Multi-fidelity Generative Deep Learning Turbulent Flows

Nicholas Geneva, Nicholas Zabaras

Documentation transient multi-fidelity Glow (TM-Glow) used for predicting high-fidelity turbulence given a low-fidelity simulation. Please see the paper for additional information regarding the model used and the underlying theory [FoDS] [Arxivs].

Multi-fidelity deep generative turbulence modeling.

User Guide

• Getting Started
  • 1. Clone the repository
  • 2. Downloading Data
  • 3. Set-up Conda Python Environment
  • 4. Start Training
  • 5. Running Pre-trained Models
• F.A.Q.

API Reference

If you are looking for information on a specific function, class or method, this part of the documentation is for you.

• main module
• args module
• nn package
  • Subpackages
  • nn.tmGlow module
  • nn.trainFlowParallel module
• pc package
  • pc.grad1Filter module
  • pc.grad2Filter module
  • pc.physicsConstrained module
• utils package
  • utils.dataLoader module
  • utils.log module
  • utils.parallel module
  • utils.utils module
  • utils.viz module

Additional Information

• Resources
• Contact
• License