Skip to content

Latest commit

 

History

History
30 lines (21 loc) · 577 Bytes

README.md

File metadata and controls

30 lines (21 loc) · 577 Bytes

Adaptive-Gradient-Clipping

Needs more testing

Yannic Kilchers Video

Usage

from agc import AGC

optimizer.zero_grad()        
loss, output = model(data)
loss.backward()

AGC(model.parameters(), args.clip)

optimizer.step()

Citations

@article{brock2021high,
  author={Andrew Brock and Soham De and Samuel L. Smith and Karen Simonyan},
  title={High-Performance Large-Scale Image Recognition Without Normalization},
  journal={arXiv preprint arXiv:},
  year={2021}
}