Skip to main content
App version: 3.4.4

Projects

A Neptune project typically represents one machine learning task. All of your tracked runs appear in the project, letting other team members explore and analyze experiments through shared views and reports.

Create a project​

Related

Via web app​

In the left nav:

  1. Click the projects menu and select Go to all projects.

  2. Select Create new project.

  3. Enter a name for the project.

    • Must consist of 3–128 characters.
    • Can include any ASCII-encoded characters, except /, \, #, ?, %, :, and 4-byte UTF characters.

  4. Configure the other options as needed and click Create.

Via API​

To create a project via the Python API, use the create_project() function:

from neptune_scale.projects import create_project

create_project(
    name="project-x",
    workspace="team-alpha",  # a workspace with this name must already exist
)

Find project name​

The project path is workspace-name/project-name. To find and copy it:

  1. In the Neptune app, open the project menu ().

  2. From the dropdown, select Settings.

    Project menu in the Neptune app.

Set target project​

The target project is the Neptune project that the runs and metadata are sent to.

As Run() argument​

To specify the Neptune project when creating a run, use the project argument:

from neptune_scale import Run

run = Run(
    run_id="astute-kittiwake-23",
    project="team-alpha/project-x",  # replace with your workspace and project name
)

As environment variable​

By saving the project path to an environment variable, you don't have to specify it in your training code every time you create a run:

Append a line with the export command to your .profile or other shell initialization file:

export NEPTUNE_PROJECT=team-alpha/project-x

For details, see Save credentials as environment variables.