{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import dotenv\n",
    "import os\n",
    "from langchain.tools import BaseTool\n",
    "from math import pi\n",
    "from typing import Union\n",
    "\n",
    "dotenv.load_dotenv()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain.chat_models import ChatOpenAI\n",
    "from langchain.chains.conversation.memory import ConversationBufferWindowMemory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# initialize LLM (we use ChatOpenAI because we'll later define a `chat` agent)\n",
    "llm = ChatOpenAI(\n",
    "    temperature=0,\n",
    "    model_name='gpt-3.5-turbo'\n",
    ")\n",
    "# initialize conversational memory\n",
    "conversational_memory = ConversationBufferWindowMemory(\n",
    "    memory_key='chat_history',\n",
    "    k=5,\n",
    "    return_messages=True\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'homeassistant_api'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[4], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39mtyping\u001b[39;00m \u001b[39mimport\u001b[39;00m Optional, List, Union\n\u001b[0;32m----> 2\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39mhomeassistant_api\u001b[39;00m \u001b[39mimport\u001b[39;00m Client\n\u001b[1;32m      5\u001b[0m haclient \u001b[39m=\u001b[39m Client(\n\u001b[1;32m      6\u001b[0m     \u001b[39m'\u001b[39m\u001b[39mhttp://homeassistant.local:8123/api\u001b[39m\u001b[39m'\u001b[39m,\n\u001b[1;32m      7\u001b[0m     \u001b[39m'\u001b[39m\u001b[39meyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJkNTVlNzMxNzAyMTE0ZmM4OTQ4NjAwMjFlN2UxNjc2YSIsImlhdCI6MTY4MjY3MTIwMCwiZXhwIjoxOTk4MDMxMjAwfQ.uwuasJHvbgesb6mQ_QdcmptYAgZxE4NTOnnpCWxLaek\u001b[39m\u001b[39m'\u001b[39m\n\u001b[1;32m      8\u001b[0m )\n\u001b[1;32m     10\u001b[0m desc \u001b[39m=\u001b[39m (\n\u001b[1;32m     11\u001b[0m     \u001b[39m\"\u001b[39m\u001b[39muse this tool when you need to turn on or off the light in the room. \u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m     12\u001b[0m     \u001b[39m\"\u001b[39m\u001b[39mgiven the light entity name and an action like turnon or turnoff, this tool will turn the lights on or off. \u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m     13\u001b[0m     \u001b[39m\"\u001b[39m\u001b[39mTo use the tool you must provide exactly two of the following parameters \u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m     14\u001b[0m     \u001b[39m\"\u001b[39m\u001b[39m[\u001b[39m\u001b[39m'\u001b[39m\u001b[39mentity_name\u001b[39m\u001b[39m'\u001b[39m\u001b[39m, \u001b[39m\u001b[39m'\u001b[39m\u001b[39maction\u001b[39m\u001b[39m'\u001b[39m\u001b[39m]\u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m     15\u001b[0m )\n",
      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'homeassistant_api'"
     ]
    }
   ],
   "source": [
    "from typing import Optional, List, Union\n",
    "from homeassistant_api import Client\n",
    "\n",
    "\n",
    "haclient = Client(\n",
    "    'http://homeassistant.local:8123/api',\n",
    "    'REPLACE_WITH'\n",
    ")\n",
    "\n",
    "desc = (\n",
    "    \"use this tool when you need to turn on or off the light in the room. \"\n",
    "    \"given the light entity name and an action like turnon or turnoff, this tool will turn the lights on or off. \"\n",
    "    \"To use the tool you must provide exactly two of the following parameters \"\n",
    "    \"['entity_name', 'action']\"\n",
    ")\n",
    "\n",
    "class HALightControl(BaseTool):\n",
    "    name = \"Control the lights in the room\"\n",
    "    description = desc\n",
    "\n",
    "\n",
    "    def _run(\n",
    "        self,\n",
    "        action: str,\n",
    "        entity_name: str,\n",
    "    ):\n",
    "        light = haclient.get_domain(\"light\")\n",
    "        if entity_name:\n",
    "            entity = \"light.protoboxer_{entity_name}\".format(entity_name=entity_name)\n",
    "            if action == \"turnon\":\n",
    "                light.turn_on(entity_id=entity)\n",
    "                return \"{entity_name} turned on\".format(entity_name=entity_name)\n",
    "            elif action == \"turnoff\":\n",
    "                light.turn_off(entity_id=entity)\n",
