bd-j / prospector

Python code for Stellar Population Inference from Spectra and SEDs
http://prospect.readthedocs.io
MIT License
157 stars 73 forks source link

Issues while fitting photometry and spectra with different wavelength coverages #295

Closed sdasrc closed 10 months ago

sdasrc commented 1 year ago

Hi,

While fitting both spectra and photometry, the spectra output by model.predict spans the input spectra wavelength, and not the total wavelength covered by both photometry and spectra. In contrast, when fitting only photometry, the output spectra covers the entire wavelength range of the photometry. For example, in the attached figure top-panel, I am fitting only photometry (UV-far IR) and the output spectra (the red plot) covers the entire wavelength range. Adding SDSS spectra (bottom panel), the output spectra (red) runs only as far as the input SDSS spectra (orange).

Is this the expected outcome, or are there certain settings/parameters I am missing here?

Attached plot.

Thanks :)

bd-j commented 11 months ago

Hi @sdasrc, yes this is the expected outcome, model.predict will try to predict the data given in the obs dictionary, but if there is no spectrum it will, for the sake of convenience, return the entire spectrum.

sdasrc commented 10 months ago

Thanks for the clarification @bd-j :)