Posterior pupil trajectories and declared estimands¶
Python-facing port of
pupil-trajectories-and-estimands.Rmdfrom the frozen R gp3bayes 0.5.0 reference. The statistical and governance framing below follows the canonical vignette; executable Python workflows use the mapped APIs listed later.
Lightweight posterior-draw contract¶
The estimand layer operates on posterior prediction draws. For documentation and tests, a deterministic draw matrix can be used without compiling Stan.
Trajectory uncertainty¶
Pointwise intervals describe uncertainty at each grid value. A finite-grid simultaneous band can be requested explicitly; it is qualified as a grid-based posterior band rather than a universal continuous-time guarantee.
Declared contrasts and windows¶
Windows are supplied by the analyst. The package does not search across time for the most favourable interval and relabel it confirmatory. Peak and peak-latency summaries propagate posterior-draw uncertainty within the declared evaluation grid.
Python API mapping¶
gp3bayespy.as_pupil_prediction_drawsgp3bayespy.estimate_pupil_aucgp3bayespy.estimate_pupil_peakgp3bayespy.estimate_pupil_peak_latencygp3bayespy.estimate_pupil_trajectorygp3bayespy.estimate_pupil_windowgp3bayespy.pupil_condition_contrastgp3bayespy.pupil_trajectory_table
Python usage¶
import gp3bayespy as gp
# All functions listed above are available from the package root.
# Use help(gp.<function>) or the API reference for the exact Python signature.
An executable workflow for this family is included in ../../examples/pupil_workflow.py.