facebookresearch / facebookresearch/moco
why pretrain from encoder_q?
Open
- Dominant language
- No language data
- Stars
- 5.1k
- Forks
- 802
- PR merge metrics
- No merged PRs in 30d
Description
Yes,i have a question is that：in liner fine turning period，why pretrain from encoder_q instead of encoder_k？
Contributor guide
Assessment
This issue has not been assessed yet.