nat.experimental.test_time_compute.functions.plan_select_execute_function

Attributes

logger

Classes

PlanSelectExecuteFunctionConfig

Defines a NAT function that performs reasoning on the input data.

Functions

plan_select_execute_function(config, builder)

Build a ExecutionPlanningFunction from the provided config.

Module Contents

logger

class PlanSelectExecuteFunctionConfig(/, **data: Any)

Bases: nat.data_models.function.FunctionBaseConfig

Defines a NAT function that performs reasoning on the input data. Output is passed to the next function in the workflow.

Designed to be used with an InterceptingFunction.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

augmented_fn: nat.data_models.component_ref.FunctionRef = None

planner: nat.data_models.component_ref.TTCStrategyRef = None

editor: nat.data_models.component_ref.TTCStrategyRef | None = None

scorer: nat.data_models.component_ref.TTCStrategyRef | None = None

selector: nat.data_models.component_ref.TTCStrategyRef = None

verbose: bool = None

agent_context_prompt_template: str = None

downstream_template: str = None

async plan_select_execute_function(

config: PlanSelectExecuteFunctionConfig,

builder: nat.builder.builder.Builder,

)

Build a ExecutionPlanningFunction from the provided config.

Args:

config (ExecutionPlanningFunctionConfig): The config for the ExecutionPlanningFunction. builder (Builder): The Builder instance to use for building the function.

Returns:

ExecutionPlanningFunction: The built ExecutionPlanningFunction.