GRAPH-0 / JODO

Learning Joint 2D & 3D Diffusion Models for Complete Molecule Generation
MIT License
40 stars 8 forks source link

Training duration #6

Open DWF-666 opened 3 days ago

DWF-666 commented 3 days ago

Hi, I see you have a large number of iterations and would like to ask you how long I have to train on v100 and whether you set up an early stop mechanism, looking forward to your reply!

GRAPH-0 commented 3 days ago

I have not trained on V100 before. If you accept the performance sacrifice, you can directly reduce the training step or the model size to speed up the training. Early stopping is not suitable for this.