allenai / longformer

Longformer: The Long-Document Transformer
https://arxiv.org/abs/2004.05150
Apache License 2.0
2.05k stars 276 forks source link

LED Training Time #242

Open gospelnnadi opened 2 years ago

gospelnnadi commented 2 years ago

Hello,

i have search through the paper and in the issues, i couldn't find the time and GPU's used for to train LED. Please, how long it took to pre-train LED-base and LED-large from BART checkpoints. How long it took to fine tune them on arXiv and Pubmed.