search

bd-j / prospector

Python code for Stellar Population Inference from Spectra and SEDs
http://prospect.readthedocs.io
MIT License
153 stars 71 forks source link

My prospector output the stellar mass always be 10^7 Msun #325

Closed Tiantian92 closed 4 months ago

Tiantian92 commented 4 months ago

Hi, Does there any idea about the output the stellar mass always be 10^7 Msun?

Thanks!!

bd-j commented 4 months ago

Sorry, this is not enough information for me to go on? Are you referring to a specific issue raised previously?

Tiantian92 commented 4 months ago

Hi, Benjamin

Since Github can't attach the .ipynb file, I directly reply here.

I am using Prospector to do non-parametric SED fitting for my paper. I attach my test program.

I meet a problem, the stellar mass is always 1e7 Msun (It didn't change, even I change the target).

Please see the attachment, the stellar mass of the target should be 6.7e9 Msun (form CIGALE result, I also attache the best model from CIGALE). The input flux is in the unit of mJy in the .ipynb attachment.

mJy = np.array([0.004101701013211826, 0.013648470958934573,

0.06038627617572804, 0.23688319329593716, 0.485215034270925, 0.8069641641717966, 1.2211422257147735, 0.9826525828295297, 0.6450598332848297]) ####

And the redshift of this target is 0.074.

Thanks!

Ben Johnson @.***> 于2024年3月14日周四 09:44写道：

Sorry, this is not enough information for me to go on? Are you referring to a specific issue raised previously?

— Reply to this email directly, view it on GitHub https://github.com/bd-j/prospector/issues/325#issuecomment-1996240783, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOYI43SAFKDOWYEZMC2NQ43YYD6HXAVCNFSM6AAAAABETUKRGOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSOJWGI2DANZYGM . You are receiving this because you authored the thread.Message ID: @.***>

bd-j commented 4 months ago

Hi, the ipynb did not get attached. I suspect the issue is the prior for mass does not extend below 1e7. If you adjust the prior then it should be fine. See e.g.

https://prospect.readthedocs.io/en/latest/models.html#parameter-specification

In your build_model function you can change mini and maxi values for many priors

Tiantian92 commented 4 months ago

Hi, It seems that changing the prior didn't much improve the result. I refer the demo to create my program(https://github.com/bd-j/prospector/blob/main/demo/InteractiveDemo.ipynb). I attach my program in the following txt file. prospector_tw.txt

Thanks!

bd-j commented 4 months ago

Your model uses the luminosity distance (the lumdist) parameter to do the distance dimming, but it does not seem that you change it from the default (10 Mpc) by adding an entry for "ldist" to run_paramters that is appropriate for your object. Thus you are getting the mass that would produce the observed fluxes if the source was at 10 Mpc.

If you want to use the redshift to set the luminosity distance then you have to remove the "lumdist" parameter from your model (e.g., _ = model_params.pop("lumdist") in your build_model() function)

Tiantian92 commented 4 months ago

Yes！！Thanks！！！ The problem is resolved now！