Xcrid

multi-gpu use

1

Hi, I just wanted to know if this implementation would work well in a DataParrallel model and if there was some adaptation to do ? Thanks

Hi, I was wondering if it was possible you had a license to your project to make it's use easier.

Pre-training

2

Hi, When you pre-train your model with the source dataset, do you also pre-train the aux_output by using the aux_loss or do you only use the main loss ?

aux_loss

2

Hi, You use two losses for your training, yet all your decoders return the main output only, the aux one being None. Which one should I add ?