Training on Multiple GPUs - IterableDataset

[alamayreh]

Folks, I have encountered an issue while attempting to run the code on multiple GPUs. I increased the number of GPUs in the Config file and used the distributed backend (DP), but then I encountered errors along the way.

It seems that MsgPackIterableDatasetMultiTargetWithDynLabels(IterableDataset) can not handle multiple GPUs.

If anyone has successfully run the system on multiple GPUs, I would greatly appreciate any guidance or insights you could provide.

Thank you for your attention and assistance.

Best regards