Implement More Methods!

[mileslucas]

Here are a list of dust extinction models that have been implemented somewhere else (along with their paper) that could be implemented into this repository.

• ccm89 code (ae9ed25)
• od94 code (ae9ed25)
• cal00 code (5e5011b)
• f99 code (Added F99 model #30)
• f04 code (Added F04 model #31)
• vcg04 code (added gcc09 and vcg04 #13 thanks @siddharthlal25 )
• gcc09 code (added gcc09 and vcg04 #13 thanks @siddharthlal25 )
• m14 code (Added M14 model #33)
• g16 code (Added G16 model #34)
• f19 code (Added F19 model #32)

[icweaver]

I'll go ahead and get started on F99

[icweaver]

Ok, just opened a PR #30! Everything looks to be working alright except for Measurements handling. I think this might be related to the package (Dierckx.jl) I am using for interpolation?

[icweaver]

Ok, just opened a PR #30! Everything looks to be working alright except for Measurements handling. I think this might be related to the package (Dierckx.jl) I am using for interpolation?

Solved: #30 (comment)

[icweaver]

Is there anything else that people would like to see in the F99 PR? If it's looking ok to everyone, I could go ahead and get started on F04 since it seems to be a natural extension of F99!

[icweaver]

Adding a quick todo here to track the general changes proposed for the other models that came up while working on the F99 PR:

• switch from randn to rand in Measurements tests so that we avoid taking +/- of potentially negative values Added F99 model #30 (comment)
• remove low-level bounds checks and corresponding tests, Added F99 model #30 (comment)
• add AD support warning (supported in Interpolations.jl, but not Dierckx.jl), Added F99 model #30 (review) Note: currently using Dierckx.jl because it has support for irregularly-gridded third order B-Splines, as used in the astropy version (Investigating in Migrate away from Dierckx.jl #54)
• switch to static data structure for internals, Added F99 model #30 (review)

Would it better to open this as another issue?

[icweaver]

Is there anything else that people would like to see in the F99 PR? If it's looking ok to everyone, I could go ahead and get started on F04 since it seems to be a natural extension of F99!

It looks like F19 is the last Fitzpatrick et al. model, so I'd be happy to close that one out too.

Just closing loose ends: F99 (#30), F04(#31), F19(#32)

[icweaver]

I can get started on the remaining two models if there are no objections

M14 (#33), G16(#34)

[icweaver]

Getting nostalgic looking at this. Are we ready to close this now and just open up new model issues as they come up?

Edit: Including #29

[icweaver]

Okie doke, will go ahead and close this. Also opened a draft PR here to start investigating the one remaining AD issue