resource_tracker.py:216: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown warnings.warn('resource_tracker: There appear to be %d ').

2.CPU Training:

When I switch to CPU training on the same machine, it runs without any issues using the same batch size of 8.

3.Google Colab Training:

There are no issues when running the same code on Google Colab.

I’m looking for insights into what might be causing these issues on MPS and how I could resolve them. Specifically, I’d like to understand the semaphore leak and bus error that seems to occur only when using MPS.

[yiliyili96]

Same for me.