facebookresearch / facebookresearch/moco
maybe wrong optimizer
Open
- Dominant language
- No language data
- Stars
- 5.1k
- Forks
- 802
- PR merge metrics
- No merged PRs in 30d
Description
**https://github.com/facebookresearch/moco/blob/colab-notebook/colab/moco_cifar10_demo.ipynb**
_def train(net, data_loader, train_optimizer, epoch, args):
net.train()
adjust_learning_rate(optimizer, epoch, args)_
adjust_learning_rate is called inside train method and therefore it should make use of the same train_optimizer.
Could this training in the notebook be with wrong results?
Contributor guide
Assessment
This issue has not been assessed yet.