    "                return \"{entity_name} turned off\".format(entity_name=entity_name)\n",
    "        return \"\"\n",
    "    \n",
    "    def _arun(self, query: str):\n",
    "        raise NotImplementedError(\"This tool does not support async\")\n",
    "\n",
    "tools = [HALightControl()]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain.agents import initialize_agent\n",
    "\n",
    "sys_msg = \"\"\"Assistant is a large language model trained by OpenAI.\n",
    "\n",
    "Assistant is designed to be able to assist with a wide range of tasks, from answering simple questions to providing in-depth explanations and discussions on a wide range of topics. As a language model, Assistant is able to generate human-like text based on the input it receives, allowing it to engage in natural-sounding conversations and provide responses that are coherent and relevant to the topic at hand.\n",
    "\n",
    "Assistant is constantly learning and improving, and its capabilities are constantly evolving. It is able to process and understand large amounts of text, and can use this knowledge to provide accurate and informative responses to a wide range of questions. Additionally, Assistant is able to generate its own text based on the input it receives, allowing it to engage in discussions and provide explanations and descriptions on a wide range of topics.\n",
    "\n",
    "Overall, Assistant is a powerful system that can help with a wide range of tasks and provide valuable insights and information on a wide range of topics. Whether you need help with a specific question or just want to have a conversation about a particular topic, Assistant is here to assist.\n",
    "\"\"\"\n",
    "\n",
    "agent = initialize_agent(\n",
    "    agent='chat-conversational-react-description',\n",
    "    system_message=sys_msg,\n",
    "    tools=tools,\n",
    "    llm=llm,\n",
    "    verbose=True,\n",
    "    max_iterations=1,\n",
    "    early_stopping_method='generate',\n",
    "    memory=conversational_memory\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
      "\u001b[32;1m\u001b[1;3m{\n",
      "    \"action\": \"Control the lights in the room\",\n",
      "    \"action_input\": {\n",
      "        \"entity_name\": \"led_2\",\n",
      "        \"action\": \"turnoff\"\n",
      "    }\n",
      "}\u001b[0m\n",
      "Observation: \u001b[36;1m\u001b[1;3mled_2 turned off\u001b[0m\n",
      "Thought:"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "variable agent_scratchpad should be a list of base messages, got {\n    \"action\": \"Control the lights in the room\",\n    \"action_input\": {\n        \"entity_name\": \"led_2\",\n        \"action\": \"turnoff\"\n    }\n}\nObservation: led_2 turned off\nThought:\n\nI now need to return a final answer based on the previous steps:",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[73], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m agent(\u001b[39m\"\u001b[39;49m\u001b[39mTurn off the light led_2\u001b[39;49m\u001b[39m\"\u001b[39;49m)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/base.py:116\u001b[0m, in \u001b[0;36mChain.__call__\u001b[0;34m(self, inputs, return_only_outputs)\u001b[0m\n\u001b[1;32m    114\u001b[0m \u001b[39mexcept\u001b[39;00m (\u001b[39mKeyboardInterrupt\u001b[39;00m, \u001b[39mException\u001b[39;00m) \u001b[39mas\u001b[39;00m e:\n\u001b[1;32m    115\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mon_chain_error(e, verbose\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mverbose)\n\u001b[0;32m--> 116\u001b[0m     \u001b[39mraise\u001b[39;00m e\n\u001b[1;32m    117\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mon_chain_end(outputs, verbose\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mverbose)\n\u001b[1;32m    118\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mprep_outputs(inputs, outputs, return_only_outputs)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/base.py:113\u001b[0m, in \u001b[0;36mChain.__call__\u001b[0;34m(self, inputs, return_only_outputs)\u001b[0m\n\u001b[1;32m    107\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mon_chain_start(\n\u001b[1;32m    108\u001b[0m     {\u001b[39m\"\u001b[39m\u001b[39mname\u001b[39m\u001b[39m\"\u001b[39m: \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m\u001b[39m__class__\u001b[39m\u001b[39m.