Note
Go to the end to download the full example code.
Recipe introspection and SED morphology comparison¶
Call tengri.list_recipes() to see the shipped menu with SSP requirements
(bare-stellar, wNE, or any) and tengri.describe_recipe(name) to fetch
a recipe’s docstring. Three models showcase the morphological diversity:
star-forming (DPL+Cue nebular, free z to 6), quiescent at z=0.05 (dexp,
lower dust ceiling), and AGN-panchromatic (full composite, z to 6).
All require bare-stellar SSP (Cue backend).
Each SED is normalized to its rest-frame value at 5500 Å to overlay the spectral shapes and highlight morphological differences across recipe types.

=== Available Recipes ===
name short_doc ssp_requirement data
─────────────────────── ─────────────────────────────────────────────────────────────────────── ──────────────────────────────────────────────────────────────────────────────── ──────────────────────────────────────────────────────────────────────
agn_panchromatic Recipe for AGN-dominated galaxies with multi-wavelength data. bare-stellar (Cue nebular backend; see star_forming_photometry for details). ready
composable_agn Fully composable AGN — all slots switchable on committed data. bare-stellar (Cue nebular backend; see star_forming_photometry for details). ready
dust_demo Recipe for forward-only dust attenuation gallery sweeps. wNE (with-nebular-emission), i.e. load_ssp("prsc_miles_chabrier_wNE"). Uses t... ready
high_z Recipe for high-redshift galaxies (z > 3.5, young starburst). with-nebular-emission (wNE) SSP grids — the ssp nebular backend reads line an... ready
mock_recovery_minimal Recipe for mock data recovery and benchmarking (minimal model). any. Nebular emission is disabled, so this recipe works with both bare-stella... ready
photoz Recipe for photometric-redshift fits (redshift-unconstrained surveys). any (nebular emission is off). ready
quiescent_z0 Recipe for quiescent galaxies at low redshift (z ~ 0.05). bare-stellar (Cue nebular backend; see star_forming_photometry for details). ready
star_forming_photometry Recipe for star-forming galaxies with broadband photometry (0 < z < 6). bare-stellar (e.g., fsps_prsc_miles_chabrier.h5, fsps_mist_c3k_a_chabrier.h5)... ready
stochastic_sfh_jwst Recipe for JWST high-redshift galaxies with stochastic SFH. bare-stellar (Cue nebular backend; see star_forming_photometry for details). ready
unified_agn Synthesizer UnifiedAGN reproduction (disc + NLR/BLR grids). bare-stellar (Cue nebular backend; see star_forming_photometry for details). ... needs Synthesizer AGN NLR grid (synthesizer-download --agn-test-grids)
[10 results — recipe]
Use: tengri.describe('agn_panchromatic') → recipes.agn_panchromatic() → SEDModel.build(ssp_data=ssp, **recipe)
/tengri/src/tengri/forward/sed_model.py:8666: WildcardPartialFreeWarning: 'all_params: FREE' freed 2 of 3 parameters in group 'dust'. These have no declared prior, only Fixed defaults, so they stay pinned:
dust_f_obscuration
The fit will run with that physics held constant. Pass explicit priors for the ones you meant to vary, e.g. dust={'f_obscuration': Uniform(lo, hi)}, or filter WildcardPartialFreeWarning if this is deliberate.
spec = parse_groups(**groups)
/tengri/src/tengri/forward/orchestrator.py:951: SFHBeforeBigBangWarning: Star formation history forms 5% of its stellar mass before the Big Bang at z=5.18 (cosmic age 1.12 Gyr). That mass is truncated, so the prediction does not reflect the requested SFH — bound the SFH age parameter or the redshift to keep star formation within cosmic time.
state = component.apply(state, sliced, ssp_data=ssp_data, template_data=template_data)
/tengri/src/tengri/forward/sed_model.py:8666: WildcardPartialFreeWarning: 'all_params: FREE' freed 2 of 3 parameters in group 'sfh'. These have no declared prior, only Fixed defaults, so they stay pinned:
sfh_dexp_start_gyr
The fit will run with that physics held constant. Pass explicit priors for the ones you meant to vary, e.g. sfh={'dexp_start_gyr': Uniform(lo, hi)}, or filter WildcardPartialFreeWarning if this is deliberate.
