facebookresearch / facebookresearch/fvcore
Quantized model size
Open
- Dominant language
- Python
- Stars
- 2.3k
- Forks
- 236
- PR merge metrics
- No merged PRs in 30d
Description
Hi
Can we use fvcore to compute the quantized model size in pytorch for :
W8A8
W4A4
W2A2
If yes, how?
Thanks
Thanks
Contributor guide
Assessment
This issue has not been assessed yet.