facebookresearch/fairscale
[chore] Cache wikitext2 dataset used for benchmarks/pipe.py to prevent throttling errors
Aperta
#743 aperta il 19 lug 2021
good first issue
Metriche repository
- Star
- (3411 stelle)
- Metriche merge PR
- (Metriche PR in attesa)
Descrizione
Chore
We should cache the dataset used for running fairscale/benchmarks/pipe.py so that we don't end up downloading this at every run. This has been done for OSS and we should do something similar for pipe.py. This is important because during periods of high PR checkins/development we could run into throttling errors if there are too many CircleCI jobs. That is something that possibly happened for OSS which is when we added the caching logic.