Hermes is a repository designed to be a modular compilation of tools for machine learning and autonomous experimentation in materials science domains.
Autonomous Experimentation is one of the latest, most powerful paradigms of research that takes advantage of Machine Learning tools. However, most machine learning code bases are designed for generic data. Contrastingly, in Materials Science, and other sciences, the data has significant physical meaning. The machine learning algorithms must be adapted to account for these physical constraints to be able to make meaningful conclusions from the data. Hermes is a code package that is designed to be a modular compilation for Machine Learning and Autonomous Experimentation in the Materials Science Domains.
The algorithms in Hermes are designed with the relevant physics built-in. This ensures that predictions and conclusions from these tools are physically meaningful. The modularity of the tools in Hermes facilitates the implementation of autonomous laboratories in the materials science domain. There are tools for communicating with instruments, analyzing data, archiving FAIR data, and constructing autonomous research campaigns.