Skip to content

This repository is the official implementation of Dataset Condensation with Contrastive Signals (DCC), accepted at ICML 2022.

Notifications You must be signed in to change notification settings

Saehyung-Lee/DCC

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Dataset Condensation with Contrastive Signals

This repository is the official implementation of Dataset Condensation with Contrastive Signals (DCC), published as a conference paper at ICML 2022. The implementation is based on (https://github.com/VICO-UoE/DatasetCondensation).

Prerequisites

  • pytorch (1.2.0)
  • numpy (1.15.1)
  • torchvision (0.4.0)
  • scipy (1.1.0)

Training and evaluation

To train the DCC (or DSAC) model in the paper, run this command:

python main.py --ipc <1, 10, or 50> --model ConvNet --dataset <CIFAR10, CIFAR100, or imagenet> (--imagenet_group <fine-grained dataset>) --method <DC or DSA> --contrast --save_path <save path name>

Please download ImageNet32x32 at (https://image-net.org/download-images)

About

This repository is the official implementation of Dataset Condensation with Contrastive Signals (DCC), accepted at ICML 2022.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages