Skip to content

Code for the paper "Understanding the Role of Momentum in Stochastic Gradient Methods"

Notifications You must be signed in to change notification settings

Kipok/understanding-momentum

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Understanding the Role of Momentum in Stochastic Gradient Methods

To reproduce experiments performed in the paper, see code and comments in experiments.ipynb.

To run logistic regression on MNIST or ResNet-18 on CIFAR, you can use scripts in lr_on_mnist and resnet_on_cifar correspondingly. Run python main.py --help for list of available hyperparameters in both scripts.

To be able to run the code, you need to install the following Python packages: numpy, matplotlib, pytorch (v0.4), tensorboardx

About

Code for the paper "Understanding the Role of Momentum in Stochastic Gradient Methods"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published