nat.experimental.test_time_compute.scoring.motivation_aware_scorer#

Attributes#

Classes#

MotivationAwareScorer

A strategy that scores an TTCItem's output based on how well it

Functions#

Module Contents#

logger#
class MotivationAwareScorer(
config: nat.data_models.ttc_strategy.TTCStrategyBaseConfig,
)#

Bases: nat.experimental.test_time_compute.models.strategy_base.StrategyBase

A strategy that scores an TTCItem’s output based on how well it addresses both the original input (task) and the ‘motivation’ from metadata.

llm_bound = None#
async build_components(builder: nat.builder.builder.Builder) None#

Build the components required for the selector.

supported_pipeline_types() list[nat.experimental.test_time_compute.models.stage_enums.PipelineTypeEnum]#

Return the stage types supported by this selector.

stage_type() nat.experimental.test_time_compute.models.stage_enums.StageTypeEnum#

Return the stage type of this strategy.

async ainvoke(
items: list[nat.experimental.test_time_compute.models.ttc_item.TTCItem],
original_prompt: str | None = None,
agent_context: str | None = None,
**kwargs,
) list[nat.experimental.test_time_compute.models.ttc_item.TTCItem]#

Scores each item by combining the original ‘task_description’ and ‘motivation’ with the ‘output’. The resulting score is stored in item.score.

async register_motivation_aware_scorer(
config: nat.experimental.test_time_compute.models.scoring_config.MotivationAwareScoringConfig,
builder: nat.builder.builder.Builder,
)#