This repository is the PyTorch implementation of "GLSearch: Maximum Common Subgraph Detection via Learning to Search" (ICML2021). For more details, please refer to our paper.
Package requirements are in requirements.txt. Please install required packages and versions. We now provide a Dockerfile to make this setup easier.
Mark /model/OurMCS/ and /src/ as the source directories.
To build and run the Docker environment use the commands below:
cd /path/to/GLSearch
nvidia-docker build -t "glsearch" .
nvidia-docker run -e "HOSTNAME=$(cat /etc/hostname)" -v /path/to/GLSearch:/workspace -it glsearch bash
Get the datasets from: https://drive.google.com/drive/folders/1l7_WmO54gPHZ4YwzOqewZtZhWIxn44ZV?usp=sharing
Place the *.klepto files under a new directory /save/OurModelData/ (ex. /save/OurModelData/aids700nef_..._None.klepto).
We noticed Google Drive sometimes renames downloaded files, which will cause issues in data loading. Please check the filenames after downloading.
Data loading is done on line 310 of model/OurMCS/data_model.py (as of commit 4e10d5f79e7f5bd53e74628f52619a0fc10f4c80). You can check here whether tp is pointing to the correct file (NOTE: it is optional whether we include the .klepto extension in tp).
Select 1 of 3 desired dataset on lines 62-77 of model/OurMCS/config.py by commenting out the rest.
Run the below command:
python3 main.py
Replace config.py with config_training.py.
Run the below command:
python3 main.py