deepinsight / deepinsight/insightface
Why do you set 'requires_grad=False' in the 'features' layer of iresnet?
Open
- Dominant language
- Python
- Stars
- 29.7k
- Forks
- 6.1k
- PR merge metrics
- No merged PRs in 30d
Description
'/insightface/recognition/arcface_torch/backbones/iresnet.py':
line 98: self.features = nn.BatchNorm1d(num_features, eps=1e-05)
line 99: nn.init.constant_(self.features.weight, 1.0)
line 100: self.features.weight.requires_grad = False
--------------------------------------------------------------------------------------------------------
What was your purpose of setting 'self.features.weight.requires_grad' as 'False'? To avoid some problem of optimizing? Or others?
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.