Prospector doesn't save model to h5 file

[grotova]

Hello, could you please help me to solve an issue:

after running my fit I try to save output to h5 file,

from prospect.io import write_results as writer
hfile = name+".h5"
writer.write_hdf5(hfile=hfile,
run_params=run_params,
model=model,
obs=obs,
sampler=output["sampling"][0],
optimize_result_list=output["optimization"][0],
tsample=output["sampling"][1],
toptimize=output["optimization"][1],
sps=sps)

however, when I then read it to do the plotting:

import prospect.io.read_results as reader
filename = "J1401_24Aug31-13.47.h5"
results1, obs1, model1 = reader.results_from(filename)

model1 = None
This causes issues when I try to use plotting functions provided.

Right before saving my model gives a normal output:
:::::::
<class 'prospect.models.sedmodel.SedModel'>

Free Parameters: (name: prior)

mass: <class 'prospect.models.priors.LogUniform'>(mini=1000000.0,maxi=1000000000000.0)
logzsol: <class 'prospect.models.priors.TopHat'>(mini=-2,maxi=0.19)
dust2: <class 'prospect.models.priors.TopHat'>(mini=0.0,maxi=2.0)
tage: <class 'prospect.models.priors.TopHat'>(mini=0.001,maxi=13.8)
tau: <class 'prospect.models.priors.LogUniform'>(mini=0.1,maxi=100.0)

Fixed Parameters: (name: value [, depends_on])

zred: [0.031]
sfh: [4]
imf_type: [2]
dust_type: [2.]
lumdist: [135.92003034]
add_dust_emission: [ True]
duste_umin: [1.]
duste_qpah: [4.]
duste_gamma: [0.001]

Could you please help me what could be a reason for that?
My goal is to plot posteriors.

Best regards,
Iuliia

[bd-j]

Hi @grotova

When saving your fit results, make sure to include a line that saves the text of the parameter file. e.g.,

run_params["param_file"] = __file__

from prospect.io import write_results as writer
hfile = name+".h5"
writer.write_hdf5(hfile=hfile,
    run_params=run_params,
    model=model,
    obs=obs,
    sampler=output["sampling"][0],
    optimize_result_list=output["optimization"][0],
    tsample=output["sampling"][1],
    toptimize=output["optimization"][1],
    sps=sps)

If you want to get more information about what might be going wrong when reading the results, try this:

import prospect.io.read_results as reader
filename = "J1401_24Aug31-13.47.h5"
results1, obs1, model1 = reader.results_from(filename)
if model1 is None:
    model2 = reader.get_model(results1)
    print(model2)

Hope this helps.