mhnn_plus

improved version of MHNN (full hypergrah message passing & hypergraph attention)

🚀 Environment Setup

• We'll use conda to install dependencies and set up the environment. We recommend using the Python 3.9 Miniconda installer.
• After installing conda, install mamba to the base environment. mamba is a faster, drop-in replacement for conda:

  conda install mamba -n base -c conda-forge

• Create a new environment named mhnnp and install dependencies.

  mamba env create -f env.yml

• Activate the conda environment with conda activate mhnnp.

📌 Datasets

Dataset	Graphs	Task type	Task number	Metric
OPV	90,823	regression	8	MAE
OCELOTv1	25,251	regression	15	MAE
PCQM4Mv2	3,746,620	regression	1	MAE

🔥 Model Training

OCELOTv1

1. We provide training scripts scripts/ocelot/train.sh for MHNN and hypergraph neural networks in DHG package by running:

   bash scripts/ocelot/train.sh [MODEL_NAME] [TASK_ID]

   For example, we can train HGNNP for one task (14: HOMO target)

   bash scripts/ocelot/train.sh HGNNP 14

   Available models can be found in the script or DHG DOCS.

2. The ocelot dataset will be downloaded automatically at the first time of training.

3. Task ID for different tasks can be found here.

4. The training results are in the folder exp_results/ocelot.

🌈 Acknowledgements

This work was supported as part of NCCR Catalysis (grant number 180544), a National Centre of Competence in Research funded by the Swiss National Science Foundation.

📝 Citation

If you find our work useful, please consider citing it:

@article{chen2024molecular,
    author = {Chen, Junwu and Schwaller, Philippe},
    title = "{Molecular hypergraph neural networks}",
    journal = {The Journal of Chemical Physics},
    volume = {160},
    number = {14},
    pages = {144307},
    year = {2024},
    doi = {10.1063/5.0193557},
    url = {https://doi.org/10.1063/5.0193557},
}

📫 Contact

If you have any question, welcome to contact me at:

Junwu Chen: [email protected]