CurrMG

Codes for "An efficient curriculum learning-based strategy for molecular graph learning"

Reference

If you make advantage of the CurrMG training strategy proposed in our paper, please cite the following in your manuscript:

@article{10.1093/bib/bbac099,
author = {Gu, Yaowen and Zheng, Si and Xu, Zidu and Yin, Qijin and Li, Liang and Li, Jiao},
title = "{An efficient curriculum learning-based strategy for molecular graph learning}",
journal = {Briefings in Bioinformatics},
year = {2022},
month = {04},
issn = {1477-4054},
doi = {10.1093/bib/bbac099},
}

Overview

Environment Requirement

• torch==1.8.0
• dgl==0.5.2
• dgl-lifesci==0.2.5
• rdkit>=2017.09.1

Easy Usage

python main.py -d {DATASET} -mo {MODEL} -me {METRIC} -cu TRUE -rp {SAVED PATH} -dt {DIFFICULTY MEASURER}
Main arguments:
    -d: FreeSolv ESOL Lipophilicity BACE BBBP Tox21 ClinTox SIDER External(your own dataset)
    -mo: GCN GAT MPNN AttentiveFP Pretrained-GIN
    -me: roc_auc_score pr_auc_score r2 mae rmse
    -dt: AtomAndBond Fsp3 MCE18 LabelDistance Joint Two_stage
Optional arguments:
    -s: Random or scaffold splitting type.
    -sr: Split ratio.
    -wt: Weight of difficulty coefficient for d_Joint and d_Two_stage.
    -ct: Power of competence function.
    -ne: Epoches. -lr: Learning rate. -bs: Batch Size. -wd: Weight decay.
For more arguments, please see main.py

Sample data

Once you want to use your own dataset, please follow the file format as JAK2.csv and Mtb.csv in 'test' folder.

Contact

We welcome you to contact us (email: gu.yaowen@imicams.ac.cn) for any questions and cooperations.