Using features for Semi-structured documents
This article mentions features used in the processing of Semi-structured documents. Your access to those features depends on your license package and pricing plan.
To learn which features are available to your organization and how to add more, contact your Hyperscience representative.
Each model is only compatible with one Semi-structured layout, but a model is not necessarily compatible with every version of a layout.
Compatibility logic is determined by comparing fields that the model was trained on (e.g. fields in the live layout version at the time of training initiation) to fields in the live layout version, excluding table-column fields.
Compatibility with Live Model
If the model includes the fields that the active layout version also includes, then the model will be compatible. However, if a live layout version includes fields that the model was not trained on, then the model will be incompatible.
If you delete a field in the layout editor and re-add it, it will be considered a new field.
A live model will continue to work even if it's not compatible with the live layout version, however, the model will not be able to find the new fields. You will see a warning on the Model Details page of which fields the model was not trained to find.
Аlways verify layout-version compatibility when switching between model versions
The Live version of a model always uses the most recent layout version, regardless of which layout version it was originally trained with.
This pairing can lead to unexpected behavior, especially if changes were made to the layout after training (e.g. new fields, field-setting updates).
Example: If a model is trained on v3 of layout, and v4 of that layout is created after the training, the model will use v4 of the layout when deployed.
Compatibility with Candidate Models
You will not be able to deploy a new Candidate Model when it is incompatible with the live layout version. The Deploy This Model button will be disabled and you will see an error on which non-table-column fields the model was not trained on.
Note that, with v32+, signatures are compatible with all Semi-structured models. Signature fields can now be trained and are eligible for Field ID automation. To learn more, see Checkboxes and Signatures.
Compatibility with Model Upload
Just like with model deployment, model upload will fail if the candidate model is not compatible with the live layout version. To ensure compatibility when uploading a model from another instance, upload and deploy the layout version first.