MolHF: A Hierarchical Normalizing Flow for Molecular Graph Generation

This is the official implementation for the paper:

MolHF: A Hierarchical Normalizing Flow for Molecular Graph Generation (IJCAI 2023)

Environment:

• Python 3.7
• Pytorch
• torch-geometric
• rdkit
• networkx

Workflow

1. Data Preprocessing

ZINC250K dataset

python preprocess.py --dataset zinc250k --in_path ./dataset/zinc250k/zinc250k.smi --out_path ./data_preprocessed/zinc250k

Polymer dataset

python preprocess.py --dataset polymer --in_path ./dataset/polymer/polymer.smi --out_path ./data_preprocessed/polymer

2. Training MolHF

Training model on the ZINC250K dataset:

python main.py \
    --dataset zinc250k --device cuda --deq_scale 0.6 \
    --train --save --batch_size 256 --lr 1e-3 \
    --squeeze_fold 2 --n_block 4 \
    --a_num_flows 6 --num_layers 2 --hid_dim 256 \
    --b_num_flows 3 --filter_size 256 \
    --temperature 0.6 --learn_prior --inv_conv --inv_rotate --condition \
    --gen_num 10000 \
    | tee training_zinc250k_molhf.log

Or downloading and using our trained models in

https://drive.google.com/drive/folders/1bgq0gQIzT4GoEfDj_Z9ZeYmXI873gTDm

Training model on the Polymer dataset:

python main.py \
    --dataset polymer --device cuda --deq_scale 0.6 \
    --train --save --batch_size 256 --lr 1e-3 \
    --squeeze_fold 2 --n_block 6 \
    --a_num_flows 8 --num_layers 4 --hid_dim 128 \
    --b_num_flows 3 --filter_size 128 \
    --temperature 0.6 --learn_prior --inv_conv --inv_rotate \
    --gen_num 10000 \
    | tee training_polymer_molhf.log

Or downloading and using our trained models in

https://drive.google.com/drive/folders/1bgq0gQIzT4GoEfDj_Z9ZeYmXI873gTDm

Citation

If you find this repository useful, please consider citing our work:

@inproceedings{zhu2023molhf,
  title={MolHF: A Hierarchical Normalizing Flow for Molecular Graph Generation},
  author={Zhu, Yiheng and Ouyang, Zhenqiu and Liao, Ben and Wu, Jialu and Wu, Yixuan and Hsieh, Chang-Yu and Hou, Tingjun and Wu, Jian},
  booktitle={Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence},
  pages={5002--5010},
  year={2023},
  month={8},
  url={https://doi.org/10.24963/ijcai.2023/556},
}