empyrean.PropagationResult

class PropagationResult(states, events, sensitivity=None, tagged_covariance=None, mixtures=<factory>)

Bases: object

Result of orbit propagation.

Parameters:

• states (CartesianOrbits)

• events (Events)

• sensitivity (StateSensitivities | None)

• tagged_covariance (TaggedCovariances | None)

• mixtures (MixtureChains)

states

Propagated Cartesian states with optional covariance.

Type:

CartesianOrbits

events

All detected dynamical events, grouped by type.

Type:

Events

sensitivity

Flat per-(orbit, epoch) sensitivity table — STMs and optional STTs. None when the propagation method did not produce sensitivities (Monte Carlo / SigmaPoint).

Type:

StateSensitivities, optional

tagged_covariance

Flat per-(orbit, epoch) provenance-tagged, resolved-kind covariance readback — the honest covariance that distinguishes a second-order close-approach ellipsoid from the bare linear Φ Σ₀ Φᵀ mapping on states. None unless propagate() was called with tagged_covariance=True. Use tagged_covariance_series() for the per-epoch view of one orbit.

Type:

TaggedCovariances, optional

mixtures

Flat per-(orbit, CA epoch, component) readback of the AGM mixture components the engine retained — the mixture itself, not its moment collapse. Empty (zero rows) when nothing split, which is every FIRST_ORDER run. Always populated; no opt-in flag, because the components ride along with the result at no extra engine cost. See mixtures for the four limits on what is retained.

Type:

MixtureChains

Methods

__init__(states, events[, sensitivity, ...])
from_dir(path)	Load a propagation result written by to_dir().
mixture_chains(orbit_index)	One orbit's retained mixture components, grouped by CA epoch.
tagged_covariance_series(orbit_index)	Per-epoch tagged covariance for one orbit, as dataclasses.
to_dir(path)	Write a propagation result to a directory of Parquet files.

Attributes

sensitivity
tagged_covariance
states
events
mixtures

states: CartesianOrbits

events: Events

sensitivity: StateSensitivities | None = None

tagged_covariance: TaggedCovariances | None = None

mixtures: MixtureChains

mixture_chains(orbit_index)

One orbit’s retained mixture components, grouped by CA epoch.

Thin accessor over to_chains(), mirroring tagged_covariance_series()’s shape. Returns an empty list for an orbit that retained nothing — a real answer, not an error.

Parameters:

orbit_index (int) – Zero-based index into the input orbits (orbit-major output order).

Return type:

list[list[MixtureComponent]]

tagged_covariance_series(orbit_index)

Per-epoch tagged covariance for one orbit, as dataclasses.

Yields one TaggedCovariance per output epoch of the orbit at orbit_index (orbit-major, matching the input order), with each matrix re-materialized as a contiguous (6, 6) array and the provenance enums decoded.

Parameters:

orbit_index (int) – Zero-based index into the input orbits (orbit-major output order).

Return type:

list[TaggedCovariance]

Returns:

list[TaggedCovariance] – The orbit’s tagged covariance at every output epoch.

Raises:

ValueError – If the result carries no tagged covariance (propagate was not called with tagged_covariance=True), or orbit_index is out of range.

to_dir(path)

Write a propagation result to a directory of Parquet files.

Return type:

None

Parameters:

path (str)

classmethod from_dir(path)

Load a propagation result written by to_dir().

Return type:

PropagationResult

Parameters:

path (str)