Graph Few-shot Class-incremental Learning (WSDM 2022)

Paper is available here.

Requirements

python==3.7.10

pytorch==1.8.1

cuda=11.1

Useage

Go to the directory

cd incremental

Pretrain

python pretrain.py --use_cuda --dataset Amazon_clothing

Meta-train and Evaluation

python meta_train.py --use_cuda --dataset Amazon_clothing --episodes 1000 --incremental --checkpoint 100 --way 3

You may need to change the pretraining checkpoint number according to your need or result.

Citation

@inproceedings{tan2022graph,
  title={Graph few-shot class-incremental learning},
  author={Tan, Zhen and Ding, Kaize and Guo, Ruocheng and Liu, Huan},
  booktitle={Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining},
  pages={987--996},
  year={2022}
}