pytorch-optimizer v1.3.2

Change Log

• torch.hub usage in docs, #76
• Adan optimizer, #77
  • fix: forgot to divide into beta_correction
  • feat: support weight_decouple

pytorch-optimizer v1.3.1

Change Log

• raw directive in RST format cannot be used due to the security issue e.g. code injection. #75

pytorch-optimizer v1.3.0

Change Log

• Support torch.hub.load to load the pytorch_optimizer, #73

Contributions

thanks to

@Bing-su

pytorch-optimizer v1.2.0

Change Log

• Add a new optimizer, Adan optimizer. #69
  • Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models

pytorch-optimizer v1.1.4

Just minor changes on the dependencies.

Change Log

• required torch version to >=1.10 (just for CI/CD), PR68
  • still supports all versions of the torch (1.x).

pytorch-optimizer v1.1.3

build(deps): numpy

pytorch-optimizer v1.1.2

Change Log

• Fix Ranger21 parameters, not to assign a default value None to num_iterations parameter. Issue

pytorch-optimizer v1.1.1

update: load_optimizer

pytorch-optimizer v1.1.0

update: remove adahessian optimizer

pytorch-optimizer v1.0.0

update: version 1.0.0