nat.experimental.test_time_compute.functions.ttc_tool_wrapper_function

Attributes

logger

Classes

TTCToolWrapperFunctionConfig

Configuration for the TTCToolWrapperFunction, which is used to wrap a function that will be executed

Functions

register_ttc_tool_wrapper_function(config, builder)

Register the TTCToolWrapperFunction with the provided builder and configuration.

Module Contents

logger

class TTCToolWrapperFunctionConfig(/, **data: Any)

Bases: nat.data_models.function.FunctionBaseConfig

Configuration for the TTCToolWrapperFunction, which is used to wrap a function that will be executed in the inference time scaling pipeline.

This function is responsible for turning an ‘objective’ or description for the tool into tool input.

NOTE: Only supports LLMs with structured output.

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

input_llm: nat.data_models.component_ref.LLMRef = None

verbose: bool = None

downstream_template: str = None

tool_description: str | None = None

async register_ttc_tool_wrapper_function(

config: TTCToolWrapperFunctionConfig,

builder: nat.builder.builder.Builder,

)

Register the TTCToolWrapperFunction with the provided builder and configuration.