Skip to content

Code for paper "Dimensionality-Driven Learning with Noisy Labels" - ICML 2018

Notifications You must be signed in to change notification settings

xingjunm/dimensionality-driven-learning

Repository files navigation

Code for ICML 2018 paper "Dimensionality-Driven Learning with Noisy Labels".

- Update (2018.07): Issues fixed on CIFAR-10.

- Update (2019.10): Start training with symmetric cross entropy (SCE) loss (replacing cross entropy).

The Symmetric Cross Entropy (SCE) was demonstrated can improve several exisiting methods including the D2L: ICCV2019 "Symmetric Cross Entropy for Robust Learning with Noisy Labels" https://arxiv.org/abs/1908.06112 https://github.com/YisenWang/symmetric_cross_entropy_for_noisy_labels

- Update (2020.03): convergence issue on CIFAR-100 when using SCE loss: learning rate, data augmentation and parameters for SCE.

1. Train DNN models using command line:

An example:

python train_model.py -d mnist -m d2l -e 50 -b 128 -r 40 

-d: dataset in ['mnist', 'svhn', 'cifar-10', 'cifar-100']
-m: model in ['ce', 'forward', 'backward', 'boot_hard', 'boot_soft', 'd2l']
-e: epoch, -b: batch size, -r: noise rate in [0, 100]

2. Run with pre-set parameters in main function of train_model.py:

    # mnist example
    args = parser.parse_args(['-d', 'mnist', '-m', 'd2l',
                              '-e', '50', '-b', '128',
                              '-r', '40'])
    main(args)
    
    # svhn example
    args = parser.parse_args(['-d', 'svhn', '-m', 'd2l',
                              '-e', '50', '-b', '128',
                              '-r', '40'])
    main(args)
    
    # cifar-10 example
    args = parser.parse_args(['-d', 'cifar-10', '-m', 'd2l',
                              '-e', '120', '-b', '128',
                              '-r', '40'])
    main(args)
    
    # cifar-100 example
    args = parser.parse_args(['-d', 'cifar-100', '-m', 'd2l',
                              '-e', '200', '-b', '128',
                              '-r', '40'])
    main(args)

Requirements:

tensorflow, Keras, numpy, scipy, sklearn, matplotlib

About

Code for paper "Dimensionality-Driven Learning with Noisy Labels" - ICML 2018

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages