Xiang

Hi, in principle, because there is a loss for predicting reward from the task state, to minimize the total loss the model should make the right attribution. I suggest trying...

Hi, several observations from your log: 1. There are some cuda packages that were not successfully loaded: ``` 2022-08-18 18:37:54.868858: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcusolver.so.10'; dlerror: libcusolver.so.10:...

I used CUDA 10.1. It is possible to have multiple CUDA versions on the same server machine, you can install CUDA 10.1 as long as you have sudo permission. Maybe...

Hi, this is strange. For DMC it took me ~10 hours to run 1M steps on a single RTX 2080 Ti. Could you check your GPU utilization? My best guess...

Hi, 1. As described in the paper, GNS was used for the single-chain CG polymer dataset. PnR would be too slow for the timestep setting. 2. Have you tried the...