management API reference

API reference for the management package and its functions.

The management API lets you perform administration tasks related to your workspace and projects.

from neptune import management

The following functions are restricted to project owners (or workspace admins):

• add_project_member()
• remove_project_service_account()

Service accounts cannot use the following functions:

• delete_project()
• invite_to_workspace()

Otherwise, service accounts can perform project actions according to their permissions.

get_project_list()

Get a list of projects you have access to.

Parameters

To keep your token secure, avoid placing it in the source code. Instead, save it as an environment variable.

Returns

List with project names in the formworkspace-name/project-name.

Example

>>> from neptune import management
>>> projects = management.get_project_list()
>>> print(projects)
['jackie/sandbox', 'ml-team/classification']

create_project()

Creates a new project in a workspace.

On older project-based subscription plans, the projects quota is managed by your workspace admin.

If the limit of projects is reached, you need to upgrade your subscription or archive another project first.

Parameters

If you leave out the workspace argument, include the workspace name here instead, in the form workspace-name/project-name. For example, ml-team/named-entity-recognition.

If you leave it out, Neptune generates a key for you based on the project name.

The level of visibility to set for the project. Users with access effectively become project owners.

`priv` : private
`workspace` : accessible to all workspace members
`pub` : public

Note: On plans without project-level access control, you need to set the value to "workspace" or "pub".

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Returns

The name of the created project.

Examples

>>> from neptune import management
>>> management.create_project(
... workspace="ml-org",
... name="named-entity-recognition",
... key="NER",
... visibility="priv",
... )
'ml-org/named-entity-recognition'

Plan note

On plans without project-level access control, private projects (accessible only to some workspace members) are not available. You need to set the privacy to "workspace" or "public".

>>> from neptune import management
>>> management.create_project(
... workspace="ml-team",
... name="classification",
... key="CLS",
... visibility="workspace",
... )
'ml-team/classification'

delete_project()

Deletes a project from a workspace.

Only workspace admins can use this function.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toprojectas of neptune-client0.16.16.

Example

>>> from neptune import management
>>> management.delete_project(project="ml-team/classification")

add_project_member()

Adds a user to a project.

Only project owners or workspace admins can use this function.

Parameters

Level of permissions the user should have in the project.

• viewer : can view project content and members.
• contributor : can view and edit project content, and view members
• owner : can view and edit project content and members

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toprojectas of neptune-client0.16.16.

Example

>>> from neptune import management
>>> management.add_project_member(
... project="ml-team/classification", username="jackie", role="contributor"
... )

Related

User roles overview

get_project_member_list()

Lists the members of a project.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toprojectas of neptune-client0.16.16.

Returns

Dictionary with usernames as keys and project roles as values.

Example

>>> from neptune import management
>>> management.get_project_member_list(project="ml-team/classification")
{'johnny': 'owner', 'jackie': 'contributor', 'janus': 'viewer'}

remove_project_member()

Removes a user from a project.

Only project owners or workspace admins can use this function.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toprojectas of neptune-client0.16.16.

Example

>>> from neptune import management
>>> management.remove_project_member(
... project="ml-team/classification",
... username="janus",
... )

get_workspace_member_list()

Lists the members of a workspace.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toworkspaceas of neptune-client0.16.16.

Returns

Dictionary with usernames as keys and workspace roles as values.

Example

>>> from neptune import management
>>> management.get_workspace_member_list(workspace="ml-team")
{'johnny': 'admin', 'jackie': 'member', 'janus': 'member'}

get_workspace_service_account_list()

Lists the service accounts of a workspace.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toworkspaceas of neptune-client0.16.16.

Returns

Dictionary with account names as keys and workspace roles as values.

Note

Service accounts can only have the "member" role in a workspace.

Example

>>> from neptune import management
>>> management.get_workspace_service_account_list(workspace="ml-team")
{'cicd@ml-team': 'member', 'test@ml-team': 'member'}

get_project_service_account_list()

Lists the service accounts assigned to a project.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toprojectas of neptune-client0.16.16.

Returns

Dictionary with account names as keys and project roles (owner,contributor, orviewer) as values.

Example

>>> from neptune import management
>>> management.get_project_service_account_list(project="ml-team/classification")
{'cicd@ml-team': 'owner', 'test@ml-team': 'viewer'}

add_project_service_account()

Adds a service account to a Neptune project.

Only project owners or workspace admins can use this function.

Parameters

Level of permissions the service account should have in the project.

• viewer : can view project content and members.
• contributor : can view and edit project content, and view members
• owner : can view and edit project content and members

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toprojectas of neptune-client0.16.16.

Example

>>> from neptune import management
>>> management.add_project_service_account(
... project="ml-team/classification",
... service_account_name="cicd@ml-team",
... role="contributor",
... )

remove_project_service_account()

Removes a service account from a Neptune project.

Only project owners or workspace admins can use this function.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Thenameparameter has been changed toprojectas of neptune-client0.16.16.

Example

>>> from neptune import management
>>> management.remove_project_service_account(
... project="ml-team/classification", service_account_name="cicd@ml-team"
... )

invite_to_workspace()

Invites a user to a workspace.

Only workspace admins can use this function.

Supply either theusernameortheemailargument (not both).

Parameters

If you provide this, leave out theemailargument.

If you provide this, leave out theusernameargument.

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Level of permissions the user should have in the workspace.

• "member" : can create projects and access projects with "workspace" privacy.
• "admin" : can manage users, projects, and the subscription.

For more, seeUser roles overview.

Examples

>>> from neptune import management
>>> management.invite_to_workspace(
... username="jackie",
... workspace="ml-team",
... role="admin",
... )

You can also use theWorkspaceMemberRoleenum values to specify the role:

• Administrator: WorkspaceMemberRole.ADMIN
• Member: WorkspaceMemberRole.MEMBER

>>> from neptune import management
>>> from management import WorkspaceMemberRole
>>> management.invite_to_workspace(
... username="jackie",
... workspace="ml-team",
... role=WorkspaceMemberRole.ADMIN,
... )

Related

User roles overview

trash_objects()

Moves one or more Neptune objects to the project trash.

Trashed objects are retained until the trash is emptied, either through the Neptune app or with thedelete_objects_from_trash()function.

The account that performs the trashing must have at least contributor permissions.

Note for model objects

Trashing aModelobject also trashes all of itsModelVersionobjects.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

The Neptune ID is a unique identifier for the run. TheExperimentstab displays it in the leftmost column.

In the run structure, the ID is stored in the system namespace (sys).

• If the run is active, you can obtain its ID withrun["sys/id"].fetch(). For example: >>>run=neptune.init_run()...>>>run["sys/id"].fetch()'CLS-26'
• If you set a custom run ID, it's stored atsys/custom_run_id: >>>run["sys/custom_run_id"].fetch()'vigilant-puffin-20bt9'

If the run is active, you can obtain its ID withrun["sys/id"].fetch(). For example:

>>> run = neptune.init_run()
...
>>> run["sys/id"].fetch()
'CLS-26'

If you set a custom run ID, it's stored atsys/custom_run_id:

>>> run["sys/custom_run_id"].fetch()
'vigilant-puffin-20bt9'

Examples

Import the management package and initialize the project to trash objects from:

import neptune
from neptune import management

project_name = "workspace-name/project-name" # (1)!
project = neptune.init_project(project=project_name, mode="read-only")

1. The full project name. For example,"ml-team/classification". To find the project string in the Neptune app, in theProject metadatasection, clickAdd new field.

The full project name. For example,"ml-team/classification".

To find the project string in the Neptune app, in theProject metadatasection, clickAdd new field.

Trashing a run with ID "CLS-1":

management.trash_objects(project=project_name, ids="CLS-1")

Fetch runs tagged as "trash" and delete them in batch:

runs_to_trash_df = project.fetch_runs_table(tag="trash").to_pandas()
runs_to_trash = runs_to_trash_df["sys/id"].tolist()

management.trash_objects(project=project_name, ids=runs_to_trash)

Trashing a newly created run:

run = neptune.init_run()
# It's a disaster, let's remove it
run_id = run["sys/id"].fetch()
management.trash_objects(project=project_name, ids=run_id)

You can also delete objects of different types at once:

run_id = run["sys/id"].fetch()
model_id = model["sys/id"].fetch()
model_version_id = model_version["sys/id"].fetch()
management.trash_objects(
 project=project_name,
 ids=[run_id, model_id, model_version_id]
)

delete_objects_from_trash()

Deletes one or more Neptune objects from the project trash.

Caution

• Deleting objects from trash is irreversible.
• Trashing a Model object also trashes all of its ModelVersion objects.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

The Neptune ID is a unique identifier for the run. TheExperimentstab displays it in the leftmost column.

In the run structure, the ID is stored in the system namespace (sys).

• If the run is active, you can obtain its ID withrun["sys/id"].fetch(). For example: >>>run=neptune.init_run()...>>>run["sys/id"].fetch()'CLS-26'
• If you set a custom run ID, it's stored atsys/custom_run_id: >>>run["sys/custom_run_id"].fetch()'vigilant-puffin-20bt9'

If the run is active, you can obtain its ID withrun["sys/id"].fetch(). For example:

>>> run = neptune.init_run()
...
>>> run["sys/id"].fetch()
'CLS-26'

If you set a custom run ID, it's stored atsys/custom_run_id:

>>> run["sys/custom_run_id"].fetch()
'vigilant-puffin-20bt9'

Examples

from neptune import management

management.delete_objects_from_trash(project="ml-team/classification", ids="CLS-1")

trash_to_delete = ["CLS-2", "CLS-3", "CLS-PRETRAINED"]
management.delete_objects_from_trash("ml-team/classification", ids=trash_to_delete)

clear_trash()

Empties the trash of the specified project.

Caution

• This action permanently deletes all of the objects in the trash.
• Trashing a Model object also trashes all of its ModelVersion objects.

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Examples

from neptune import management

management.clear_trash(project="ml-team/classification")

get_workspace_status()

This function is not available for self-hosted (onprem) Neptune.

Retrieves status information about a Neptune workspace, as a dictionary.

Includes the following:

• Storage usage and limit
• Active project count and limit
• Member count

Parameters

To keep your token secure, avoid placing it in the source code. Instead,save it as an environment variable.

Example

>>> from neptune import management
>>> management.get_workspace_status(workspace="ml-team")
... {'storageBytesAvailable': 214747451765,
... 'storageBytesLimit': 214748364800,
... 'storageBytesUsed': 913035,
... 'activeProjectsUsage': 1,
... 'activeProjectsLimit': 2,
... 'membersCount': 5}

Related

• Workspaces and projects
• Create a Neptune project
• User roles overview

Related Documentation

• Project API
• Neptune Package
• Creating and Deleting Projects

This page is originally sourced from the legacy docs.