Transkribus Community · PyLaia · Published June 23, 2023
The English Eagle
Text Recognition
Description
"The English Eagle" is a robust AI model for Transkribus, rooted in the English Handwriting M3 model, which not only features Bentham's handwriting but also integrates a broad range of other sources. These sources span several centuries, from the 17th to the 20th, contributing to the model's robustness and broad application potential. The continuous and collective efforts of the Transkribus community further enhance the model. They provide ongoing data contributions, facilitating periodic updates that ensure the model's constant evolution, improvement, and adaptability to diverse and changing research requirements.
Please avoid using this model as a basemodel when training custom models, or with the language model or SmartSearch option, as it will not work properly due to its size.
Try this model
Use this modelOpen in Transkribus
Low error rate8.3% CER
Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 8.3% on its validation set. As a rule of thumb, a CER below 10% is considered good for most handwritten material. This is a larger model trained on diverse material, which generally makes it more robust across different handwriting styles. That said, larger training sets also make it harder to push the CER down further.
Measured on the model's own validation data. Results on your documents may differ depending on handwriting style, document condition, language, and how closely your material resembles the training data.
Words3,822,933
Lines558,632
Training Pages17,285
Model ID53042