Stego-SandBox

dataset

download url

methods

The implementation is based on the repo GraphSaint

requirements

• python == 3.8.11
• pytorch == 1.8.0
• cython == 0.29.24
• g++ == 5.4.0
• numpy == 1.20.3
• scipy == 1.6.2
• scikit-learn == 0.24.2
• pyyaml == 5.4.1

before running

We have a cython module which need compilation before training can start. Compile the module by running the following from the root directory:

python setup.py build_ext --inplace

baseline

python main.py --data_prefix ${relative_data_dir} --train_config ./train_config/no_graph.yml --no_graph --repeat_time 1 

using context information

python main.py --data_prefix ${relative_data_dir} --train_config ./train_config/gat_192_8_with_graph.yml --repeat_time 1

relative_data_dir

decompress the downloaded data, then relative_data_dir can be any one of the sub_dir, for example:

relative_data_dir=./data/reddit-ac-1-onlyends_with_isolated_bi_10percent_hop1

For details of the methods and results, please refer to our paper.