Can you release some examples about how to use RoleReactMode

[GitEasonXu]

Feature description

All examples is react_mode = BY_ORDER.
I know MetaGPT supporting REACT or PLAN_AND_ACT model, so could you release more example how to use REACT or PLAN_AND_ACT.

[iorisa]

The Data Interpreter utilizes the PLAN_AND_ACT mode.
You can find the detailed code in metagpt/roles/di/data_interpreter.py.
You can also find relevant demos at Data Interpreter (DI) | MetaGPT.

Regarding REACT, it consists of three actions: observe, think, and act. You can see these three operations (_observe(.), _think(.), _act(.)) in the base class Role in metagpt/roles/role.py:

    @role_raise_decorator
    async def run(self, with_message=None) -> Message | None:
        """Observe, and think and act based on the results of the observation"""
        ......
        if not await self._observe():
            ......
            return

        rsp = await self.react()

        .......
        return rsp

    async def react(self) -> Message:
        """Entry to one of three strategies by which Role reacts to the observed Message"""
        if self.rc.react_mode == RoleReactMode.REACT or self.rc.react_mode == RoleReactMode.BY_ORDER:
            rsp = await self._react()
        elif ......
        return rsp

    async def _react(self) -> Message:
        ......
        while actions_taken < self.rc.max_react_loop:
            # think
            todo = await self._think()
            if not todo:
                break
            # act
            rsp = await self._act()
            actions_taken += 1
        return rsp  # return output from the last action

When self.rc.max_react_loop of the _react is greater than 1, it is the BY_ORDER mode.

[github-actions]

This issue has no activity in the past 30 days. Please comment on the issue if you have anything to add.

[github-actions]

This issue was closed due to 45 days of inactivity. If you feel this issue is still relevant, please reopen the issue to continue the discussion.