Vertex AI Feature Store is now powered by BigQuery
With Vertex AI’s new Feature Repository, you can now use your data from BigQuery directly in your MLOps processes.
In this way, you will avoid the expense of duplicating and keeping your data.
In addition, the ability to use features in real time (~2ms latency) and local placement support thanks to the vector import function make the new Feature Store even more powerful.
You can try the new version, which is currently in public preview, right away using the BigQuery engine in your project.
More details can be found here .
Using AI/ML and generative AI in Python with BigQuery Dataframes
Using your knowledge of Python with BigQuery Dataframes, you can now perform big data operations on BigQuery.
You can use BigQuery Dataframes in your own Python environment, as well as in Google Cloud environments such as BigQuery Studio and Colab Enterprise, and in solutions from Google partners such as Hex and Deepnote.
The library, which is currently in preview, includes two APIs, bigframes.pandas and bigframes.ml.
More details can be found here .
Are you enterpriseally ready for generative AI?
Today, there are many factors to consider in the enterprise use of generative AI models. To use enterprise-generative AI, especially privacy, security, control, and compliance, you need a platform that provides them.
True to its principles, Google Cloud is focused on helping your organization use generative AI to its full potential.
More details can be found here .
Events:
- You can watch the recent Generative AI with Google Cloud: Getting Started with Model Garden on Vertex AI seminar from the link.
You can register for the free Google Cloud Platform Fundamentals: Big Data & Machine Learning training via the link .
