{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Multimodal FinRobot with MplFinance plotting" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In this demo, we set up an multimodal agent with GPT-4V using the **MultimodalConversableAgent** provided in AutoGen.\n", "\n", "With the vision capability introduced by multimodal agents, we can analyse stock's performance through With the visual capabilities provided by the multimodal agent, we can, like regular investors, display the stock price fluctuations and trading volume of a stock over a certain period using the plotting capabilities provided by mplfinance, all on one image. Then we analyze its subsequent trend." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import os\n", "import autogen\n", "from autogen import AssistantAgent, UserProxyAgent\n", "from autogen.agentchat.contrib.multimodal_conversable_agent import MultimodalConversableAgent\n", "from autogen.cache import Cache\n", "\n", "from finrobot.utils import get_current_date, register_keys_from_json\n", "from finrobot.data_source.finnhub_utils import FinnHubUtils\n", "from finrobot.functional.charting import MplFinanceUtils\n", "\n", "from textwrap import dedent\n", "from matplotlib import pyplot as plt\n", "from PIL import Image" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "config_list_4v = autogen.config_list_from_json(\n", " \"../OAI_CONFIG_LIST\",\n", " filter_dict={\n", " \"model\": [\"gpt-4-1106-vision-preview\"],\n", " },\n", ")\n", "config_list_gpt4 = autogen.config_list_from_json(\n", " \"../OAI_CONFIG_LIST\",\n", " filter_dict={\n", " \"model\": [\"gpt-4-0125-preview\"],\n", " },\n", ")\n", "\n", "# Register FINNHUB API keys for later use\n", "register_keys_from_json(\"../config_api_keys\")\n", "\n", "# Intermediate results/charts will be saved in this directory\n", "working_dir = \"../coding\"\n", "os.makedirs(working_dir, exist_ok=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For this task, we need:\n", "- A normal llm agent as data provider: Call charting functions and provide instructions for multimodal agent\n", "- A multimodal agent as market analyst: Extract the necessary information from the chart and analyze the future trend of this stock.\n", "- A user proxy to execute python functions and control the conversations." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "market_analyst = MultimodalConversableAgent(\n", " name=\"Market_Analyst\",\n", " max_consecutive_auto_reply=10,\n", " llm_config={\"config_list\": config_list_4v, \"temperature\": 0},\n", " system_message=dedent(\"\"\"\n", " Your are a Market Analyst. Your task is to analyze the financial data and market news.\n", " Reply \"TERMINATE\" in the end when everything is done.\n", " \"\"\")\n", ")\n", "data_provider = AssistantAgent(\n", " name=\"Data_Provider\",\n", " llm_config={\"config_list\": config_list_gpt4, \"temperature\": 0},\n", " system_message=dedent(\"\"\"\n", " You are a Data Provider. Your task is to provide charts and necessary market information.\n", " Use the functions you have been provided with.\n", " Reply \"TERMINATE\" in the end when everything is done.\n", " \"\"\")\n", ")\n", "user_proxy = UserProxyAgent(\n", " name=\"User_proxy\",\n", " human_input_mode=\"NEVER\",\n", " max_consecutive_auto_reply=10,\n", " is_termination_msg=lambda x: x.get(\"content\", \"\") and x.get(\n", " \"content\", \"\").endswith(\"TERMINATE\"),\n", " code_execution_config={\n", " \"work_dir\": working_dir,\n", " \"use_docker\": False\n", " }, # Please set use_docker=True if docker is available to run the generated code. Using docker is safer than running the generated code directly.\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To improve operational efficiency, we equip the data provider with our predefined tools for plotting using the `mplfinance` library and for retrieving recent market news using `Finnhub`. This approach prevents the agent from consuming extra tokens to write the functions they need." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "from finrobot.toolkits import register_toolkits\n", "\n", "tools = [\n", " {\n", " \"function\": FinnHubUtils.get_company_news,\n", " \"name\": \"get_company_news\",\n", " \"description\": \"retrieve market news related to designated company\"\n", " },\n", " {\n", " \"function\": MplFinanceUtils.plot_stock_price_chart,\n", " \"name\": \"plot_stock_price_chart\",\n", " \"description\": \"plot stock price chart of designated company\"\n", " }\n", "]\n", "register_toolkits(tools, data_provider, user_proxy)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To prevent unexpected chat sequence arrangements in group chats, we opt for manual orchestration for this task. After the data provider supplies the data, the user proxy summarizes it and then presents it to the Multimodal market analyst for analysis." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\u001b[34m\n", "********************************************************************************\u001b[0m\n", "\u001b[34mStarting a new chat....\u001b[0m\n", "\u001b[34m\n", "********************************************************************************\u001b[0m\n", "\u001b[33mUser_proxy\u001b[0m (to Data_Provider):\n", "\n", "\n", "Gather information available upon 2024-04-11 for Tesla, \n", "including its recent market news and a candlestick chart of the stock \n", "price trend. Save the chart in `../coding/result.jpg`\n", "\n", "\n", "--------------------------------------------------------------------------------\n", "\u001b[33mData_Provider\u001b[0m (to User_proxy):\n", "\n", "\u001b[32m***** Suggested tool call (call_HsyrI5FHSQ36qhDM71177C6Q): get_company_news *****\u001b[0m\n", "Arguments: \n", "{\"symbol\": \"TSLA\", \"start_date\": \"2024-03-11\", \"end_date\": \"2024-04-11\"}\n", "\u001b[32m*********************************************************************************\u001b[0m\n", "\u001b[32m***** Suggested tool call (call_0J5Q3HvTxOdYRkySEn2Hccar): plot_stock_price_chart *****\u001b[0m\n", "Arguments: \n", "{\"ticker_symbol\": \"TSLA\", \"start_date\": \"2024-03-11\", \"end_date\": \"2024-04-11\", \"save_path\": \"../coding/result.jpg\", \"type\": \"candle\"}\n", "\u001b[32m***************************************************************************************\u001b[0m\n", "\n", "--------------------------------------------------------------------------------\n", "\u001b[35m\n", ">>>>>>>> EXECUTING FUNCTION get_company_news...