{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ed724944-f8db-4e10-9eef-549ffe63079e",
   "metadata": {},
   "source": [
    "# ConversationBufferWindowMemory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "360289bc-f61a-472b-b619-fbe0d236364b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain.chat_models import ChatOpenAI\n",
    "from langchain.chains import ConversationChain\n",
    "from langchain.memory import ConversationBufferWindowMemory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2b47dee4-d516-4eff-a29c-89879d2f8212",
   "metadata": {},
   "outputs": [],
   "source": [
    "llm = ChatOpenAI(temperature=0)\n",
    "memory = ConversationBufferWindowMemory(k=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9fd619b6-8232-431c-be08-12b2906da968",
   "metadata": {},
   "source": [
    "# Connect conversation to memory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d5b40fda-30b8-4b0e-a0a7-147412683036",
   "metadata": {},
   "outputs": [],
   "source": [
    "conversation = ConversationChain(\n",
    "    llm=llm,\n",
    "    memory=memory,\n",
    "    verbose=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4f01de60-34fd-4e70-8a9e-89c69fe36a88",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\u001b[1m> Entering new ConversationChain chain...\u001b[0m\n",
      "Prompt after formatting:\n",
      "\u001b[32;1m\u001b[1;3mThe following is a friendly conversation between a human and an AI. The AI is talkative and provides lots of specific details from its context. If the AI does not know the answer to a question, it truthfully says it does not know.\n",
      "\n",
      "Current conversation:\n",
      "\n",
      "Human: Hello, nice to meet you!\n",
      "AI:\u001b[0m\n",
      "\n",
      "\u001b[1m> Finished chain.\u001b[0m\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "\"Hello! Nice to meet you too! I'm an AI trained to have conversations and provide information. How can I assist you today?\""
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "conversation.predict(input=\"Hello, nice to meet you!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "239f0816-33ea-44d7-8324-4d5499b2fc14",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[HumanMessage(content='Hello, nice to meet you!', additional_kwargs={}, example=False),\n",
       " AIMessage(content=\"Hello! Nice to meet you too! I'm an AI trained to have conversations and provide information. How can I assist you today?\", additional_kwargs={}, example=False)]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "memory.buffer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0ff140ca-e9bd-4053-a61b-89e6caceb379",
   "metadata": {},
   "outputs": [],
   "source": [
    "memory.save_context({\"input\": \"Very Interesting.\"},\n",
    "                    {\"output\": \"Yes, it was my pleasure as and AI to answer.\"})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f0bae707-b0de-4538-ac84-c609447c9cb6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'history': 'Human: Very Interesting.\\nAI: Yes, it was my pleasure as and AI to answer.'}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "memory.load_memory_variables({})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "884f3bdc-c555-4266-bf35-ab088e9511a7",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}