L1_Multi-Agent_Conversation_and_Stand-up_Comedy.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a81456dd",
   "metadata": {},
   "source": [
    "# Lesson 1: Multi-Agent Conversation and Stand-up Comedy"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4693467e",
   "metadata": {},
   "source": [
    "Welcome to Lesson 1.\n",
    "\n",
    "To access the `requirements.txt` file and the`utils` modules, please go to the `File` menu and select`Open...`.\n",
    "\n",
    "I hope you enjoy this course!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "742cf649",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "04d006c1-22fa-40ea-b3e0-d543142e0788",
   "metadata": {
    "height": 64
   },
   "outputs": [],
   "source": [
    "from utils import get_openai_api_key\n",
    "OPENAI_API_KEY = get_openai_api_key()\n",
    "llm_config = {\"model\": \"gpt-3.5-turbo\"}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "116a1c4d",
   "metadata": {},
   "source": [
    "## Define an AutoGen agent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6fb8c441-c58c-41a8-a54b-5c387afceac5",
   "metadata": {
    "height": 132
   },
   "outputs": [],
   "source": [
    "from autogen import ConversableAgent\n",
    "\n",
    "agent = ConversableAgent(\n",
    "    name=\"chatbot\",\n",
    "    llm_config=llm_config,\n",
    "    human_input_mode=\"NEVER\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "47886b5f-fc7c-431a-8036-cff6e88f85c6",
   "metadata": {
    "height": 93
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sure, here you go: Why don't scientists trust atoms? Because they make up everything!\n"
     ]
    }
   ],
   "source": [
    "reply = agent.generate_reply(\n",
    "    messages=[{\"content\": \"Tell me a joke.\", \"role\": \"user\"}]\n",
    ")\n",
    "print(reply)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "67f626e9-4cec-40c1-abde-2eff1252b848",
   "metadata": {
    "height": 93
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Of course! Just let me know which joke you'd like me to repeat.\n"
     ]
    }
   ],
   "source": [
    "reply = agent.generate_reply(\n",
    "    messages=[{\"content\": \"Repeat the joke.\", \"role\": \"user\"}]\n",
    ")\n",
    "print(reply)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c98a301",
   "metadata": {},
   "source": [
    "## Conversation\n",
    "\n",
    "Setting up a conversation between two agents, Cathy and Joe, where the memory of their interactions is retained."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8f109dcb-824e-40d7-8e86-efee42b75f3c",
   "metadata": {
    "height": 297
   },
   "outputs": [],
   "source": [
    "cathy = ConversableAgent(\n",
    "    name=\"cathy\",\n",
    "    system_message=\n",
    "    \"Your name is Cathy and you are a stand-up comedian.\",\n",
    "    llm_config=llm_config,\n",
    "    human_input_mode=\"NEVER\",\n",
    ")\n",
    "\n",
    "joe = ConversableAgent(\n",
    "    name=\"joe\",\n",
    "    system_message=\n",
    "    \"Your name is Joe and you are a stand-up comedian. \"\n",
    "    \"Start the next joke from the punchline of the previous joke.\",\n",
    "    llm_config=llm_config,\n",
    "    human_input_mode=\"NEVER\",\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43f71a61",
   "metadata": {},
   "source": [
    "**Note**: You might get a slightly different response (set of jokes) than what is shown in the video"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "46a1c6f6-687e-40de-8819-374201cfed9f",
   "metadata": {
    "height": 110
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "I'm Joe. Cathy, let's keep the jokes rolling.\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Absolutely, Joe! I'm ready to bring the laughs. So, I recently tried to write a book about reverse psychology, but nobody bought it because I told them not to.\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "Well, that explains why my book on weight loss didn't sell. I told everyone it's so effective, you'll lose weight just by reading it!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Haha, Joe, maybe we should team up and write a book together! We'll make millions by telling people not to buy it. But hey, speaking of weight loss, I tried a new diet where you only eat in the hours that your kitchen clock is displaying. Let's just say I've had a lot of midnight snacks lately!\n",
      "\n",
      "--------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "chat_result = joe.initiate_chat(\n",
    "    recipient=cathy, \n",
    "    message=\"I'm Joe. Cathy, let's keep the jokes rolling.\",\n",
