Denghui Lu

[Feature Request] Reduce MD loop time

2

**Summary** The current dpgen workflow is divided into three parts: vasp labeling, training, and MD exploration. An ideal implementation attemps to assign as large as possible part of (financial/temporal) cost...

### Reminder - [x] Have you linked an issue with this pull request? - [ ] Have you added adequate unit tests and/or case tests for your pull request? -...

### Background The current BPCG method has low GPU memory efficiency, and all K-point Psi are stored in the GPU memory. ### Describe the solution you'd like Improve memory efficiency...

### Background A GPU program may encounter issues if it runs out of memory, so it's important to alert our users about this potential problem. ### Describe the solution you'd...

### Details  Reduce redundant computing of variables like 'ib * max_npw + ig ' ### Task list for Issue attackers (only for developers) - [X] Reproduce the performance issue...