December 14-15, 2023 — 7.44 Release Notes¶

Organization Usage: Improved views for billing reconciliation — General Availability¶

With this release, we are pleased to announce that the following views in the Organization Usage schema have been improved to make it easier to reconcile Snowflake usage with monthly billing statements:

• CONTRACT_ITEMS

• RATE_SHEET_DAILY

• REMAINING_BALANCE_DAILY

• USAGE_IN_CURRENCY_DAILY

During a transition period, only new accounts have these upgraded views available by default. To inquire about upgrading your account, contact Snowflake Support.

• CONTRACT_ITEMS View

• RATE_SHEET_DAILY View

• REMAINING_BALANCE_DAILY View

• USAGE_IN_CURRENCY_DAILY View

Snowflake Cortex ML-Based Time-Series Functions — General Availability¶

With this release, we are pleased to announce the general availability of the Snowflake Cortex ML-Based functions Forecasting (SNOWFLAKE.ML.FORECAST) and Anomaly Detection (SNOWFLAKE.ML.ANOMALY_DETECTION), which were previously available as preview features. These functions use a machine learning model trained on your historical data to make predictions and detect unexpected events. You can also obtain evaluation metrics and feature importance data for these models to learn what factors are driving trends and causing anomalies.

For more information, see Forecasting and Anomaly Detection.

Snowpark ML Modeling API — General Availability¶

With this release, we are pleased to announce the general availability of the Snowpark ML Modeling API, which was previously available as a preview feature. Snowpark ML Modeling lets you train Python models inside Snowflake using APIs similar to those provided by Scikit-Learn, LightGBM, and XGBoost. Many preprocessing classes run in distributed fashion, on as many nodes as are available in your warehouse, cutting runtime significantly.

This feature is available in Snowpark ML 1.1.1 and later. For more information, see Snowpark ML Modeling.

Snowpark ML Distributed Hyperparameter Optimization — Preview¶

With this release, we are pleased to announce the preview of distributed hyperparameter optimization in the Snowpark ML Modeling API. Hyperparameter optimization allows you to find the best parameters for your models and can now be run in distributed fashion, cutting runtime significantly. Distributed processing is enabled by default, but may be disabled.

This feature is available in Snowpark ML 1.1.1 and later. For more information, see Snowpark ML Modeling.

Cross-Cloud/Cross-Region Support for Iceberg Tables — Preview¶

With this release, we are pleased to announce cross-cloud/cross-region support for Iceberg tables that use an external Iceberg catalog.

For more information, see Cross-cloud/cross-region support.