Getting the current sample variable value so it can be used to compute the observed inputs in a secondary model likelihood

[Vital]

@junpenglao

I am still a bit puzzled by your problem here. To me, if you are trying to integrate the red part and use as an observed, you can compute the area as:

    # Remove from the observation the proposed continuum
    emission_obs = obs_flux - continuum_flux

    # Integrate the line area
    line_area_obs = tt.sum(emission_obs[idcs_mask])/(obs_wave[idcs_mask

That is perfect!! (Too obvious for me to see!) Thanks a lot!!!.

Sorry for the simple example, I have made it more complicated than necessarily: The second approach is much nicer but I cannot apply it: In reality, we do not have an equation to model the continuum in galaxies, we just fit a linear system with the light histogram of each star times how many stars of that type there are. So I have to integrate the area (which is not linear either but has a drop in flux due to the atoms absorbing light at that wavelength). And yes the lines lines area is computed different as it is done here :)

Thanks a lot!!!