zjfheart / badlabels Goto Github PK
View Code? Open in Web Editor NEWChallenging label noise called BadLabel; Robust label-noise learning called Robust DivideMix
Home Page: https://arxiv.org/abs/2305.18377
Challenging label noise called BadLabel; Robust label-noise learning called Robust DivideMix
Home Page: https://arxiv.org/abs/2305.18377
Hi,
First of all, congratulations for your work !
When I run
python robust_dividemix.py --gpu 0 --r 02 --dataset 'cifar100' --log 'robdm_bad' --noise-file '../eval_badlabels/noise/badlabels_cifar100_r0.2.json' --warmup 20 --lam 1 --tol 0.01 --max-iter 500 --noise-mode 'bad' &
it outputs this
load noise labels from ../eval_badlabels/noise/badlabels_cifar100_r0.1.json ..
===> epoch: 20, lambda: 1.0.
gmm.means: [[0.52740902]
[0.42178395]]
gmm.means: [[0.40338232]
[0.5466281 ]]
gmm0.means: [[0.42655838]
[0.43710862]]
gmm0.converged: True
gmm0.n_iter_ 127
gmm0.means: [[0.46136058]
[0.44749019]]
gmm0.converged: True
gmm0.n_iter_ 162
Train Net1
load noise labels from ../eval_badlabels/noise/badlabels_cifar100_r0.1.json ..
labeled data has a size of 1645
num of clean data is 1135 (69.00%)
load noise labels from ../eval_badlabels/noise/badlabels_cifar100_r0.1.json ..
unlabeled data has a size of 48355
num of clean data is 43865 (90.71%)
756
0it [00:00, ?it/s]
Traceback (most recent call last):
File "robust_dividemix.py", line 77, in train
inputs_u, inputs_u2 = unlabeled_train_iter.next()
AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute 'next'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "robust_dividemix.py", line 374, in <module>
train(epoch, net1, net2, optimizer1, labeled_trainloader, unlabeled_trainloader) # train net1
File "robust_dividemix.py", line 80, in train
inputs_u, inputs_u2 = unlabeled_train_iter.next()
AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute 'next'
it seems that the code does not go to except.
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.