Removing a Dataset from a Project

When you remove a dataset from a project, that dataset is no longer used as an input dataset for the project. Removing an input dataset from a project does not delete that dataset from Tamr Core. The dataset remains uploaded and available to add to other projects.

Tip: Removing an input dataset early in the project workflow can result in fewer additional changes to other projects. Assistance from an admin might be needed.

Team members with the admin role can delete datasets from Tamr Core and all projects. See Deleting a Dataset From All Projects.

To remove a dataset from a project:

1. Open a mastering, categorization, or schema mapping project.
2. On the Unified Dataset page, select Show Transformations.
3. Update any transformations that specify use of the dataset you intend to remove.
   Important: There is no need to update the unified dataset at this point but if you do, be sure to run the full pipeline before proceeding to step 4
4. On the Schema Mapping page, filter to the dataset you intend to remove and unmap all of its attributes. See Mapping Unified Attributes.
5. On the Datasets page, select the checkbox for the dataset(s) you want to remove.
6. Choose Remove from project > Confirm.
7. Run Update Unified Dataset and any additional, subsequent jobs as needed for the project.
8. For established projects in which all jobs have been run previously, work with an admin to identify any other projects that rely on this project for an input dataset. For example, if you remove an input dataset from a mastering project, an admin could check for a golden records project for this mastered data. Update jobs might also be needed in each of those "downstream" projects. See Deleting a Dataset From All Projects.