Skip to main content

​
Parameter Objects

Parameter for a function. Arguments:
  • name (str): The name of the parameter.
  • description (Optional[str]): The description of the parameter.
  • type (str): The type of the parameter.
  • default (Any): The default value of the parameter.
  • required (bool): Whether the parameter is required.

​
to_ollama

Convert the parameter to a dictionary for an Ollama-compatible function schema. Returns: Dict[str, Any]: The parameter in dictionary format.

​
FunctionSchema Objects

Class that consumes a function and can return a schema required by different LLMs for function calling.

​
__init__

Initialize the FunctionSchema. Arguments:
  • function (Union[Callable, BaseModel]): The function to consume.

​
to_ollama

Convert the FunctionSchema to an Ollama-compatible function schema dictionary. Returns: Dict[str, Any]: The function schema in dictionary format.

​
get_schema_list

Get a list of function schemas from a list of functions or Pydantic BaseModels. Arguments:
  • items (List[Union[BaseModel, Callable]]): The functions or BaseModels to get the schemas for.
Returns: List[Dict[str, Any]]: A list of function schemas.

​
get_schema

Get a function schema from a function or Pydantic BaseModel. Arguments:
  • item (Union[BaseModel, Callable]): The function or BaseModel to get the schema for.
Returns: Dict[str, Any]: The function schema.

​
convert_python_type_to_json_type

Convert a Python type to a JSON type. Arguments:
  • param_type (str): The type of the parameter.
Returns: str: The JSON type.

​
route_and_execute

Route and execute a function. Arguments:
  • query (str): The query to route and execute.
  • llm (BaseLLM): The LLM to use.
  • functions (List[Callable]): The functions to execute.
  • layer (Layer): The layer to use.
Returns: Any: The result of the function.