Code developed during Summer 2020 to implement machine learning and neural networks to predict artificial viscosity values in Pyranda.

Work Title - "Approximating an artificial viscosity operator with neural networks in a shock-capturing scheme"

The images produced by the code match those from the paper.

Reference:

For installation instructions and source code:
Pyranda - https://github.com/LLNL/pyranda/tree/master

Includes:

• .py files to create datasets
• .py files to implement TensorFlow Keras neural networks
  
• .py files that incorporate neural networks into shock dominated problems
  
• .py files that generate plots of density and artificial viscosity
  

Shock-Dominated Problems:

• Burgers' Equation
  
• 1D Sod Shock Tube
  
• Shu-Osher Problem
  
• 2D Sod Shock Tube
  
• Sedov Blast Wave
  
• Triple Point Problem