Feinabstimmung

AI Center includes the capability of fine-tuning ML models using data that has been validated by a human using Validation Station.

As your RPA workflow processes documents using an existing ML model, some documents may require human validation using the Present Validation Station activity (available on attended bots or in the browser using Orchestrator Action Center).

Die in der Validation Station generierten validierten Daten können mit der „Machine Learning Extractor Trainer“-Aktivität exportiert und zur Feinabstimmung von ML-Modellen im AI Center verwendet werden.

Es wird nicht empfohlen, ML-Modelle von Grund auf (d. h. das ML-Paket von DocumentUnderstanding) mit Daten aus der Validation Station zu trainieren, es sei denn vorhandene ML-Modelle (einschließlich out-of-the-box Modelle) sollen fein abgestimmt werden.

For the detailed steps involved in fine-tuning an ML model, check the Import Documents section of the Document Manager documentation.

Wichtig:

It if often wrongly assumed that the way to use Validation Station data is to retrain the previous model version iteratively, so the current batch is used to train package X.1 to obtain X.2. Then the next batch trains on X.2 to obtain X.3 and so on. This is the wrong way to use the product. Each Validation Station batch needs to be imported into the same Document Manager session as the original manually labeled data making a larger dataset, which must be used to train always on the X.0 ML Package version.