Skip to main content
After you publish your Document Definition, close the Document Definition dialog box, go to the Field Extraction Training Batches section, and create a training batch.

​
Create a training batch

1

Create the batch

Click File → New Batch. Select the Document Definition you created and the section you configured fields for, and click OK.
2

Choose the training variant

In the Look up Variant for Training Batch window, select the variant to use for training.
3

Mark it as an NLP batch

Select the new batch and select NLP batch, or click Field extraction training → NLP batch.
Screenshot of the Field Extraction Training Batches view in ABBYY FlexiCapture, with a batch selected and its shortcut menu open and the NLP batch option checked.

​
Load the training documents

1

Open the batch

Double-click the batch you created.
2

Load images

Click File → Load Images. Click Image Processing Settings, select One document per file, and click OK, and then choose the documents to train on.
3

Match the Document Definition

Select the loaded documents and click Recognition → Match Document Definition (or right-click the selection and click Match Document Definition), and then choose the Document Definition.
The quality of a trained NLP model depends on the number of training documents and the quality of their markup:
  • Mark up every field described by the Document Definition in the training documents.
  • Include 100 to 500 documents in each training batch. This range lets FlexiCapture choose the best parameters without slowing training.

​
Mark up the documents

Mark up the fields on each document so the model knows where to look for entities.
1

Open a document

Double-click a document to open it.
2

Mark up each field

Select a field to extract, and then either choose its value on the document or draw a rectangle around it. Repeat for each field.
3

Move to the next document

Click the Next document button, and repeat for the remaining documents.
4

Save

Save the changes.
After marking up all documents, return to the Field Extraction Training Batches view, right-click the batch, and click Train. After training, the model is ready to use. To disable or delete training results, right-click the training batch and select Disabled or Delete on the shortcut menu. To use a trained NLP model in another project, import the training batch and its Document Definition into that project.