Data management is Model Studio's unified management feature for training sets and evaluation sets, supporting dataset creation, import, version management, publishing and deletion, as well as data cleansing and augmentation through data flows.
Feature overview
Alibaba Cloud Model Studio Data Management feature provides unified management of all large model-related datasets in your workspace. Go to the Model Studio console · Data Management page. The top of the page provides two tabs: Datasets and Data Flows:
- Datasets tab: Displays all datasets in a list, including dataset name, type, latest version, data volume, import status, publish status, storage location, import method, and operation columns. The upper-right corner provides a Create Dataset button. The operation column supports adding versions, publishing, and deleting.
- Data Flows tab: Used for data cleansing and data augmentation, with a visual data flow canvas for building processing pipelines. See Data cleansing and augmentation.