Light

moskomule / shampoo.pytorch Goto Github PK

View Code? Open in Web Editor NEW

73.0 3.0 8.0 79 KB

An implementation of shampoo

Python 100.00%

shampoo pytorch

shampoo.pytorch's Introduction

shampoo.pytorch

An implementation of shampoo, proposed in Shampoo : Preconditioned Stochastic Tensor Optimization by Vineet Gupta, Tomer Koren and Yoram Singer.

#  Suppose the size of the tensor grad (i, j, k),
#  dim_id = 1 and dim = j
grad = grad.transpose_(0, dim_id).contiguous()  # (j, i, k)
transposed_size = grad.size()
grad = grad.view(dim, -1)  # (j, i x k)
grad_t = grad.t()  # (i x k, j)
precond.add_(grad @ grad_t)  # (j, j)
inv_precond.copy_(_matrix_power(state[precond, -1 / order))  # (j, j)
grad = grad = inv_precond @ grad  # (j, i x k)
grad = grad.view(transposed_size)   # (j, i, k)

shampoo.pytorch's People

Stargazers

Watchers

Forkers

murari023 shubhampachori12110095 geazy zhengmingzhang ls6468 ml-lab whu-dft lencof

shampoo.pytorch's Issues

Problem about Cifar10 Expriements Reproduction

Hi there,
Thanks for your great work of shampoo implementation in Pytorch. I'm trying to reproduce the cifar10 results in the Shampoo paper. But I got a much lower testing results. I have tried changing the learning rate form 0.01 to 10(according to the paper suggests), but still got a near 85% acc. Here are my experiments results:

We use the Resnet32 network in Cifar10 experiments.
--momentum, 0.9
--epsilon, 1e-4
--batchSize, 128

lr=0.1:(250 epochs)

Training Loss	Training Acc	Testing loss	Testing Acc
0.65	77.03%	0.68	76.39%

lr=1: (250 epochs)

Training Loss	Training Acc	Testing loss	Testing Acc
0.25	91.33%	0.57	84.04%

lr=2: (250 epochs)

Training Loss	Training Acc	Testing loss	Testing Acc
0.23	91.87%	0.72	82.02%

lr=5: (250 epochs)

Training Loss	Training Acc	Testing loss	Testing Acc
0.22	92.33%	0.75	82.04%

When training for 500 epochs for different lr above, the testing acc ramains almost the same. Still can't reach even 90% acc.

Any idea or suggestions about this problem? Thanks for your time.

Wrong matrix power?

In your pseudocode, you take H_t^i to the power -1/(2k), but in your code you take precond to the power -1 / order, not -1 / (2 * order). Is this a bug or am I missing something?

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.