Welcome to GrainLearning’s documentation!

GrainLearning is a Bayesian uncertainty quantification and propagation toolbox for computer simulations of granular materials. The software is primarily used to infer and quantify parameter uncertainties in computational models of granular materials from observation data, which is also known as inverse analyses or data assimilation. Implemented in Python, GrainLearning can be loaded into a Python environment to process the simulation and observation data, or alternatively, as an independent tool where simulation runs are done separately, e.g., via a shell script.

If you use GrainLearning, please cite the version of the GrainLearning software you used and the following paper:

H. Cheng, T. Shuku, K. Thoeni, P. Tempone, S. Luding, V. Magnanimo. An iterative Bayesian filtering framework for fast and automated calibration of DEM models. Comput. Methods Appl. Mech. Eng., 350 (2019), pp. 268-294, 10.1016/j.cma.2019.01.027

Contents:

• Introduction
• Installation
  • Package management and dependencies
  • Documentation
• Dynamic systems
  • The dynamic system module
  • Interact with third-party software
• Bayesian filtering
  • Bayes’ theorem
  • The inference module
  • The sampling module
  • Iterative Bayesian filter
• RNN Module
  • Train a RNN with your own data
  • Make a prediction with a pre-trained model
  • Use a trained RNN model in grainLearning calibration process
  • The RNN model
• Tutorials
  • Linear regression with a Python model
  • Linear regression with a “software” model
  • GrainingLearning as a postprocessing tool
• Examples
  • Bayesian calibration of DEM models
• Contributing guidelines
  • You have a question
  • You think you may have found a bug
  • You want to make some kind of change to the code base
• Python API
  • Calibration toolbox
  • Dynamic systems
  • Iterative Bayesian Filter
  • Inference
  • Sampling
  • RNN Module

Indices and tables

• Index

• Module Index

• Search Page