Managing upgrades and dependencies for Vertex AI Workbench user-managed notebooks

Sample Cloud Build pipelines for Vertex AI Workbench user-managed notebooks upgrades and dependency management.

How-to use

0 - Prepare:

• Clone this repository.
• Choose a Deep Learning Container (DLCs) image and specifiy it in the Dockerfile
• Add notebook dependencies to the Dockerfile.
• Add notebooks for automated testing in folder test_notebooks
• Specify an _IMAGE_NAME in each of of the pipeline yaml files.

1 - Publish staging image:

• Create a trigger to run the staging pipeline on each commit to branch called staging.
• Commit into to the staging branch and push to your own remote repository.
• Cloud Build automatically runs the test_notebooks and creates the staging image.

2 - Test staging image

• (Optional) Create notebook instances to manually test image using the staging environment.

  gcloud notebooks instances create <INSTANCE_NAME> \
    --location=<NOTEBOOK_LOCATION> \
    --environment=<IMAGE_NAME>.staging

• Access your instance from the Vertex AI Workbench console

3 - Publish production image:

• Create a trigger to run the production pipeline on each commit to a branch called production.
• Commit into the production branch and push to your own remote repository.
• Cloud Build creates the production image.

4 - Create notebook instances using image from the production environment

  gcloud notebooks instances create <INSTANCE_NAME> \
    --location=<NOTEBOOK_LOCATION> \
    --environment=<IMAGE_NAME>.production

• Access your instance from the Vertex AI Workbench console

Both pipelines use the original commit SHA as the Docker tag, representing the image version ID in Container Registry.

Licensing

Copyright 2021 Google LLC

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    https://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

Contributing

• Contributing guidelines
• Code of conduct