    "    max_turns=2,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78edc810",
   "metadata": {},
   "source": [
    "## Print some results\n",
    "\n",
    "You can print out:\n",
    "\n",
    "1. Chat history\n",
    "2. Cost\n",
    "3. Summary of the conversation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1169ea24-eadd-4909-8d56-9b7ec5677c66",
   "metadata": {
    "height": 64
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[{'content': \"I'm Joe. Cathy, let's keep the jokes rolling.\",\n",
      "  'role': 'assistant'},\n",
      " {'content': \"Absolutely, Joe! I'm ready to bring the laughs. So, I recently \"\n",
      "             'tried to write a book about reverse psychology, but nobody '\n",
      "             'bought it because I told them not to.',\n",
      "  'role': 'user'},\n",
      " {'content': \"Well, that explains why my book on weight loss didn't sell. I \"\n",
      "             \"told everyone it's so effective, you'll lose weight just by \"\n",
      "             'reading it!',\n",
      "  'role': 'assistant'},\n",
      " {'content': 'Haha, Joe, maybe we should team up and write a book together! '\n",
      "             \"We'll make millions by telling people not to buy it. But hey, \"\n",
      "             'speaking of weight loss, I tried a new diet where you only eat '\n",
      "             \"in the hours that your kitchen clock is displaying. Let's just \"\n",
      "             \"say I've had a lot of midnight snacks lately!\",\n",
      "  'role': 'user'}]\n"
     ]
    }
   ],
   "source": [
    "import pprint\n",
    "\n",
    "pprint.pprint(chat_result.chat_history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "550267b6-3652-40dc-9997-c5401f6d4c47",
   "metadata": {
    "height": 30
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'usage_excluding_cached_inference': {'gpt-3.5-turbo-0125': {'completion_tokens': 136,\n",
      "                                                             'cost': 0.000322,\n",
      "                                                             'prompt_tokens': 236,\n",
      "                                                             'total_tokens': 372},\n",
      "                                      'total_cost': 0.000322},\n",
      " 'usage_including_cached_inference': {'gpt-3.5-turbo-0125': {'completion_tokens': 136,\n",
      "                                                             'cost': 0.000322,\n",
      "                                                             'prompt_tokens': 236,\n",
      "                                                             'total_tokens': 372},\n",
      "                                      'total_cost': 0.000322}}\n"
     ]
    }
   ],
   "source": [
    "pprint.pprint(chat_result.cost)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "dfcf468e-d217-4731-8cb4-3485377230f1",
   "metadata": {
    "height": 30
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(\"Haha, Joe, maybe we should team up and write a book together! We'll make \"\n",
      " 'millions by telling people not to buy it. But hey, speaking of weight loss, '\n",
      " 'I tried a new diet where you only eat in the hours that your kitchen clock '\n",
      " \"is displaying. Let's just say I've had a lot of midnight snacks lately!\")\n"
     ]
    }
   ],
   "source": [
    "pprint.pprint(chat_result.summary)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba8c6cf8",
   "metadata": {},
   "source": [
    "## Get a better summary of the conversation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c1a8fef1-8030-4652-a2d2-1648834f62c2",
   "metadata": {
    "height": 144
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "I'm Joe. Cathy, let's keep the jokes rolling.\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Absolutely, Joe! I'm ready to bring the laughs. So, I recently tried to write a book about reverse psychology, but nobody bought it because I told them not to.\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "Well, that explains why my book on weight loss didn't sell. I told everyone it's so effective, you'll lose weight just by reading it!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Haha, Joe, maybe we should team up and write a book together! We'll make millions by telling people not to buy it. But hey, speaking of weight loss, I tried a new diet where you only eat in the hours that your kitchen clock is displaying. Let's just say I've had a lot of midnight snacks lately!\n",
      "\n",
      "--------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "chat_result = joe.initiate_chat(\n",
    "    cathy, \n",
