Pytorch Implementation of Continual Learning With Filter Atom Swapping (ICLR'22 Spolight) Paper
If you find this repo helpful for your work, please cite the following bib entries:
@inproceedings{Miao_2022_CLatoms,
title={Continual Learning With Filter Atom Swapping},
author={Miao, Zichen and Wang, Ze and Chen, Wei and Qiu, Qiang},
booktitle={International Conference on Learning Representations},
year={2022}}
CIFAR100 will be automatically downloaded to 'Datasets' dir. Please download ImageNet to 'Datasets/ImageNet'
cifar100 (N=5)
$ ./train_dcf_c100.sh 6 GPU_TO_USE
cifar100 (N=10)
$ ./train_dcf_c100.sh 11 GPU_TO_USE
ImageNet-Sub (N=5)
$ ./train_dcf_img100.sh 6 GPU_TO_USE
ImageNet-Sub (N=10)
$ ./train_dcf_img100.sh 11 GPU_TO_USE