spec = parse_groups(**groups)
/tengri/src/tengri/forward/sed_model.py:8666: WildcardPartialFreeWarning: 'all_params: FREE' freed 2 of 3 parameters in group 'dust'. These have no declared prior, only Fixed defaults, so they stay pinned:
dust_f_obscuration
The fit will run with that physics held constant. Pass explicit priors for the ones you meant to vary, e.g. dust={'f_obscuration': Uniform(lo, hi)}, or filter WildcardPartialFreeWarning if this is deliberate.
spec = parse_groups(**groups)
/tengri/src/tengri/forward/orchestrator.py:951: SFHBeforeBigBangWarning: Star formation history forms 5% of its stellar mass before the Big Bang at z=5.18 (cosmic age 1.12 Gyr). That mass is truncated, so the prediction does not reflect the requested SFH — bound the SFH age parameter or the redshift to keep star formation within cosmic time.
state = component.apply(state, sliced, ssp_data=ssp_data, template_data=template_data)
import warnings
import jax
import matplotlib.pyplot as plt
import numpy as np
import tengri
from tengri import recipes
from tengri.plot import setup_style
setup_style()
warnings.filterwarnings("ignore", message=".*BakedInBackend.*")
warnings.filterwarnings("ignore", message=".*wNE.*")
C_AA_PER_S = 2.998e18
# Load bare-stellar SSP (required for Cue nebular backend in all recipes)
ssp = tengri.load_ssp("fsps_prsc_miles_chabrier")
# Print the recipe menu to stdout for documentation
print("\n=== Available Recipes ===")
recipe_table = tengri.list_recipes()
print(recipe_table)
print("\n")
# Define the three recipes to showcase
RECIPE_CONFIGS = [
("star_forming_photometry", recipes.star_forming_photometry, "#3377cc"),
("quiescent_z0", recipes.quiescent_z0, "#cc3333"),
("agn_panchromatic", recipes.agn_panchromatic, "#9933aa"),
]
# Compute and plot rest-frame SEDs
fig, ax_seds = plt.subplots(figsize=(8, 5.5))
plotted = 0
first_failure: Exception | None = None
for name, recipe_fn, color in RECIPE_CONFIGS:
try:
model = tengri.SEDModel.build(ssp_data=ssp, **recipe_fn())
except Exception as e:
if first_failure is None:
first_failure = e
continue
p = dict(model.spec.sample(jax.random.PRNGKey(0)))
out = model.predict(p)
wave = np.asarray(model.wavelengths)
nu_l_nu = C_AA_PER_S / wave * np.asarray(out.rest_sed())
# Normalize to the value at 5500 Å rest to show morphology shapes
i_norm = int(np.argmin(np.abs(wave - 5500.0)))
if nu_l_nu[i_norm] > 0:
nu_l_nu_norm = nu_l_nu / nu_l_nu[i_norm]
else:
nu_l_nu_norm = nu_l_nu
ax_seds.semilogx(wave, nu_l_nu_norm, color=color, lw=1.4, label=name)
plotted += 1
# Error check: ensure at least one recipe built
if plotted == 0:
raise RuntimeError(
f"none of the {len(RECIPE_CONFIGS)} recipes built, so the SED panel is "
f"empty. First failure: {type(first_failure).__name__}: {first_failure}"
) from first_failure
ax_seds.set(
xlim=(700, 5e6),
xlabel=r"Rest-frame wavelength $\lambda$ [$\mathrm{\AA}$]",
ylabel=r"$\nu L_\nu$ / $\nu L_\nu(5500\,\mathrm{\AA})$ (normalized shape)",
)
ax_seds.legend(frameon=False, fontsize=10, loc="lower left")
ax_seds.grid(True, alpha=0.2, which="both")
fig.tight_layout()
plt.savefig("plot_recipe_introspection_tour.png", dpi=150, bbox_inches="tight")
Total running time of the script: (0 minutes 33.897 seconds)