facebookresearch / facebookresearch/SlowFast
About training setting of released X3D models
Open
- Dominant language
- Python
- Stars
- 7.4k
- Forks
- 1.3k
- PR merge metrics
- No merged PRs in 30d
Description
In the original paper, X3D models are trained with 128 GPUs, while the released configs are for 8 GPUs. Are those released checkpoints trained with 8GPUs or 128GPUs?
Contributor guide
Assessment
This issue has not been assessed yet.