jina-hubble-sdk 0.14.0

SDK for Hubble API at Jina AI.

jina-hubble-sdk

Install

pip install jina-hubble-sdk

Core functionality

• Python API and CLI.
• Authentication and token management.
• Artifact management.

Python API

Get a token

Including login if not.

import hubble
hubble.get_token()

If you are using inside an interactive environment, i.e. user can input via stdin:

import hubble
hubble.get_token(interactive=True)

Mark a function as login required,

import hubble

@hubble.login_required
def foo():
    pass

Login to Hubble

import hubble

# Open browser automatically and login via 3rd party.
# Token will be saved locally.
hubble.login()

Logout

import hubble

# If there is a valid token locally, 
# this will disable that token and remove it from local config.
hubble.logout()

Authentication and Token Management

After calling hubble.login(), you can use the client with:

import hubble

client = hubble.Client(
    max_retries=None,
    jsonify=True
)
# Get current user information.
response = client.get_user_info()
# Create a new personally access token for longer expiration period.
response = client.create_personal_access_token(
    name='my-pat',
    expiration_days=30
)
# Query all personal access tokens.
response = client.list_personal_access_tokens()

Artifact Management

import hubble
import io

client = hubble.Client(
    max_retries=None,
    jsonify=True
)

# Upload artifact to Hubble Artifact Storage by providing path.
response = client.upload_artifact(
    f='~/Documents/my-model.onnx',
    is_public=False
)

# Upload artifact to Hubble Artifact Storage by providing `io.BytesIO`
response = client.upload_artifact(
    f=io.BytesIO(b"some initial binary data: \x00\x01"),
    is_public=False
)

# Get current artifact information.
response = client.get_artifact_info(id='my-artifact-id')

# Download artifact to local directory.
response = client.download_artifact(
    id='my-artifact-id',
    f='my-local-filepath'
)
# Download artifact as an io.BytesIO object
response = client.download_artifact(
    id='my-artifact-id',
    f=io.BytesIO()
)

# Get list of artifacts.
response = client.list_artifacts(filter={'metaData.foo': 'bar'}, sort={'type': -1})

# Delete the artifact.
response = client.delete_artifact(id='my-artifact-id')

Error Handling

import hubble

client = hubble.Client()

try:
    client.get_user_info()
except hubble.excepts.AuthenticationRequiredError:
    print('Please login first.')
except Exception:
    print('Unknown error')

CLI

Login to Jina Cloud

Open browser automatically and login via 3rd party. Token will be saved locally.

jina auth login

Logout

If there is a valid token locally, this will disable that token and remove it from local config.

jina auth logout

Personal access token (PAT) management

Create a new PAT

jina auth token create <name of PAT> -e <expiration days>

List PATs

jina auth token list

Delete PAT

jina auth token delete <name of PAT>

Development

Local test

• Make a new virtual env. make env
• Install dependencies. make init
• The test should be run in a logged in environment. So need to login to Jina. jina auth login
• Test locally. make test

Release cycle

• Each time new commits come into main branch, CD workflow will generate a new release both on GitHub and Pypi.
• Each time new commits come into alpha branch, CD workflow will generate a new pre-release both on GitHub and Pypi.

FAQ (Frequently Asked Questions)

Run into RuntimeError: asyncio.run() cannot be called from a running event loop in Google Colab?

You could run into a problem when you trying to run these codes in Google Colab

import hubble

hubble.login()

The way to bypass this problem is adding nest_asyncio and use it first.

import hubble
import nest_asyncio

nest_asyncio.apply()
hubble.login()

Support

• Use Discussions to talk about your use cases, questions, and support queries.
• Join our Slack community and chat with other Jina community members about ideas.
• Join our Engineering All Hands meet-up to discuss your use case and learn Jina's new features.
  • When? The second Tuesday of every month
  • Where? Zoom (see our public events calendar/.ical) and live stream on YouTube
• Subscribe to the latest video tutorials on our YouTube channel

Join Us

Hubble Python SDK is backed by Jina AI and licensed under Apache-2.0. We are actively hiring AI engineers, solution engineers to build the next neural search ecosystem in opensource.