\u001b[39m\u001b[39m__name__\u001b[39m},\n\u001b[1;32m    109\u001b[0m     inputs,\n\u001b[1;32m    110\u001b[0m     verbose\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mverbose,\n\u001b[1;32m    111\u001b[0m )\n\u001b[1;32m    112\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m--> 113\u001b[0m     outputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_call(inputs)\n\u001b[1;32m    114\u001b[0m \u001b[39mexcept\u001b[39;00m (\u001b[39mKeyboardInterrupt\u001b[39;00m, \u001b[39mException\u001b[39;00m) \u001b[39mas\u001b[39;00m e:\n\u001b[1;32m    115\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mon_chain_error(e, verbose\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mverbose)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/agents/agent.py:807\u001b[0m, in \u001b[0;36mAgentExecutor._call\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m    805\u001b[0m     iterations \u001b[39m+\u001b[39m\u001b[39m=\u001b[39m \u001b[39m1\u001b[39m\n\u001b[1;32m    806\u001b[0m     time_elapsed \u001b[39m=\u001b[39m time\u001b[39m.\u001b[39mtime() \u001b[39m-\u001b[39m start_time\n\u001b[0;32m--> 807\u001b[0m output \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49magent\u001b[39m.\u001b[39;49mreturn_stopped_response(\n\u001b[1;32m    808\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mearly_stopping_method, intermediate_steps, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49minputs\n\u001b[1;32m    809\u001b[0m )\n\u001b[1;32m    810\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_return(output, intermediate_steps)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/agents/agent.py:515\u001b[0m, in \u001b[0;36mAgent.return_stopped_response\u001b[0;34m(self, early_stopping_method, intermediate_steps, **kwargs)\u001b[0m\n\u001b[1;32m    513\u001b[0m new_inputs \u001b[39m=\u001b[39m {\u001b[39m\"\u001b[39m\u001b[39magent_scratchpad\u001b[39m\u001b[39m\"\u001b[39m: thoughts, \u001b[39m\"\u001b[39m\u001b[39mstop\u001b[39m\u001b[39m\"\u001b[39m: \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_stop}\n\u001b[1;32m    514\u001b[0m full_inputs \u001b[39m=\u001b[39m {\u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mnew_inputs}\n\u001b[0;32m--> 515\u001b[0m full_output \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mllm_chain\u001b[39m.\u001b[39;49mpredict(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mfull_inputs)\n\u001b[1;32m    516\u001b[0m \u001b[39m# We try to extract a final answer\u001b[39;00m\n\u001b[1;32m    517\u001b[0m parsed_output \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39moutput_parser\u001b[39m.\u001b[39mparse(full_output)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/llm.py:151\u001b[0m, in \u001b[0;36mLLMChain.predict\u001b[0;34m(self, **kwargs)\u001b[0m\n\u001b[1;32m    137\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mpredict\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs: Any) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m \u001b[39mstr\u001b[39m:\n\u001b[1;32m    138\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"Format prompt with kwargs and pass to LLM.\u001b[39;00m\n\u001b[1;32m    139\u001b[0m \n\u001b[1;32m    140\u001b[0m \u001b[39m    Args:\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    149\u001b[0m \u001b[39m            completion = llm.predict(adjective=\"funny\")\u001b[39;00m\n\u001b[1;32m    150\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 151\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m(kwargs)[\u001b[39mself\u001b[39m\u001b[39m.\u001b[39moutput_key]\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/base.py:116\u001b[0m, in \u001b[0;36mChain.