Page management
This section only applies if you want to process files that consist of multiple documents, e.g. a PDF file of a loan request that consists of different document types such as wage slips, quotations or a purchase order.
To train the page management model you will have to split the files that have been uploaded into separate documents and choose the right document type and language for each document.
To label training data for page management, go to the uploads page in the training module. Click on an upload, then click on the "Page management" section in the details:
You will arrive upon the following screen:
The screen consists of the following sections, going from left to right:
- Upload with files
- Files - a list of the different uploaded files with per file the different pages.
- Document creation - the different documents that have been created. This is initially empty because no documents have been created yet
- Preview - preview of the currently selected page or document.
You can then create documents by dragging page by page, or multiple pages at once to the middle panel.
Selecting pages for a new document
When clicking a page it will be highlighted, and the selected page(s) can be dragged into a document in the documents section.
Clicking on the eye icon at the bottom of a page allows you to preview the page in the preview section without selecting the page for dragging.
To create a new document, drag and drop the selected pages to the box with "+ Add document".
A random name for the newly created document will be proposed, which can be changed to whichever name you like. Before you can save the documents, you will need to specify the document type and the language.
Duplicating documents
You can simply repeat the same steps as explained above. You can drag the same selected pages to the section "+ Add document".
Deleting page(s) & document(s)
You can delete pages & documents when you accidentally made a mistake. Below you can find the different ways of doing that:
- Delete a document
- Hovering over a page allows you to delete the page
- Selecting pages in a document and dragging them to the uploaded files section
Notice that deleting all pages from a document doesn't delete the document. Make sure to delete the document if it is not needed anymore.
Move page(s) to another existing document
You can move pages from one document to another, by selecting them and dragging them. A line will show up to indicate the place where the pages will be dropped, making it easier to correctly order the pages.
Move page(s) to a new document
You can move one or more selected pages from a document to a new document by dragging the pages to the "Add document" field.
Undo page(s) selection
You can undo a page selection by clicking on te page again to deselect it, it won't be highlighted anymore.
Preview of a page
The right pane shows you the preview of the selected page or document. You can zoom in and out.
Saving documents
When you are done with assigning pages to the correct documents, click "Save documents". The number between brackets indicates how many documents you have created. Note that you cannot save the documents if they do not all have a name, a document type and a language. It is not necessary to assign all pages to a document, you can leave irrelevant pages (such as pages without any text) unassigned.