Governed end-to-end analysis pipelines

In eyeprocess: Harmonize Eye-Tracking, Pupillometry, Biometrics, and Psychometric Process Data

knitr::opts_chunk$set(collapse = TRUE, comment = "#>")

The pipeline layer links import, measurement quality, preprocessing, feature construction, modeling, diagnostics, sensitivity analysis, and reporting while preserving the researcher’s declared choices. Pipeline steps are explicit functions with declared dependencies; eyeprocess does not silently choose preprocessing or statistical specifications.

spec <- eye_analysis_spec(blink_correction="linear", pupil_baseline=c(-500,0), fixation_algorithm="ivt", aoi_rule="probabilistic")
p <- eye_analysis_pipeline(list(
  eye_pipeline_step("import", read_fun),
  eye_pipeline_step("quality", quality_fun, requires="import"),
  eye_pipeline_step("model", model_fun, requires="quality")
), spec = spec)
validate_eye_pipeline(p)
r <- run_eye_pipeline(p, context=list(path="study.csv"))
audit_eye_pipeline(r)
plot(p)

eye_targets_manifest() and write_eye_targets_template() provide interoperability scaffolding without pretending arbitrary closures can be losslessly translated into another pipeline engine.

eyeprocess documentation built on Sept. 28, 2026, 5:08 p.m.