jrc1995 / demonrangeroptimizer Goto Github PK
View Code? Open in Web Editor NEWQuasi Hyperbolic Rectified DEMON Adam/Amsgrad with AdaMod, Gradient Centralization, Lookahead, iterative averaging and decorrelated Weight Decay
Quasi Hyperbolic Rectified DEMON Adam/Amsgrad with AdaMod, Gradient Centralization, Lookahead, iterative averaging and decorrelated Weight Decay
This is a greate repo, my respect.
I think DEMON based optimizers have some issue. Line 375 momentum.div_(1 - (beta1 ** state['step'])).mul_(nu1).add_(1-nu1, grad)
gives the following error after few iterations of training.
OverflowError: (34, 'Numerical result out of range')
It seems to be a general issue:
https://discuss.pytorch.org/t/overflowerror-34-numerical-result-out-of-range/1907
I tried their recommendation and it still didn't work. My value of beta1 was around -3 when this overflow error was raised. I believe beta1 should stay in [0,1] ?
This problem occurs in other optimizers also. Removing DEMON option seems to get rid of this issue.
Thanks for your contribution. Is there a guideline/comparison you provide for these optimizers? Which one would be best for standard tasks like image classification?
/content/DemonRangerOptimizer/optimizers.py in step(self, display, activate_IA, closure)
883 denom = vt.pow_(group['p']).add_(group['eps'])
884
--> 885 n = state['lr']/denom
886
887 if beta3 > 0.0: # apply AdaMod
KeyError: 'lr'
https://github.com/pytorch/pytorch/commits/master/torch/optim/adam.py
Where is a lot of changes like no grads cope decorator.
Not sure if it affects math/performance.
For your consideration.
I tried it in two tasks, but got nans during training, any suggestions?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.