-
Hello, was reading your DPA-1 paper [1], and wanted to try out the efficacy of model pre-training on my own dataset / problem. I was wondering if the checkpoints for the OC20 pre-trained model are available anywhere? If not, I can pre-train the model myself, but wanted to ask in case I can save a fair bit of GPU time. [1] Zhang, Duo, et al. "DPA-1: Pretraining of Attention-based Deep Potential Model for Molecular Simulation." arXiv preprint arXiv:2208.08236 (2022). |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 1 reply
-
Hi, we recently uploaded the trained model on AIS-Square, you can find manual and directly download it here. |
Beta Was this translation helpful? Give feedback.
Hi, we recently uploaded the trained model on AIS-Square, you can find manual and directly download it here.