Rumi4
Results
3
comments of
Rumi4
I tested the model on GigaWord dataset and it seems to copy the original article rather than giving an abstractive summary closer to the gold summary. Are there any changes...
Hi! Just type 'optims' in checkpoint['optim'] it works.
while training on 1 gpu we need to set the grad accum count greater than 5, u said. how much should that be? Please help. > With only 1 gpu...