Prerequisites
Training is available only if ABBYY FlexiCapture can reliably identify the company by finding the corresponding record in a database. If you have no databases but still want to use field training, you can accumulate company information by adding records to your data sets while capturing documents. For more information, see Looking up vendors and business units in the database.Train the program
1
Collect and feed a batch of documents
Collect a batch of documents (for example, invoices processed within the past month) and start feeding them to the program. For more information, see How to capture invoices.
2
Let the program recognize and check the documents
Once fed to the program, the documents are automatically recognized (only if the Recognize added images automatically option is enabled on the Document Processing tab of the Options dialog box — to open it, click Tools → Options), and the data is checked by validation rules.
3
Open a document that needs correction
If a recognized document’s status is other than Valid, or you have other reasons to believe the program failed to detect some fields, open the document in the document editor.
4
Confirm the company
Review the document form. The Vendor/Issuer group of fields must be filled out correctly.Training is done independently for each document variant, and documents from the same company are treated as the same variant. If the program fails to identify the issuing company, either select it from your company database or key it in from the document image and save it by clicking Save.Depending on your project’s settings, you may also have to specify the issuing company’s unique ID to use its document for training. To do this, type the unique ID in the VATID field (this field may have a different name in some projects, depending on the country). The VATID is a unique identification number assigned to companies for tax purposes.If documents from the same company have widely varying layouts, use the clustering feature. For more information, see Training with clustering.
5
Adjust the field regions
Training succeeds only if all field regions are identified correctly, so make sure the regions match the actual field locations on the image. In the image window of the document editor, adjust the regions or draw regions for the fields the program failed to detect. For more information about marking up line items, see Training line items.You can change a field’s region in several ways:
- Place the mouse pointer in a field on the data form, find the corresponding region on the image (highlighted in blue), and click it (or draw a rectangle with the mouse).
- Place the mouse pointer over a region on the image (highlighted in blue), click it (or draw the region), and then select the corresponding field from the drop-down list that opens.
- Adjust a region by moving its boundaries with the mouse.
- Delete an incorrectly located region: place the pointer over its rectangle and, when a red cross appears in the top-right corner, click the cross. Then create the correct region for the field.
- On the data form, start typing a value into a field. A drop-down list shows the words captured from the image that resemble what you are typing. Select the right word, and the position of that word on the image becomes the field’s region.
The program is trained on all the document’s fields, not just those whose regions you drew or adjusted.
6
Repeat for the remaining documents
Open the next document and repeat the review and markup steps.
7
Let training run
To initiate training, a training batch must contain at least one document. If clustering is used, a separate FlexiLayout is created for each cluster; otherwise, a FlexiLayout is created for each company. For more information, see Training with clustering.
8
Review the training results
The program tests the trained FlexiLayout variant by applying it to all documents in the training batch and comparing the results with the markup you adjusted. If the trained FlexiLayout delivers better results than its earlier version, it is used the next time you recognize documents belonging to this document variant.