RubricBasedEvaluatorDefinition Class

Rubric-based evaluator definition — stores dimensions produced by the generate API. Used for both quality and safety evaluators.

Constructor

RubricBasedEvaluatorDefinition(*args: Any, **kwargs: Any)

Variables

Name	Description
init_parameters	dict[str, any] The JSON schema (Draft 2020-12) for the evaluator's input parameters. This includes parameters like type, properties, required.
data_schema	dict[str, any] The JSON schema (Draft 2020-12) for the evaluator's input data. This includes parameters like type, properties, required.
metrics	dict[str, EvaluatorMetric] List of output metrics produced by this evaluator.
type	str or <xref:azure.ai.projects.models.RUBRIC> Required. Rubric-based evaluator definition. Stores dimensions (the scoring blueprint) for both quality and safety evaluators. Can be created via the generate API or manually via createVersion.
dimensions	list[Dimension] The set of dimensions — the scoring blueprint used by the LLM judge. Quality evaluators include a non-editable residual dimension with id 'general_quality' (always_applicable: true); safety evaluators include 'general_policy_compliance'. Both use the same Dimension structure. Required.
pass_threshold	float Pass/fail threshold for the aggregate rubric score, on the same normalized 0.0-1.0 scale as the emitted score. When the runtime weighted average meets or exceeds this value, the result is pass. Defaults to 0.5 (equivalent to a raw 1-5 weighted average of 3.0). The 'any dimension scored 1 → fail' rule still applies regardless of this threshold.

Methods

as_dict	Return a dict that can be turned into json using json.dump.
clear	Remove all items from D.
copy
get	Get the value for key if key is in the dictionary, else default. :param str key: The key to look up. :param any default: The value to return if key is not in the dictionary. Defaults to None :returns: D[k] if k in D, else d. :rtype: any
items
keys
pop	Removes specified key and return the corresponding value. :param str key: The key to pop. :param any default: The value to return if key is not in the dictionary :returns: The value corresponding to the key. :rtype: any :raises KeyError: If key is not found and default is not given.
popitem	Removes and returns some (key, value) pair :returns: The (key, value) pair. :rtype: tuple :raises KeyError: if D is empty.
setdefault	Same as calling D.get(k, d), and setting D[k]=d if k not found :param str key: The key to look up. :param any default: The value to set if key is not in the dictionary :returns: D[k] if k in D, else d. :rtype: any
update	Updates D from mapping/iterable E and F. :param any args: Either a mapping object or an iterable of key-value pairs.
values

as_dict

Return a dict that can be turned into json using json.dump.

as_dict(*, exclude_readonly: bool = False) -> dict[str, Any]

Keyword-Only Parameters

Name	Description
exclude_readonly	bool Whether to remove the readonly properties. Default value: False

Returns

Type	Description
dict	A dict JSON compatible object

clear

Remove all items from D.

clear() -> None

copy

copy() -> Model

get

Get the value for key if key is in the dictionary, else default. :param str key: The key to look up. :param any default: The value to return if key is not in the dictionary. Defaults to None :returns: D[k] if k in D, else d. :rtype: any

get(key: str, default: Any = None) -> Any

Parameters

Name	Description
key Required
default	Default value: None

items

items() -> ItemsView[str, Any]

Returns

Type	Description
ItemsView	set-like object providing a view on D's items

keys

keys() -> KeysView[str]

Returns

Type	Description
KeysView	a set-like object providing a view on D's keys

pop

Removes specified key and return the corresponding value. :param str key: The key to pop. :param any default: The value to return if key is not in the dictionary :returns: The value corresponding to the key. :rtype: any :raises KeyError: If key is not found and default is not given.

pop(key: str, default: ~typing.Any = <object object>) -> Any

Parameters

Name	Description
key Required
default

popitem

Removes and returns some (key, value) pair :returns: The (key, value) pair. :rtype: tuple :raises KeyError: if D is empty.

popitem() -> tuple[str, Any]

setdefault

Same as calling D.get(k, d), and setting D[k]=d if k not found :param str key: The key to look up. :param any default: The value to set if key is not in the dictionary :returns: D[k] if k in D, else d. :rtype: any

setdefault(key: str, default: ~typing.Any = <object object>) -> Any

Parameters

Name	Description
key Required
default

update

Updates D from mapping/iterable E and F. :param any args: Either a mapping object or an iterable of key-value pairs.

update(*args: Any, **kwargs: Any) -> None

values

values() -> ValuesView[Any]

Returns

Type	Description
ValuesView	an object providing a view on D's values

Attributes

data_schema

The JSON schema (Draft 2020-12) for the evaluator's input data. This includes parameters like type, properties, required.

data_schema: dict[str, Any] | None

dimensions

The set of dimensions — the scoring blueprint used by the LLM judge. Quality evaluators include a non-editable residual dimension with id 'general_quality' (always_applicable: true); safety evaluators include 'general_policy_compliance'. Both use the same Dimension structure. Required.

dimensions: list['_models.Dimension']

init_parameters

The JSON schema (Draft 2020-12) for the evaluator's input parameters. This includes parameters like type, properties, required.

init_parameters: dict[str, Any] | None

metrics

List of output metrics produced by this evaluator.

metrics: dict[str, '_models.EvaluatorMetric'] | None

pass_threshold

Pass/fail threshold for the aggregate rubric score, on the same normalized 0.0-1.0 scale as the emitted score. When the runtime weighted average meets or exceeds this value, the result is pass. Defaults to 0.5 (equivalent to a raw 1-5 weighted average of 3.0). The 'any dimension scored 1 → fail' rule still applies regardless of this threshold.

pass_threshold: float | None

type

Required. Rubric-based evaluator definition. Stores dimensions (the scoring blueprint) for both quality and safety evaluators. Can be created via the generate API or manually via createVersion.

type: RUBRIC: 'rubric'>]