Skip to content

API Reference

The everyday names are importable from the top level, e.g. from virgil import OIData, System, PointSource, likelihood_grid.

Everything here is stable except the entries marked Provisional, which may change in any 0.4.x release without a deprecation. See Stability for the list.

  • OIData: observables and their conventions
  • OIFITS: reading and writing OIFITS files
  • AMIGO: AMIGO mixed-DISCO products
  • Models: source models and visibilities
  • Harmonics: azimuthal harmonics of a MultipoleDisk and conversions between their forms
  • Likelihood: likelihoods and numpyro models
  • Gains: calibration gains correlated across channels, marginalized analytically
  • Linear marginalization: the shared algebra for parameters marginalized analytically
  • Observables: OI_FLUX spectra, |V|, T3AMP and differential phases after the visibilities and phases
  • Inference: Laplace and Fisher curvature
  • Grid Fit: grid searches
  • Limits: contrast limits and flux/contrast/Δmag conversions
  • Priors: isotropic inclination and latitude priors, and hierarchical (population) scales
  • Detection: detection statistics for false-alarm rates and ROC curves
  • Spectra: wavelength-dependent fluxes
  • Orbits: Keplerian orbits in virgil's conventions, solved with jaxoplanet
  • Orbit aliases: fit_orbit_aliases, evidences of the period alias bands of an undersampled orbit (provisional)
  • Tempering: the effective-channel factor that tempers an orbit evidence for correlated spectral channels (provisional)
  • Epochs: datasets grouped into epochs, and starting multi-epoch orbit fits (provisional)
  • Radio: ComplexVisData, CalibrationNuisance, complex visibilities from CASA measurement sets with per-observation calibration nuisances (provisional)
  • Orbit Search: score_orbits, candidate orbits scored on all epochs with shared nuisances (provisional)
  • Radio Toeplitz: ToeplitzVisData, the radio χ² of a one-Image fit through precomputed dirty beams (provisional)
  • Radio CLEAN: clean, CLEAN deconvolution of complex visibilities (provisional)
  • Cross-validation: K-fold cross-validation of a hyperparameter (provisional)
  • Short arcs: each night's velocity, the periods its positions and velocities admit, and each alias band's seed orbits (provisional)
  • Fields: Gaussian-process log-brightness priors for pixel images
  • Fitting: fit, maximum a posteriori fits; gauss_newton_mass, a NUTS mass matrix from a fit
  • SVI: variational, a guide fitted to the posterior of numpyro_model with numpyro's SVI
  • Angles: AngleVector, angles sampled as 2-D vectors with no wrap boundary, and von Mises priors in least-squares form
  • Imaging: regularizers and helpers for image reconstruction
  • Scenes: synthetic truth images for testing reconstructions
  • Metrics: image-recovery scores for reconstructions
  • Ensemble: randomized reconstruction ensembles, selected and averaged into a mean image with a per-pixel spread
  • Coverage: synthetic uv coverage and noise for simulations
  • Pipeline: BinaryPipeline, StarPipeline, DiskPipeline, OrbitPipeline and ImagingPipeline, scriptable pipelines that write a run folder
  • Simulate: a scene observed with a template's sampling, and bias tests
  • Plotting: figures
  • Legacy: ImPlaneIA-derived OIFITS tools