This is a place for experimental addons to the dense_basis
SED fitting code to live. Documentation and usage can be found at the dense_basis docs. The main reason for making this a separate package is the additional dependences this needs (including, but not limited to: pytorch, ranger, emcee, sklearn and joblib)
This includes
- Sampling methods (e.g., MCMC, Nested Sampling and more)
- Optimization techinques (e.g., efficient parallelization, a NN+PCA backend (similar to Speculator; Alsing et al. 2019) or NDinterpolation coupled to the inference tools)
- More visualization/interpretation tools