This respository is based on the Continual Graph Learning Benchmark CGLB

Get Started | Evaluation & Visualization Toolkit | Acknowledgement

This is the official repository of Sparsified Subgraph Memory for Continual Graph Representation Learning (SSM), which was published in ICDM 2022.

@inproceedings{zhang2022sparsified,
 title={Sparsified Subgraph Memory for Continual Graph Representation Learning},
 author={Zhang, Xikun and Song, Dongjin and Tao, Dacheng},
 booktitle={IEEE International Conference on Data Mining (ICDM)},
 pages={1335--1340},
 year={2022},
 organization={IEEE}
}

Get Started

To run the code, the following packages are required to be installed:

• python==3.7.10
• scipy==1.5.2
• numpy==1.19.1
• torch==1.7.1
• networkx==2.5
• scikit-learn~=0.23.2
• matplotlib==3.4.1
• ogb==1.3.1
• dgl==0.6.1

For all experiments, the starting point is the train.py file, and the different configurations are assigned through the keyword arguments of the Argparse module. For example,

python train.py --dataset Arxiv-CL \
      --backbone GCN \
      --gpu 0 \
      --epochs 200 \
      --sample_nbs False \
      --minibatch False \
      --repeats 5 \
      --ratio_valid_test 0.2 0.2 \
      --overwrite_result False \
      --perform_testing True

Since the graphs can be too large to be processed in one batch on most devices, the --minibatch argument could be specified to be True for training with the large graphs in mini-batches.

 python train.py --dataset Arxiv-CL \
       --backbone GCN \
       --gpu 0 \
       --epochs 200 \
       --sample_nbs False \
       --minibatch True \
       --batch_size 2000 \
       --repeats 5 \
       --ratio_valid_test 0.2 0.2 \
       --overwrite_result False \
       --perform_testing True

In the above example, besides specifying the --minibatch, the size of each mini-batch is also specified through --batch_size. Moreover, some graphs are extremely dense and will run out the memory even with mini-batch training, which could be addressed through the neighborhood sampling specified via --sample_nbs. And the number of neighbors to sample for each hop is specified through --n_nbs_sample. There are also other customizable arguments, the full list of which can be found in train.py.

Modifying the train-validation-test Splitting

The splitting can be simply specified via the arguments when running the experiments. In our implemented pipeline, the corresponding arguments are the validation and testing ratios. For example,

 python train.py --dataset Arxiv-CL \
       --backbone GCN \
       --gpu 0 \
       --epochs 200 \
       --sample_nbs False \
       --minibatch True \
       --batch_size 2000 \
       --repeats 5 \
       --ratio_valid_test 0.4 0.4 \
       --overwrite_result False \
       --perform_testing True

The example above set the data ratio for validation and testing as 0.4 and 0.4, and the training ratio is automatically calculated as 0.2.

Evaluation and Visualization Toolkit

The introduction on the evaluation and visualization toolkit can be found in the official respository of CGLB CGLB.

Acknowledgement

The construction of this repository is based on CGLB.