a.anabitarte · Baselines · Published February 8, 2024

HABE_Baseline new

Layout Analysis

Description

Using the GT for text recognition for baselines (deleting header, footer, etc.)
Open in Transkribus
Very low loss2.91% loss

Loss indicates how far the predicted text regions deviate from the ground truth (lower is better). This model achieved 2.91% on its validation set. A loss below 10% generally indicates reliable baseline detection.

Layout detection quality depends heavily on your document's structure. Pages with columns, marginalia, or non-standard layouts may produce different results.

Words33,928
Lines5,515
Training Pages254
Model ID59444