Skip to contents

The default output contains only identifiers and the prespecified primary endpoint. Use `diagnostics = TRUE` to add calibration fields and one nested diagnostic record per row.

Usage

# S3 method for class 'gaze_transport_fit'
tidy(x, diagnostics = FALSE, ...)

Arguments

x

A fitted object returned by [gaze_transport_cv()].

diagnostics

Include candidate/calibration columns and a nested `diagnostics` list-column.

...

Unused.

Value

A tibble. The default has identifiers and `gaze_info_bits` only.