search

gregory-kyro / HAC-Net

HAC-Net: A Hybrid Attention-Based Convolutional Neural Network for Highly Accurate Protein-Ligand Binding Affinity Prediction
MIT License
50 stars 7 forks source link

How to train 2 GCNs? #8

Closed yuwang-langrui closed 3 weeks ago

yuwang-langrui commented 3 weeks ago

Hello Gregory, First of all, thank you very much for sharing this amazing work. I have a question while trying to run your code. Your paper mentions that you trained two GCN networks, and I also noticed that the prediction examples show the use of parameters from two GCNs. However, I am not sure how to train two different GCNs. Should I train the dataset twice? Or have I misunderstood something? I hope to get your response.

Regards,

gregory-kyro commented 3 weeks ago

Hello,

We train two GCNs separately; the architecture and training data are identical for both cases. Simply train the same architecture twice with the same data, and you will obtain two sets of slightly different parameters.

I hope this helps. Please feel free to reach out with any additional questions.

Best, Gregory