A fingerprint of the rows an analysis was run on

Factors are hashed as their values and not as their level sets, because mm_model() drops a level no fitted row carries and a caller re-deriving the same frame has not. Numeric columns are hashed as they are, so nothing is rounded on the way in.

Usage

mm_sample_digest(project, frame)

Arguments

project

A mm_project().

frame

The rows.

Value

A SHA-256 string. Not a copy of the data, and not invertible.