Training on multiple GPUs in parallel

[vnik18]

Hello, I am able to run your code, however I do not have access to a TPU.

Do you know if it is possible to perform training of the masker-corrector model on multiple GPUs instead of a single one? Thank you!

[j6mes]

Hi, while i did all my experiments on a single GPU. The code is built on huggingface and should support distributed data parallel training automatically detecting multiple GPUs or by manually setting the --num_gpus X flag

[vnik18]

Thank you, I will try that