UiPath Data Manager is the tool which must be used to prepare datasets for Training and Evaluation of Document Understanding Machine Learning models.
It is available in 3 deployment methods:
- Data Manager standalone docker container for On Premises. This is Generally Available and fully supported for Production scenarios. Strongly recommended for On Premises.
- Data Manager in AI Center in Automation cloud. This is Generally Available and fully supported for Production scenarios. It has a limitation on size of datasets that can be imported to 1500 images per import. The volume of data is not limited, so multiple imports can be done in succession.
- Data Manager in AI Center On Premises. This is available in Private Preview, it is not supported, and it is recommended only for Trial or Demo scenarios involving datasets of less than 500 images.
Data Manager enables multiple users to perform a variety of operations involved with managing data batches, data preparation and model configuration:
Define and configure the fields to be extracted by an ML model.
Import documents for labeling.
Prelabel documents using a preexisting ML model such as Invoice Extraction or Receipt Extraction provided by UiPath out-of-the-box, or by using a model trained using AI Center.
Export documents in the format expected by the AI Center Training pipelines.
Updated 3 days ago