nuclear-multimessenger-astronomy / nmma

A pythonic library for probing nuclear physics and cosmology with multimessenger analysis
https://nuclear-multimessenger-astronomy.github.io/nmma/
GNU General Public License v3.0
30 stars 58 forks source link

model_name Bu2022mv not found in models list #382

Open phelipedarc opened 2 weeks ago

phelipedarc commented 2 weeks ago

Description: I encountered an issue when attempting to use the SVDLightCurveModel with the model Bu2022mv. The model is not found in the list of available models, resulting in an error.

Steps to Reproduce:

from nmma.em.model import SVDLightCurveModel

kn_model = SVDLightCurveModel(model="Bu2022mv", sample_times=sample_times, filters=colors.keys())

The SVDLightCurveModel should initialize correctly with the Bu2022mv model without any errors, indicating that the model is present in the list because it appears when I run nmma.em.model.model_parameters_dict

{'Bu2019nsbh': ['log10_mej_dyn', 'log10_mej_wind', 'KNtheta'],
 'Bu2019lm': ['log10_mej_dyn', 'log10_mej_wind', 'KNphi', 'KNtheta'],
 'Bu2019lm_sparse': ['log10_mej_dyn', 'log10_mej_wind'],
 'Ka2017': ['log10_mej', 'log10_vej', 'log10_Xlan'],
 'TrPi2018': ['inclination_EM',
  'log10_E0',
  'thetaCore',
  'thetaWing',
  'b',
  'L0',
  'q',
  'ts',
  'log10_n0',
  'p',
  'log10_epsilon_e',
  'log10_epsilon_B',
  'xi_N',
  'd_L'],
 'Piro2021': ['log10_Menv', 'log10_Renv', 'log10_Ee'],
 'Me2017': ['log10_mej', 'log10_vej', 'beta', 'log10_kappa_r'],
 'Bu2022mv': ['log10_mej_dyn',
  'vej_dyn',
  'log10_mej_wind',
  'vej_wind',
  'KNtheta'],

ERROR:

    139     raise ValueError("model_name must be specified, got None")
    140 if model_name not in MODELS:
--> 141     raise ValueError(f"model_name {model_name} not found in models list")
    142 model_info = MODELS[model_name]
    144 if not exists(models_home):

ValueError: model_name Bu2022mv not found in models list
mcoughlin commented 2 weeks ago

At this point, I think we do not suggest using the 2022 model, but instead focus on 2019 and hope to release a new one soon. Right @tsunhopang and @ThibeauWouters ?

ThibeauWouters commented 2 weeks ago

Indeed, there have been issues reported with the 2022 model that got resolved with the updated model. However, a reliable surrogate of that model does not exist yet.

mcoughlin commented 2 weeks ago

@ThibeauWouters so maybe we remove from the list?

phelipedarc commented 2 weeks ago

Also related to this issue, I experienced the same error with the models: Bu2022Ye, Bu2023Ye, LANLTP(1/2) and LANLTS(1/2).

Thanks for helping me and addressing this issue so quickly.

ThibeauWouters commented 2 weeks ago

@ThibeauWouters so maybe we remove from the list?

Yes that sounds good. This is also what I suggested some time ago in one of the NMMA calls

mcoughlin commented 2 weeks ago

@bking-astro @atoivonen13 Maybe a discussion for the call next week? Would be good to get someone to update the docs / the code to explain this.