huggingface / huggingface/evaluate
Evaluation took so much disk
Open
- Dominant language
- Python
- Stars
- 2.5k
- Forks
- 341
- PR merge metrics
- No merged PRs in 30d
Description
I am try to train SegFormer model and use evaluate module to eval the IoU, but when there is so many data and epoch, the metrics looks like take so much space.
is there any method so I can free space while training still going?
Contributor guide
Assessment
This issue has not been assessed yet.