usage: lifelong.py [-h] [-c CONFIG] [--device DEVICE] [--data-root DATA_ROOT] [--dataset DATASET] [--model MODEL] [--load LOAD]
                    [--save SAVE] [--optm OPTM] [--lr LR] [--batch-size BATCH_SIZE] [--jump JUMP] [--iteration ITERATION]
                    [--memory-size MEMORY_SIZE] [--seed SEED] [-p] [--eval EVAL] [--sample-rate SAMPLE_RATE] [--k K]
                    [--hidden HIDDEN [HIDDEN ...]] [--drop DROP [DROP ...]] [--merge MERGE]

    optional arguments:
    -h, --help            show this help message and exit
    -c CONFIG, --config CONFIG
                            config file path
    --device DEVICE       cuda or cpu
    --data-root DATA_ROOT
                            dataset location
    --dataset DATASET     cora, citeseer, or pubmed
    --model MODEL         LGL or SAGE
    --load LOAD           load pretrained model file
    --save SAVE           model file to save
    --optm OPTM           SGD or Adam
    --lr LR               learning rate
    --batch-size BATCH_SIZE
                            minibatch size
    --jump JUMP           reply samples
    --iteration ITERATION
                            number of training iteration
    --memory-size MEMORY_SIZE
                            number of samples
    --seed SEED           Random seed.
    -p, --plot            increase output verbosity
    --eval EVAL           the path to eval the acc
    --sample-rate SAMPLE_RATE
                            sampling rate for test acc, if ogb datasets please set it to 200
    --k K                 the level of k hop.
    --hidden HIDDEN [HIDDEN ...]
    --drop DROP [DROP ...]
    --merge MERGE         Merge some class if needed.

    @inproceedings{wang2022lifelong,
      title={Lifelong graph learning},
      author={Wang, Chen and Qiu, Yuheng and Gao, Dasong and Scherer, Sebastian},
      booktitle={2022 Conference on Computer Vision and Pattern Recognition (CVPR)},
      year={2022}
    }