search

Starfish-develop / Starfish

Tools for Flexible Spectroscopic Inference
https://starfish.readthedocs.io/en/latest/
BSD 3-Clause "New" or "Revised" License
70 stars 22 forks source link

Extinction #121

Closed mileslucas closed 5 years ago

mileslucas commented 5 years ago

Documenting a discrepancy- in the paper eqn. 3 (Arxiv) allows for extinction, however, this doesn't exist anywhere in the code. Extinction is one of my major use-cases. I say we implement the extinction package. Perhaps wrapped in a transform as we've discussed at #64. I'm partial to the Cardelli 89 equation by default because my work is in the 1-5um regime, but maybe we could have a semi-intelligent way to choosing an equation based on the incoming wavelength range (if not manually configured).

I think accompanying this we should be able to specify a model without doing any extinction calcs, but I think that falls under the larger rewrite discussions we've had a la #103 .

iancze commented 5 years ago

Agreed! This was something that we implemented in other forms (as I'm sure most astronomers have at some point) but we never properly included it in Starfish.

We could also use some of the existing extinction (mature) Python packages. Both extinction and dust_extinction look to be pretty good.

mileslucas commented 5 years ago

This is being incorporated on refactor/core-numpy branch under model/transforms.py