Skip to content

ZJULearning/AdaDB

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AdaDB: An Adaptive Gradient Method with Data-Dependent Bound

Requirements

  • Python 3.6.8
  • PyTorch 1.2.0
  • CUDA Version 10.0.130
  • GPU: GTX 1080Ti

Usage

You can use AdaDB like any other PyTorch optimizers.

optimizer = optimizers.AdaDB(model.parameters(), lr=1e-3, final_lr=0.1)

Run Demos on CIFAR-100

Train model ResNet-18 on CIFAR-100 dataset using optimizer AdaDB.

cd AdaDB
python cifar.py --model=resnet18 --optim=adadb --lr=1e-3 --final_lr=0.1 --gamma=1e-5

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages