DU Trainer module enables a Smart Extract model training option through document annotation. Using DU Trainer module, user will have the option to annotate the required fields on the document used for training the model. The primary objective of the module is to onboard various document types and new FOIs at a faster pace. The approach primarily works for SmartExtract engine.
Only those models which have completed the training process and have been published can be used inside DU so that next time when a project is created the desired model (published from DUT) can be loaded for extraction and processing of data.
The process of DU Trainer module consists of:
Creating a project and uploading related document(s)
Annotate the required fields of interest (FOIs) for both headers and tables
Generating Smart Extract models through training
Publishing the model
Related Topics: