CiderPress: Machine Learning Exchange-Correlation Functionals

Welcome to the CiderPress documentation! CiderPress is a Python package built for running machine-learned exchange-correlation functionals within the CIDER framework. This documentation is currently under construction. In the meantime, please feel free to post a Github issue or reach out the developers at kylebystrom@gmail.com.

Contents:

• Overview and Installation Instructions
  • What is the CIDER formalism?
  • WARNING: The CiderPress Code Base is Experimental
  • Installation
  • How can I run a CIDER calculation?
  • How can I train a CIDER functional?
  • Known Issues
  • Questions and Comments
  • Citing
• Theory Overview
  • Uniform Scaling
  • Machine Learning Framework for CIDER Functionals
  • Numerical Evaluation of NLDF Features
• Density and Orbital Features in CiderPress
  • Semilocal Features (SL)
  • Nonlocal Density Features (NLDF)
  • Nonlocal Orbital Features (NLOF)
  • Smooth Density Matrix Exchange (SDMX)
• DFT Module
  • The Settings Module
  • The Plans Module
  • The Feature Normalizer Module
  • The Transform Data Module
  • The XC Evaluator Modules
• The Models Module
  • ciderpress.models.train
  • ciderpress.models.dft_kernel
  • ciderpress.models.kernels
  • Kernel Tools
• PySCF Interface
  • dft
  • pbc.dft
  • analyzers
  • descriptors
• GPAW Interface
  • GPAW Calculator Interface
• C Extension Libraries
  • Plane-wave Utilities (pwutil)