Stay organized with collectionsSave and categorize content based on your preferences.
Model development in a managed notebooks instance
Vertex AI Workbench managed notebooks isdeprecated. On
April 14, 2025, support for
managed notebooks will end and the ability to create managed notebooks instances
will be removed. Existing instances will continue to function
but patches, updates, and upgrades won't be available. To continue using
Vertex AI Workbench, we recommend that youmigrate
your managed notebooks instances to Vertex AI Workbench instances.
This page describes common ways to develop a machine learning (ML) model
in Vertex AI Workbench managed notebooks. You can use
pre-installed Python packages that are commonly used for ML model development,
Vertex AI custom training, and BigQuery ML.
Common Python packages
By default, managed notebooks instances are pre-installed
with Python packages that are commonly used for model development.
Import these packages into your notebook file and they are ready to use.
Vertex AI custom training
You can useVertex AI custom trainingto create and train models from within
your managed notebooks instance.
UsingBigQuery ML, you can train models that use
your BigQuery data, all from within
your managed notebooks instance.
For example, by using thePython client for
BigQuery,
you can send SQL commands from your notebook file
to create a model, and then use the model to get batch predictions.
BigQuery ML leverages the BigQuery computational engine,
so you don't need to deploy the compute resources
required for batch predictions or model training.
This can reduce the time it takes to set up training, evaluation,
and prediction.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-04 UTC."],[],[],null,["# Model development in Vertex AI Workbench managed notebooks\n\nModel development in a managed notebooks instance\n=================================================\n\n\n| Vertex AI Workbench managed notebooks is\n| [deprecated](/vertex-ai/docs/deprecations). On\n| April 14, 2025, support for\n| managed notebooks will end and the ability to create managed notebooks instances\n| will be removed. Existing instances will continue to function\n| but patches, updates, and upgrades won't be available. To continue using\n| Vertex AI Workbench, we recommend that you\n| [migrate\n| your managed notebooks instances to Vertex AI Workbench instances](/vertex-ai/docs/workbench/managed/migrate-to-instances).\n\n\u003cbr /\u003e\n\nThis page describes common ways to develop a machine learning (ML) model\nin Vertex AI Workbench managed notebooks. You can use\npre-installed Python packages that are commonly used for ML model development,\nVertex AI custom training, and BigQuery ML.\n\n### Common Python packages\n\nBy default, managed notebooks instances are pre-installed\nwith Python packages that are commonly used for model development.\nImport these packages into your notebook file and they are ready to use.\n\n### Vertex AI custom training\n\nYou can use [Vertex AI custom training](/vertex-ai/docs/training/overview)\nto create and train models from within\nyour managed notebooks instance.\n\n[Install one of the Vertex AI\nclient libraries](/vertex-ai/docs/start/client-libraries) on your\ninstance, or use the [Vertex AI API](/vertex-ai/docs/reference/rest)\nto send API requests from a Jupyter notebook file.\n\n### BigQuery ML\n\nUsing [BigQuery ML](/bigquery-ml/docs), you can train models that use\nyour BigQuery data, all from within\nyour managed notebooks instance.\nFor example, by using the [Python client for\nBigQuery](/python/docs/reference/bigquery/latest),\nyou can send SQL commands from your notebook file\nto create a model, and then use the model to get batch predictions.\n\nBigQuery ML leverages the BigQuery computational engine,\nso you don't need to deploy the compute resources\nrequired for batch predictions or model training.\nThis can reduce the time it takes to set up training, evaluation,\nand prediction.\n\nWhat's next\n-----------\n\n- To learn more about Vertex AI custom training, see\n [Understand the custom training service](/vertex-ai/docs/training/understanding-training-service).\n\n- To learn more about BigQuery ML, see\n [What is BigQuery ML?](/bigquery-ml/docs/introduction)"]]
