Project Properties

Project properties include:

• Type (required): One of Categorization, Mastering, Golden Records, Schema Mapping, or Enrichment. If you are creating an enrichment project, see Creating an Enrichment Project for instructions.
• Name (required): A project's name must be unique. Project names cannot include the characters /, \, or :, a leading ., or leading or trailing white spaces.
  Note: Avoid using other special characters in project names. Project names that include the characters ?, #, or % can cause errors later in the project workflow.
• External Id: A unique project ID by which to reference the project from an external client, such as Tamr Python Client.
  On project creation, the external ID defaults to a generated unique value. You can later update the external ID using the project endpoint (Update a Project).

Creating a Project

To create a project:

1. Navigate to the home page.
2. Select Create New Project, and then choose a project type.
3. In the Project Name text box, enter the project name.
4. (Optional) Further configure your project:

• All project types include a Description field which can be used to provide more details about your project.
• Mastering and categorization projects include an Enable Transformations checkbox. If selected, a Unified Dataset page for transforming data is included in the project.
• Categorization projects include a field to input the type of data you are categorizing. Common values include "parts", "products", or "transactions". Tamr uses this value to customize the UI.
• Mastering projects include a field to input the type of data you are mastering. Common values include "suppliers", "customers", or "employees". Tamr uses this value to customize the UI. These projects also include a field to customize the term for potential duplicates. Most commonly, duplicates are called "pairs".
• Note: You can enable the learned pairs feature for a project by setting a value for Maximum Inferred Pairs later in the project workflow, after you run Update Unified Dataset for the first time. See learned pairs for more.

5. Select Create.
6. If you are creating a golden records project, select the mastering project with the record clusters that you want to group into golden records.
7. To ensure that users in your group have access to this project, attach a policy to the project.