Creating an analytics project (Watson Studio and Watson Knowledge Catalog)

You create an analytics project to work with data and other assets to achieve a particular goal, such as building a model or integrating data. You can create an empty project, create a project from file by importing existing assets from an exported project file on your system, or create a project integrated with Git.

Required permissions
You must have one of these user permissions to create a project:

• Manage projects
• Create projects

Your project resources can include data, collaborators, tools, operational assets that run code, like notebooks and models, and other types of assets.

To create a project:

1. From the navigation menu, choose Projects > All projects.
2. Click New project.

3. Choose whether to create:

   • An empty project
   • A project from file by importing a project ZIP file
   • A project integrated with a Git repository

4. If you create an empty project, you have these project configuration options:

   • Mark the project as sensitive. The project has a sensitive tag and project collaborators can't move data assets out of the project. You can enable or disable this option at any time on the project Settings page.
   • Log project activity. Detailed project activity is tracked in a full activities log that you can download. You can enable or disable this option at any time on the project Settings page.

5. If you create a project from file, select the project ZIP file on your local system to upload. See Import from a local file.

   You have these project configuration options:

   • Mark the project as sensitive. The project has a sensitive tag and project collaborators can't move data assets out of the project. You can enable or disable this option at any time on the project Settings page.
   • Log project activity. Detailed project activity is tracked in a full activities log that you can download. You can enable or disable this option at any time on the project Settings page.

6. If you create a project with a Git repository, select the integration type:

   • Default Git integration for true Git-based version control when working on files. You need to:

     • Configure access to a Git repository. This enables you to work in your clone of a branch from the remote repository associated with the project.

   • Deprecated Git integration for Git integration with locking. Although you can still create and use projects with this integration type, this project creation option is deprecated. You have these project configuration options:

     1. Either create a new empty project or import an existing project from a Git repository.
     2. Configure access to a Git repository.
     3. Mark the project as sensitive. You can enable or disable this option at any time on the projects Settings page.
     4. Log project activity. You can enable or disable this option at any time on the project Settings page.
     5. Designate the JupyterLab IDE as the only editor tool for notebooks and scripts by selection. You can enable this option at any time on the project Settings page.

7. Click Create. You can start adding resources if your project is empty, or begin working with the assets and files you imported.

   Before you begin working with the imported assets, you should check for missing credentials, for example in notebooks and data connections, to enable successful relinking between the assets.

Next steps

• Create tokens for collaboration in projects with Git integration
• Add collaborators
• Add data
• Choose a tool

Parent topic: Projects