IQ Bot 11.x: Preview extracted data

After completing field and table mapping, click the See Extraction Results button to view OCR and extracted results to verify the accuracy of the training.

The extraction result displays information that the Validator view would display and lets you do the following:
  • See the listed fields and values. Display an indication if value was successfully extracted, if there was a validation error, or lack of OCR confidence.
  • See all table headers and the values for each row, also indicating if there was an error on validation or lack of confidence.

At this point, you can export the preview data to a CSV file for further analysis by clicking on the Export to CSV link at the top right header area. Alternatively, click Back to Training button to continue with the training. Once all fields and tables complete the training and pass validation, a green check mark displays next to the group title in the middle panel.

重要： A training document can pass in spite of an error flagged by the validation pattern. Sometimes we want a validation pattern to flag an error, so it would not pass specific documents at a later stage in production; instead enter them in the manual validation queue.
After training the documents groups, choose Save and close. At this point choose any of the following:
  • Choose Cancel to stay on the current training.
  • Choose Save to save the training.
  • Choose Save and send to production to send the learning instance to the Production environment.

Preview any time during or after the document training, or after completing the mapping of fields and tables for each document group. The See Extraction Results button is available for a current document that lets you see the extracted results with the mapped fields and table columns. Errors, if any, are visible and highlighted within a red box. Hover on the error to know more about the error, and correct it.

Click the Back to Training button to return to the training.

重要： If you upload 30 documents and all documents are classified into one group, you can see only one document for each unique layout.