__call__\u001b[0;34m(self, inputs, return_only_outputs)\u001b[0m\n\u001b[1;32m    114\u001b[0m \u001b[39mexcept\u001b[39;00m (\u001b[39mKeyboardInterrupt\u001b[39;00m, \u001b[39mException\u001b[39;00m) \u001b[39mas\u001b[39;00m e:\n\u001b[1;32m    115\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mon_chain_error(e, verbose\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mverbose)\n\u001b[0;32m--> 116\u001b[0m     \u001b[39mraise\u001b[39;00m e\n\u001b[1;32m    117\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mon_chain_end(outputs, verbose\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mverbose)\n\u001b[1;32m    118\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mprep_outputs(inputs, outputs, return_only_outputs)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/base.py:113\u001b[0m, in \u001b[0;36mChain.__call__\u001b[0;34m(self, inputs, return_only_outputs)\u001b[0m\n\u001b[1;32m    107\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mon_chain_start(\n\u001b[1;32m    108\u001b[0m     {\u001b[39m\"\u001b[39m\u001b[39mname\u001b[39m\u001b[39m\"\u001b[39m: \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m\u001b[39m__class__\u001b[39m\u001b[39m.\u001b[39m\u001b[39m__name__\u001b[39m},\n\u001b[1;32m    109\u001b[0m     inputs,\n\u001b[1;32m    110\u001b[0m     verbose\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mverbose,\n\u001b[1;32m    111\u001b[0m )\n\u001b[1;32m    112\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m--> 113\u001b[0m     outputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_call(inputs)\n\u001b[1;32m    114\u001b[0m \u001b[39mexcept\u001b[39;00m (\u001b[39mKeyboardInterrupt\u001b[39;00m, \u001b[39mException\u001b[39;00m) \u001b[39mas\u001b[39;00m e:\n\u001b[1;32m    115\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mon_chain_error(e, verbose\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mverbose)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/llm.py:57\u001b[0m, in \u001b[0;36mLLMChain._call\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m     56\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_call\u001b[39m(\u001b[39mself\u001b[39m, inputs: Dict[\u001b[39mstr\u001b[39m, Any]) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m Dict[\u001b[39mstr\u001b[39m, \u001b[39mstr\u001b[39m]:\n\u001b[0;32m---> 57\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mapply([inputs])[\u001b[39m0\u001b[39m]\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/llm.py:118\u001b[0m, in \u001b[0;36mLLMChain.apply\u001b[0;34m(self, input_list)\u001b[0m\n\u001b[1;32m    116\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mapply\u001b[39m(\u001b[39mself\u001b[39m, input_list: List[Dict[\u001b[39mstr\u001b[39m, Any]]) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m List[Dict[\u001b[39mstr\u001b[39m, \u001b[39mstr\u001b[39m]]:\n\u001b[1;32m    117\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"Utilize the LLM generate method for speed gains.\"\"\"\u001b[39;00m\n\u001b[0;32m--> 118\u001b[0m     response \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mgenerate(input_list)\n\u001b[1;32m    119\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcreate_outputs(response)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/llm.py:61\u001b[0m, in \u001b[0;36mLLMChain.generate\u001b[0;34m(self, input_list)\u001b[0m\n\u001b[1;32m     59\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mgenerate\u001b[39m(\u001b[39mself\u001b[39m, input_list: List[Dict[\u001b[39mstr\u001b[39m, Any]]) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m LLMResult:\n\u001b[1;32m     60\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"Generate LLM result from inputs.\"\"\"\u001b[39;00m\n\u001b[0;32m---> 61\u001b[0m     prompts, stop \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mprep_prompts(input_list)\n\u001b[1;32m     62\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mllm\u001b[39m.\u001b[39mgenerate_prompt(prompts, stop)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/chains/llm.py:79\u001b[0m, in \u001b[0;36mLLMChain.prep_prompts\u001b[0;34m(self, input_list)\u001b[0m\n\u001b[1;32m     77\u001b[0m \u001b[39mfor\u001b[39;00m inputs \u001b[39min\u001b[39;00m input_list:\n\u001b[1;32m     78\u001b[0m     selected_inputs \u001b[39m=\u001b[39m {k: inputs[k] \u001b[39mfor\u001b[39;00m k \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mprompt\u001b[39m.\u001b[39minput_variables}\n\u001b[0;32m---> 79\u001b[0m     prompt \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mprompt\u001b[39m.