Czm369 / simmatch

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SimMatch: Semi-supervised Learning with Similarity Matching (CVPR2022)

This repository contains PyTorch evaluation code, training code and pretrained models for SimMatch. Most of the code in this repository is adapt from here.

For details see SimMatch: Semi-supervised Learning with Similarity Matching by Mingkai Zheng, Shan You, Fei Wang, Chen Qian, and Chang Xu



This repository is based on ImageNet dataset, We also provide the training code and logs for cifar10/100, please download it from this link.


To run the code, you probably need to change the Dataset setting (ImagenetPercentV2 function in dataset/, and Pytorch DDP setting (dist_init function in util/ for your server environment.

The distributed training of this code is based on slurm environment, we have provided the training scrips in script/

We also provide the pre-trained model.

Arch Setting Epochs Accuracy Download
SimMatch ResNet50 1% 400 67.2 % simmatch-1p.pth
SimMatch ResNet50 10% 400 74.4 % simmatch-10p.pth

If you want to test the pre-trained model, please download the weights from the link above, and move them to the checkpoints folder. The evaluation scripts also have been provided in script/


If you find that SimMatch interesting and help your research, please consider citing it:

    author    = {Zheng, Mingkai and You, Shan and Huang, Lang and Wang, Fei and Qian, Chen and Xu, Chang},
    title     = {SimMatch: Semi-Supervised Learning With Similarity Matching},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2022},
    pages     = {14471-14481}



Language:Python 98.1%Language:Shell 1.9%