featurebyte.Feature.update_readiness

update_readiness(

readiness: Literal["DEPRECATED", "DRAFT", "PUBLIC_DRAFT", "PRODUCTION_READY"],

ignore_guardrails: bool=False

)

Description

Updates readiness of a feature version. A Feature version can have one of four readiness levels:

1. PRODUCTION_READY: Assigned to Feature versions that are ready for deployment in production environments.
2. PUBLIC_DRAFT: For Feature versions shared for feedback purposes.
3. DRAFT: For Feature versions in the prototype stage.
4. DEPRECATED: For feature versions not advised for use in either training or online serving.

When a new feature version is created, its status is DRAFT. Only a Draft feature version can be deleted. Feature versions with other status cannot be reverted to DRAFT. Only one version of a feature can be labeled as PRODUCTION_READY.

Once a feature version is promoted as PRODUCTION_READY, guardrails are automatically applied to ensure that the cleaning operations and FeatureJob settings are consistent with the defaults defined at the table level. These guardrails can be disregarded by setting the ignore_guardrails as True if the user is confident that the settings of the promoted feature version adhere to equally robust practices.

Parameters

• readiness: Literal["DEPRECATED", "DRAFT", "PUBLIC_DRAFT", "PRODUCTION_READY"]
  Feature readiness level
  
  
• ignore_guardrails: bool
  default: False
  Allow a user to specify if they want to ignore any guardrails when updating this feature. This should currently only apply of the FeatureReadiness value is being updated to PRODUCTION_READY. This should be a no-op for all other scenarios.
  
  

Examples

>>> feature = catalog.get_feature("InvoiceCount_60days")
>>> feature.update_readiness(readiness="PRODUCTION_READY")