Description
The Marginalia Monarch is an AI model trained on 5,200 pages with over 8,000 instances of marginalia. This model is designed to detect and analyze handwritten notes and annotations in the margins of historical documents. It is a useful tool for researchers and archivists looking to study these often overlooked elements in historical texts. While effective, users are encouraged to test the model on their own datasets to evaluate its performance and suitability for specific projects.
Open in Transkribus
Good precision68.47% MaP
Mean Average Precision (MaP) measures how accurately the model detects field regions (higher is better). This model scored 68.47% on its validation set. MaP is harder to compare across models than CER, because the score depends heavily on how many distinct region types the model must distinguish. A model detecting a handful of simple fields will naturally score higher than one trained to recognise many fine-grained regions, even if both perform well in practice.
This score reflects performance on the model's own validation data. Your results will depend on how closely your documents match the training material and the complexity of the structures you need to detect.
Words1,337,905
Lines276,390
Training Pages5,291
Model ID159405