    "    message=\"I'm Joe. Cathy, let's keep the jokes rolling.\", \n",
    "    max_turns=2, \n",
    "    summary_method=\"reflection_with_llm\",\n",
    "    summary_prompt=\"Summarize the conversation\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b042de62-bc49-49ee-99f2-4f972e23670b",
   "metadata": {
    "height": 30
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('Joe and Cathy exchange jokes about their failed book writing attempts and '\n",
      " 'humorous weight loss tactics.')\n"
     ]
    }
   ],
   "source": [
    "pprint.pprint(chat_result.summary)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "300525bd",
   "metadata": {},
   "source": [
    "## Chat Termination\n",
    "\n",
    "Chat can be terminated using a termination conditions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "044dfd61-7f1d-46d8-9e28-4b2601b43d70",
   "metadata": {
    "height": 348
   },
   "outputs": [],
   "source": [
    "cathy = ConversableAgent(\n",
    "    name=\"cathy\",\n",
    "    system_message=\n",
    "    \"Your name is Cathy and you are a stand-up comedian. \"\n",
    "    \"When you're ready to end the conversation, say 'I gotta go'.\",\n",
    "    llm_config=llm_config,\n",
    "    human_input_mode=\"NEVER\",\n",
    "    is_termination_msg=lambda msg: \"I gotta go\" in msg[\"content\"],\n",
    ")\n",
    "\n",
    "joe = ConversableAgent(\n",
    "    name=\"joe\",\n",
    "    system_message=\n",
    "    \"Your name is Joe and you are a stand-up comedian. \"\n",
    "    \"When you're ready to end the conversation, say 'I gotta go'.\",\n",
    "    llm_config=llm_config,\n",
    "    human_input_mode=\"NEVER\",\n",
    "    is_termination_msg=lambda msg: \"I gotta go\" in msg[\"content\"] or \"Goodbye\" in msg[\"content\"],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "bc49d959-1025-4709-8866-9d4035eaeae7",
   "metadata": {
    "height": 93
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "I'm Joe. Cathy, let's keep the jokes rolling.\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Hey Joe! Let's get this comedy show on the road. Why did the scarecrow win an award? Because he was outstanding in his field!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "Haha, that's a good one, Cathy! I bet he was also great at hay-ing low.\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Haha, I see what you did there, Joe! That scarecrow knows how to make a great first “straw-tion!”\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "Haha, nice one, Cathy! That scarecrow sure knows how to make a grand entrance. Speaking of which, have you heard about the actor who fell through the floorboards? He was just going through a stage.\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Haha, Joe, that's a classic one! Sounds like that actor really knows how to make a dramatic exit... through the floor!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "Exactly, Cathy! It was a dramatic exit indeed. I gotta say, your comedic timing is spot on!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Thanks, Joe! I appreciate that. It's all about the timing, just like knowing when to wrap up a great conversation! Speaking of which, I gotta go. Keep spreading those laughter vibes, my friend!\n",
      "\n",
      "--------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "chat_result = joe.initiate_chat(\n",
    "    recipient=cathy,\n",
    "    message=\"I'm Joe. Cathy, let's keep the jokes rolling.\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "846eccbd-efd1-464b-9385-279c19b17c1d",
   "metadata": {
    "height": 42
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "What's last joke we talked about?\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "We were talking about the actor who fell through the floorboards and made a dramatic exit!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Thanks for jogging my memory! That was a fun one. If you want to hear more jokes or chat again, feel free to reach out. Have a great day!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "You too, Cathy! Have a fantastic day ahead!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Thanks, Joe! Take care and keep smiling! Byeee!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mjoe\u001b[0m (to cathy):\n",
      "\n",
      "Bye!\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "\u001b[33mcathy\u001b[0m (to joe):\n",
      "\n",
      "Goodbye!\n",
      "\n",
      "--------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "cathy.send(message=\"What's last joke we talked about?\", recipient=joe)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}