\u001b[39;49mformat_prompt(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mselected_inputs)\n\u001b[1;32m     80\u001b[0m     _colored_text \u001b[39m=\u001b[39m get_colored_text(prompt\u001b[39m.\u001b[39mto_string(), \u001b[39m\"\u001b[39m\u001b[39mgreen\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[1;32m     81\u001b[0m     _text \u001b[39m=\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mPrompt after formatting:\u001b[39m\u001b[39m\\n\u001b[39;00m\u001b[39m\"\u001b[39m \u001b[39m+\u001b[39m _colored_text\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/prompts/chat.py:127\u001b[0m, in \u001b[0;36mBaseChatPromptTemplate.format_prompt\u001b[0;34m(self, **kwargs)\u001b[0m\n\u001b[1;32m    126\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mformat_prompt\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs: Any) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m PromptValue:\n\u001b[0;32m--> 127\u001b[0m     messages \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mformat_messages(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m    128\u001b[0m     \u001b[39mreturn\u001b[39;00m ChatPromptValue(messages\u001b[39m=\u001b[39mmessages)\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/prompts/chat.py:186\u001b[0m, in \u001b[0;36mChatPromptTemplate.format_messages\u001b[0;34m(self, **kwargs)\u001b[0m\n\u001b[1;32m    180\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39misinstance\u001b[39m(message_template, BaseMessagePromptTemplate):\n\u001b[1;32m    181\u001b[0m     rel_params \u001b[39m=\u001b[39m {\n\u001b[1;32m    182\u001b[0m         k: v\n\u001b[1;32m    183\u001b[0m         \u001b[39mfor\u001b[39;00m k, v \u001b[39min\u001b[39;00m kwargs\u001b[39m.\u001b[39mitems()\n\u001b[1;32m    184\u001b[0m         \u001b[39mif\u001b[39;00m k \u001b[39min\u001b[39;00m message_template\u001b[39m.\u001b[39minput_variables\n\u001b[1;32m    185\u001b[0m     }\n\u001b[0;32m--> 186\u001b[0m     message \u001b[39m=\u001b[39m message_template\u001b[39m.\u001b[39;49mformat_messages(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mrel_params)\n\u001b[1;32m    187\u001b[0m     result\u001b[39m.\u001b[39mextend(message)\n\u001b[1;32m    188\u001b[0m \u001b[39melse\u001b[39;00m:\n",
      "File \u001b[0;32m~/opt/anaconda3/envs/openai/lib/python3.10/site-packages/langchain/prompts/chat.py:43\u001b[0m, in \u001b[0;36mMessagesPlaceholder.format_messages\u001b[0;34m(self, **kwargs)\u001b[0m\n\u001b[1;32m     41\u001b[0m value \u001b[39m=\u001b[39m kwargs[\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mvariable_name]\n\u001b[1;32m     42\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39misinstance\u001b[39m(value, \u001b[39mlist\u001b[39m):\n\u001b[0;32m---> 43\u001b[0m     \u001b[39mraise\u001b[39;00m \u001b[39mValueError\u001b[39;00m(\n\u001b[1;32m     44\u001b[0m         \u001b[39mf\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mvariable \u001b[39m\u001b[39m{\u001b[39;00m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mvariable_name\u001b[39m}\u001b[39;00m\u001b[39m should be a list of base messages, \u001b[39m\u001b[39m\"\u001b[39m\n\u001b[1;32m     45\u001b[0m         \u001b[39mf\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mgot \u001b[39m\u001b[39m{\u001b[39;00mvalue\u001b[39m}\u001b[39;00m\u001b[39m\"\u001b[39m\n\u001b[1;32m     46\u001b[0m     )\n\u001b[1;32m     47\u001b[0m \u001b[39mfor\u001b[39;00m v \u001b[39min\u001b[39;00m value:\n\u001b[1;32m     48\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39misinstance\u001b[39m(v, BaseMessage):\n",
      "\u001b[0;31mValueError\u001b[0m: variable agent_scratchpad should be a list of base messages, got {\n    \"action\": \"Control the lights in the room\",\n    \"action_input\": {\n        \"entity_name\": \"led_2\",\n        \"action\": \"turnoff\"\n    }\n}\nObservation: led_2 turned off\nThought:\n\nI now need to return a final answer based on the previous steps:"
     ]
    }
   ],
   "source": [
    "agent(\"Turn off the light led_1\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "openai",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.11"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}