Skip to content

Code for "Candidate-aware Selective Disambiguation Based On Normalized Entropy for Instance-dependent Partial-label Learning" (ICCV 2023).

Notifications You must be signed in to change notification settings

shuohe1995/NE_PLL

Repository files navigation

NE_PLL

Code for "Candidate-aware Selective Disambiguation Based On Normalized Entropy for Instance-dependent Partial-label Learning" (ICCV 2023).

Requirements:

python 3.9.12

torch 1.12.1

torchvision 0.15.2

Dataset:

./"dataset_root"

Candidate label generation:

Model checkpoints saved in ./model_path

Run:

sh run.sh 0 cifar10 10

sh run.sh 1 cifar100 100

sh run.sh 2 fmnist 10

sh run.sh 3 kmnist 10

Parameters varied in run.sh.

Log file saved in ./output_log

About

Code for "Candidate-aware Selective Disambiguation Based On Normalized Entropy for Instance-dependent Partial-label Learning" (ICCV 2023).

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published