kingjamessong / batched Goto Github PK
View Code? Open in Web Editor NEWECCV22 Paper "Batch-efficient EigenDecomposition for Small and Medium Matrices"
Home Page: https://arxiv.org/abs/2207.04228
License: MIT License
ECCV22 Paper "Batch-efficient EigenDecomposition for Small and Medium Matrices"
Home Page: https://arxiv.org/abs/2207.04228
License: MIT License
Hi,
When running inside network training, after a while with some probability i'm getting nan for one instance in the batch. All the eigen values and the eigen vecotrs are nan.
The matrix batch in this case is 1024X64X64 and it is real and symmetric. I'm using ed_plus.
It seems to have a relation to the batch size, as the same matrix that caused the nan, doesn't nan when taken alone or with a smaller batch.
Ant idea of where I should investigate this?
Thanks!
BatchED.utils_ed_plus.Batched_ED_Plus.apply just gives an error... Generally, its quite hard to find a use for such a code.
[utils_ed_plus.py:217](https://file+.vscode-resource.vscode-cdn.net/mnt/hd1/Dropbox/tmp/projects/SNIPPETS/torch_eigs/BatchED/utils_ed_plus.py:217), in DivideStep(x_orig)
[215](file:///mnt/hd1/Dropbox/tmp/projects/SNIPPETS/torch_eigs/BatchED/utils_ed_plus.py?line=214) t1 = x[:, :half_size, :half_size]
[216](file:///mnt/hd1/Dropbox/tmp/projects/SNIPPETS/torch_eigs/BatchED/utils_ed_plus.py?line=215) t2 = x[:, half_size:, half_size:]
--> [217](file:///mnt/hd1/Dropbox/tmp/projects/SNIPPETS/torch_eigs/BatchED/utils_ed_plus.py?line=216) return torch.cat([t1,t2],dim=0), xnorm, beta
RuntimeError: Sizes of tensors must match except in dimension 0. Expected size 4 but got size 5 for tensor number 1 in the list.
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.