Skip to content

Pingala: Bayesian inference library for deep learning and geoscientifc models

Notifications You must be signed in to change notification settings

sydney-machine-learning/pingala

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 

Repository files navigation

pingala

Pingala is a Python based machine learning software library that features Bayesian inference for deep learning and geoscientific models via MCMC

It is named after "Pingala" - to pay tribute to ancient Hindu scientist, who invented the binary number system: https://en.wikipedia.org/wiki/Pingala

Features

  • MCMC (sequential and parallel implementation)
  • Parallel tempering MCMC for statistical models
  • Parallel tempering MCMC for neural networks: 1. FNN, 2. RNN, 3. LSTM, 4. GRU
  • Parallel tempering MCMC for geoscientific models - Badlands lansacpe evolution models
  • Parallel tempering MCMC for geoscientific models - pyReef-Core evolution models
  • Surrogate assisted parallel tempering MCMC for neural networks
  • Surrogate assited parallel tempering MCMC for geoscientific models - Badlands lansacpe evolution models

Research

The library will combine the Python software frameworks from the following papers:

About

Pingala: Bayesian inference library for deep learning and geoscientifc models

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published