Repository navigation
Deprecate GaussianAbsorption1D model - #6200
Conversation
| methods. [#6170] | ||
|
|
||
| - Deprecated ``GaussianAbsorption1D`` model, as it can be better represented | ||
| by subtracting ``Gaussian1D`` from a continuum model. [#6200] |
There was a problem hiding this comment.
Maybe you can say a Const1D model instead of continuum, and say that this allows better control over the continuum level?
There was a problem hiding this comment.
Isn't that too specific? Technically, your continuum can be any model you want, even a blackbody...
|
|
||
| # TODO: Don't need BaseGaussian1D anymore when this is removed. | ||
| @deprecated( | ||
| '2.0', alternative='Gaussian1D and subtract it off continuum model') |
astrofrog
left a comment
astrofrog
left a comment
There was a problem hiding this comment.
Just a couple of small comments and otherwise looks good to me.
e02765d to
75130c3
Compare
|
@astrofrog , I applied your comments. |
75130c3 to
fa0c32d
Compare
|
@pllim - There are some conflicts now. Are we still waiting until tomorrow with this? |
fa0c32d to
aaf6d8e
Compare
|
@bsipocz , rebased. Travis is running on my fork; I cancelled the one here on purpose. |
|
Travis passed on my fork. As for merging, so far I got one 👍 for this approach and no outright objections (both on GitHub issue and on |
|
Let's merge this then. Thanks! |
|
Representing it as a compound model is fine. But has anyone actually checked that the compound model fitting works in this case? Is there a fitting test? |
|
@nden , does this example answer your question? http://synphot.readthedocs.io/en/latest/synphot/tutorials.html#fitting-equivalent-width It works for me but I haven't tested with the new unit support yet. |
|
Yes, something like this. |
4 participants
Note: Only merge if there is no objection by Friday...
Fix #6195
Travis tests on my fork -- https://travis-ci.org/pllim/astropy/builds/243298429