Skip to content

Latest commit

 

History

History
11 lines (8 loc) · 624 Bytes

README.md

File metadata and controls

11 lines (8 loc) · 624 Bytes
Raw

LAMP

This is the official implementation of the paper: "Layerwise Sparsity for Magnitude-based Pruning", ICLR 2021.

  • The key file is the tools/pruners.py, where we implement various layerwise sparsity determination methods for the magnitude-based pruning.
  • Run iterate.py to run simulations.
  • This codebased only contains CIFAR-10 experiments in the paper. To add more experiments, you may want to add models to tools/models/, datasets to tools/datasets/ and tools/dataloaders.py, hyperparameter setups to tools/modelloaders.py.

Enjoy!
Best,
Authors.