BayeSED3 Bayesian SED Analysis Workflow

Agent workflow router. Canonical copy: repo skills/ (sync to ~/.agents/skills/bayesed3/ after edits).

Happy path: Observation → Input catalog → SEDModel → SEDInference.priors_init → (optional set_prior) → inference.run → Results → analyze / compare.

Decision Tree

1. Need an input catalog?
   → Existing .txt, or data-preparation.md (SEDObservation / filters)
2. Build the SED model
   → Default: SEDModel → custom-models.md
   → Shortcut: .galaxy() / .agn() → galaxy-fitting.md / agn-fitting.md
   → Fine control: *Params → same refs + run_test.py
3. Advanced run mode? (AB mag, phot/spec-only, NNLM/RDF/SNR, MultiNest)
   → advanced-run-modes.md
4. priors_init(params); optional set_prior → prior-management.md
5. inference.run(params) → Results
   → Batch/diagnostics: Interface.run → Execution, then BayeSEDResults(outdir)
6. Analyze → results-analysis.md
7. Compare model configurations → model-comparison.md
8. CLI / Launcher (`bayesed3`) / --import → binary-cli.md

Reference Files

File	Open when
references/data-preparation.md	Arrays → catalog, filters, nondetections, spectroscopy
references/custom-models.md	Canonical SEDModel, dust, torus, mixed components
references/galaxy-fitting.md	Factory + low-level galaxy; SSP/SFH/DAL tables
references/agn-fitting.md	Factory + low-level AGN
references/prior-management.md	set_prior, prior types, .iprior
references/advanced-run-modes.md	AB mag, phot/spec-only, NNLM/RDF, MultiNest
references/results-analysis.md	Load Results, stats, plots, GetDist
references/model-comparison.md	Evidence ranking, posterior comparison
references/binary-cli.md	Launcher (bayesed3) and binary flags / --import

Happy-path steps

Each step ends when its Done criterion holds. Follow in order; open a reference only when the decision tree points there.

1. Input catalog — path to a BayeSED .txt catalog (build via data-preparation.md if needed). Done: input_file points at an existing catalog file.

2. SED model — default SEDModel.create_galaxy / create_agn + params.add_* (custom-models.md). Done: params has components attached and save_sample_par=True when posteriors or comparison are required.

3. Priors — SEDInference().priors_init(params); optional set_prior (prior-management.md). Done: priors_init has run for this params (and every set_prior finished before the run).

4. Run — results = inference.run(params) → Results. Batch path: Interface.run → Execution, then BayeSEDResults(outdir, ...). Done: you hold a BayeSEDResults instance (from inference.run or explicit load).

5. Analyze — summary, best-fit, posteriors, evidence (results-analysis.md). Done: print_summary() (or equivalent) succeeded and get_evidence() returned log_evidence / log_evidence_error when evidence was requested.

6. Compare (optional) — separate outdir per model configuration (model-comparison.md). Done: each configuration has Results; evidence or posterior comparison produced if asked.

Skeleton

from bayesed import BayeSEDParams, SEDInference
from bayesed.model import SEDModel

input_file = 'observation/test/gal.txt'  # step 1

galaxy = SEDModel.create_galaxy(
    ssp_model='bc2003_hr_stelib_chab_neb_2000r',
    sfh_type='exponential',
    dal_law='calzetti',
)
params = BayeSEDParams(
    input_type=0,
    input_file=input_file,
    outdir='tests/output_skill_happy_path',
    save_sample_par=True,
)
params.add_galaxy(galaxy)  # step 2

inference = SEDInference()
inference.priors_init(params)  # step 3
# Optional: inference.set_prior('log(age/yr)', min_val=8.5, max_val=9.8, confirm=False)

results = inference.run(params)  # step 4 → Results

results.print_summary()  # step 5
results.plot_bestfit()
results.plot_posterior_free()
evidence = results.get_evidence()  # log_evidence, log_evidence_error
# step 6 optional → model-comparison.md

Branches

• Factory: .galaxy() / .agn() then same steps 3–5 → galaxy/agn refs.
• Mixed / dust / torus: custom-models.md.
• Batch / diagnostics: execution = BayeSEDInterface(...).run(params) then load Results.
• Advanced modes / CLI: advanced-run-modes.md / binary-cli.md (Launcher: bayesed3).

Imports (happy path)

from bayesed import BayeSEDParams, SEDInference, BayeSEDResults
from bayesed.model import SEDModel

Low-level *Params, SEDObservation, BayeSEDExecution, standardize_parameter_names, list_catalog_names: see the reference for that branch.

Output conventions

• Outdirs: tests/output_<name>/ or observation/<testname>/output/
• Posteriors: HDF5; best-fit SED: FITS
• Parameter names: param_name[igroup,id] (e.g. log(age/yr)[0,1]) — offsets in custom-models.md

FAQ

Execution vs Results? inference.run returns Results. BayeSEDInterface.run returns Execution (exit code, paths, timing); load Results from outdir afterward. Analyze posteriors only via Results.

Why priors_init every run? Happy-path convention: load/generate .iprior before fitting. Call it before set_prior / print_priors / validate_priors.

Quick test? inference.run(params, Ntest=2), or Ntest=2 on BayeSEDInterface, or params.configure_multinest(nlive=40).

Evidence keys? Prefer log_evidence / log_evidence_error from get_evidence(). Raw INSlogZ may also appear in the dict when present in HDF5.