Update: It turned out that I just forgot the --multiprocessing-distributed flag, which is critical for multi-GPU instance. See Bizarre InternalTorchDynamoError with locally and formerly working code - #5 by ptrblck - PyTorch Forums for details.
Update: It turned out that I just forgot the --multiprocessing-distributed flag, which is critical for multi-GPU instance. See Bizarre InternalTorchDynamoError with locally and formerly working code - #5 by ptrblck - PyTorch Forums for details.