Types

A primary objective of our framework is to simplify its usage for those who wish to employ it without delving into development. We aim to offer them a straightforward and effortless setup process, minimizing the complexities and initializations associated with numerous classes. This approach reduces potential bugs and allows users to focus on their primary task: developing an application using our framework. To ensure uniformity throughout the framework and to effectively manage new Planners while avoiding incorrect class initializations, we have introduced Types identifiers. These identifiers, at a high level, determine the specific type of planner required. Consequently, CHA users are relieved from the intricate details of initializing Planner objects, integrating them into the CHA, and handling setups.

Planner Types

This enumeration class defines different types of planner. This ensures consistency in case the planner developer decides to change the name of their planner, the end user need not to change their code cause they use the keys. It inherits from the str class and the Enum class in Python’s enum module. Each value in this enumeration represents a specific type of planner. The key naming convention should be all uppercase with underscore, and the value naming convention should be underscore_case: NAME_OF_PLANNER = this_is_a_sample_planner_name

from enum import Enum


class PlannerType(str, Enum):
    ZERO_SHOT_REACT_PLANNER = "zero_shot_react_planner"
    TREE_OF_THOUGHT = "tree_of_thought"

Types

This dictionary is used to map each PlannerType value to its corresponding Planner class. It allows for easy retrieval of the appropriate class based on the planner type.

from typing import Dict
from typing import Type

from planners.planner import BasePlanner
from planners.planner_types import PlannerType
from planners.react.base import ReActPlanner
from planners.tree_of_thought import TreeOfThoughtPlanner


PLANNER_TO_CLASS: Dict[PlannerType, Type[BasePlanner]] = {
    PlannerType.ZERO_SHOT_REACT_PLANNER: ReActPlanner,
    PlannerType.TREE_OF_THOUGHT: TreeOfThoughtPlanner,
}