The density functional theory (DFT) is a workhorse of modern computational chemistry and materials science. It has brought practically useful accuracy to models of catalysis, electrochemistry, photonics, etc. Computing power growth along with emergence of machine learning (ML) methods raised the demand of complex simulation workflows that include DFT codes. Efficient data exchange with DFT software is challenging due to size of arrays that describe electronic structure of simulated systems. Aiming to provide a fast and portable way to import and export large data structures of DFT codes we have developed a new applications programming interface, Atomic Simulation Interface (ASI), and implemented it in established electronic structure software packages. ASI offers a set of functions that are of utility for interoperation with ML models and for embedding in multi scale simulation frameworks. ASI is designed to be easily and efficiently implemented in existing DFT codes. A Python wrapper is provided to simplify ASI usage by client codes.
This online session is open to all. It will use the Blackboard Collaborate platform.