amoeba2

amoeba2 is a Bayesian model of OH emission and absorption spectra. Written in the bayes_spec Bayesian modeling framework, amoeba2 implements two models, one for predicting optical depth spectral only, and one for predicting both optical depth spectra and emission spectra. The bayes_spec framework provides methods to fit these models to data using Monte Carlo Markov Chain methods.

Useful information can be found in the amoeba2 Github repository, the bayes_spec Github repository, and in the tutorials below.

Installation

conda create --name amoeba2 -c conda-forge pymc pytensor pip
conda activate amoeba2
pip install amoeba2

Tutorials:

• AbsorptionModel Tutorial
  • Simulating Data
  • Model Definition
  • Variational Inference
  • Posterior Sampling: MCMC
  • Visualization using Mean Point Estimate
• EmissionAbsorptionModel Tutorial
  • Simulating Data
  • Model Definition
  • Variational Inference
  • Posterior Sampling: MCMC
• Optimization Tutorial
  • Model Definition and Simulated Data
  • Optimize
  • Optimize with MCMC
• Real Data Example
  • Load the Data
  • Optimization

API:

• amoeba2
  • amoeba2