Create a new document group

Manually define a user-created document group to process documents with an identical layout. For example, invoices from a specific supplier.

先决条件

If you have not done so already, Create a learning instance. After you create a learning instance, you are redirected to the document groups page, where you create a new document group.
Identify the group label. This is the name of the folder in the documents\classify\train directory where documents with an identical layout are sent. You will provide the group label in the steps below.
Identify sample documents (maximum 150) that have an identical layout. You can upload more documents to train the group after the group is created.

Navigate to the learning instance details page and click Create group.
In the Group label field, provide the folder name that contains the sorted documents in the documents\classify\train directory.
Click Browse to upload sample documents to define the layout.
Click Save.

Configure the mechanism to sort incoming documents to the correct group:
- To use the Classifier, run the bot with the Train Classifier action to create the classification model and sub-folders to hold the sorted documents in the documents\classify\train directory. See Using Train Classifier action.
- Alternatively, create RPA bots identify and sort incoming documents using conditional statements, such as string-matching the file name or sender. See the String condition in the Loop package or If package.
Update the following actions:
- Optional: Update the Classify Document action to use Express mode. See step 6 in Using the Classify document action.
- Update the Upload document action to provide the group label. See step 4 in Upload document action.
Upload more documents to train the group. When you are done with training, set the group to production.
After you set the group to production: If you have a bot with the Process Documents action, you must update the action with the group label. See step 7 in Using IQ Bot Process documents action.
注： During the production process, you must not move the learning instances or bots from Staging to Production or vice-versa.