Mercurial > repos > goeckslab > multimodal_learner

diff README.md @ 3:25bb80df7c0c draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/goeckslab/gleam.git commit f0daa5846b336584c708d88f6d7f1b5ee8dc3093
author goeckslab
date Sat, 17 Jan 2026 22:53:42 +0000
parents a92f200d296e
children
line wrap: on
line diff
--- a/README.md	Sat Jan 10 16:13:19 2026 +0000
+++ b/README.md	Sat Jan 17 22:53:42 2026 +0000
@@ -12,6 +12,7 @@
 ## Inputs
 - `Training dataset (CSV/TSV)`: includes the label column and any feature columns; image columns should contain file paths that exist in the provided ZIP archives (or absolute paths).
 - Optional `Test dataset (CSV/TSV)`: if omitted, the tool performs train/validation/test splitting or k-fold CV.
+- Optional `Sample ID column`: when provided, related rows are grouped into the same split and cross-validation fold to reduce leakage.
 - Optional `Image archive(s) (ZIP)`: one or more archives containing the image files referenced in the table.
 - Optional overrides: text and image backbones, evaluation metric, quality preset, threshold for binary tasks, and extra hyperparameters (JSON/YAML string or file path).