DeePKS is a machine-learning aided density funcitonal model that fits the energy difference between highly accurate but computationally demanding method and effcient but less accurate method via neural-network. As such, the trained DeePKS model can provide highly accurate energetics (and forces) with relatively low computational cost, and can therefore act as a bridge to connect expensive quantum mechanic data and machine-learning-based potentials. While the original framework of DeePKS is for molecular systems, please refer to this reference for the application of DeePKS in periodic systems.

Detailed instructions on installing and running DeePKS can be found on this website. An example for training DeePKS model with ABACUS is also provided. The DeePKS-related keywords in INPUT file can be found here.

Note: Use the LCAO basis for DeePKS-related calculations