Tecton

Introducing Tecton SDK 0.4

  • Mahesh Murag
By
Last updated: November 1, 2023
Share

Today we released version 0.4 of the Tecton SDK. The major changes in 0.4 include:

  • A new, simplified API for defining features in Tecton. More on these API changes below
  • Integration with Snowflake for offline feature computation and storage, as previously announced
  • Materialization info in ‘tecton plan’ output and the web apply log

See the release notes for more information on these enhancements.

API changes

In 0.4, we set out to further simplify Tecton’s declarative API and align on a standard specification with Feast, the leading open source feature store.

You can see our documentation for a full list of what’s changing. The core concepts stay the same, while many classes and parameters were cleaned up to be more intuitive. Also note the documentation version dropdown selector at the top of the page – if you don’t see it, try a hard refresh with shift+click refresh on mac or ctrl+click refresh on windows.

One of the great improvements is the new option for incremental backfills with Batch Feature Views. This option really simplifies authoring custom aggregations when Tecton’s built-in aggregations are not an option.

Updating to 0.4 Tecton SDK

If you are leveraging the Tecton 0.2 SDK or are in/near production as part of an automated process (such as CI/CD), please consult with your CSM and Consulting Architect before making this change. To stay on the 0.3 version series, change your requirements to ‘tecton[pyspark]==0.3.*’.

To check your currently installed version, run tecton version in your command line or tecton.version.summary() in your notebook.

  • In order to upgrade an existing feature view, you can use the compatibility library. If a feature view has materialized data, taking this step will re-materialize it which we do not recommend if it is in production. Support for no-materialization upgrades will be released later this year. We recommend avoiding upgrades for Feature Views that are costly to re-materialize until this functionality is available. 
  • We have extended support for Tecton 0.3 through September 1st, 2022
  • Tecton 0.4 will be supported until June 8th, 2023
  • If you need any help, please reach out to your CSM & Consulting Architect

Finally, please remember to update your local CLI and Notebook Cluster versions at the same time!

Share

Request a Demo

Unfortunately, Tecton does not currently support these clouds. We’ll make sure to let you know when this changes!

However, we are currently looking to interview members of the machine learning community to learn more about current trends.

If you’d like to participate, please book a 30-min slot with us here and we’ll send you a $50 amazon gift card in appreciation for your time after the interview.

CTA link

or

CTA button

Contact Sales

Interested in trying Tecton? Leave us your information below and we’ll be in touch.​

Unfortunately, Tecton does not currently support these clouds. We’ll make sure to let you know when this changes!

However, we are currently looking to interview members of the machine learning community to learn more about current trends.

If you’d like to participate, please book a 30-min slot with us here and we’ll send you a $50 amazon gift card in appreciation for your time after the interview.

CTA link

or

CTA button

Request a free trial

Interested in trying Tecton? Leave us your information below and we’ll be in touch.​

Unfortunately, Tecton does not currently support these clouds. We’ll make sure to let you know when this changes!

However, we are currently looking to interview members of the machine learning community to learn more about current trends.

If you’d like to participate, please book a 30-min slot with us here and we’ll send you a $50 amazon gift card in appreciation for your time after the interview.

CTA link

or

CTA button