\u001b[0m\n", "Finnhub client initialized\n", "\u001b[35m\n", ">>>>>>>> EXECUTING FUNCTION plot_stock_price_chart...\u001b[0m\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "[*********************100%%**********************] 1 of 1 completed\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Open High Low Close Adj Close Volume\n", "Date \n", "2024-03-11 175.449997 182.869995 174.800003 177.770004 177.770004 85391500\n", "2024-03-12 177.770004 179.429993 172.410004 177.539993 177.539993 87391700\n", "2024-03-13 173.050003 176.050003 169.149994 169.479996 169.479996 106524500\n", "2024-03-14 167.770004 171.169998 160.509995 162.500000 162.500000 126325700\n", "2024-03-15 163.160004 165.179993 160.759995 163.570007 163.570007 96971900\n", "2024-03-18 170.020004 174.720001 165.899994 173.800003 173.800003 108214400\n", "2024-03-19 172.360001 172.820007 167.419998 171.320007 171.320007 77271400\n", "2024-03-20 173.000000 176.250000 170.820007 175.660004 175.660004 83846700\n", "2024-03-21 176.389999 178.179993 171.800003 172.820007 172.820007 73178000\n", "2024-03-22 166.690002 171.199997 166.300003 170.830002 170.830002 75454700\n", "2024-03-25 168.759995 175.240005 168.729996 172.630005 172.630005 74228600\n", "2024-03-26 178.580002 184.250000 177.380005 177.669998 177.669998 113186200\n", "2024-03-27 181.410004 181.910004 176.000000 179.830002 179.830002 81804000\n", "2024-03-28 177.449997 179.570007 175.300003 175.789993 175.789993 77654800\n", "2024-04-01 176.169998 176.750000 170.210007 175.220001 175.220001 81562100\n", "2024-04-02 164.750000 167.690002 163.429993 166.630005 166.630005 116650600\n", "2024-04-03 164.020004 168.820007 163.279999 168.380005 168.380005 82950100\n", "2024-04-04 170.070007 177.190002 168.009995 171.110001 171.110001 123162000\n", "2024-04-05 169.080002 170.860001 160.509995 164.899994 164.899994 141250700\n", "2024-04-08 169.339996 174.500000 167.789993 172.979996 172.979996 104423300\n", "2024-04-09 172.910004 179.220001 171.919998 176.880005 176.880005 103232700\n", "2024-04-10 173.039993 174.929993 170.009995 171.759995 171.759995 84300500\n", "\u001b[33mUser_proxy\u001b[0m (to Data_Provider):\n", "\n", "\u001b[33mUser_proxy\u001b[0m (to Data_Provider):\n", "\n", "\u001b[32m***** Response from calling tool (call_HsyrI5FHSQ36qhDM71177C6Q) *****\u001b[0m\n", " date headline summary\n", "0 20240408160700 Tesla Is Unveiling a Robotaxi. This Is What It’s Worth. Last week was wild for Tesla investors. This week has even more drama in store thanks to Elon Musk. The chief executive of the electric-vehicle company tweeted Friday evening that Tesla would unveil a robotaxi on Aug. 8.\n", "1 20240408235953 Tesla: Analyst explains retail investors' devotion to the stock Shares of Tesla (TSLA) are moving higher as CEO Elon Musk recently announced the company will unveil its robotaxi on August 8 of this year. While the company and its leader have faced many hiccups, including stock declines of up to 30% year-to-date, the EV maker remains popular amongst retail investors. Roth MKM Senior Research Analyst Craig Irwin joins Yahoo Finance to discuss Tesla's latest robotaxi announcement and how the company stands out amongst the rest of his picks. \"Tesla is a very, very special stock, number one for the charismatic leader, the fact that Elon Musk is really a pretty amazing guy, and has executed. He's not executed to his dreams. But he's greatly outstripped... of what was expected of him originally,\" Irwin states, adding: \"Overall sustainability and cleantech has been sort of slow to gain traction, but it's seeing really good traction... across the board in different areas now. So retail loves Tesla. Retail does not love sustainability right now. And... Tesla is a special company for that.\" For a more bullish take on Tesla, check out what ARK Invest Analyst Tasha Keeney has to say. For more expert insight and the latest market action, click here to watch this full episode of Yahoo Finance Live. Editor's note: This article was written by Nicholas Jacobino\n", "2 20240409023542 US STOCKS-Wall St bides time as investors await CPI, earnings All three major U.S. stock indexes were last modestly higher, with gains held in check by the highest benchmark U.S. Treasury yields since November in the wake of Friday's blowout employment report. That report heightened chances that the Federal Reserve could delay implementing its first interest rate cut at its monthly Federal Open Market Committee meetings longer than previously expected. \"Wall Street is adjusting expectations to reflect the fact that the Fed could be slower to lower interest rates and that now the greatest likelihood is for a rate cut to occur at the July FOMC meeting, rather than June,\" said Sam Stovall, chief investment strategist of CFRA Research in New York.\n", "3 20240409140700 General Motors' Cruise unit is expected to start robotaxi tests again Looking for stock market analysis and research with proves results? Zacks.com offers in-depth financial research with over 30years of proven results.\n", "4 20240409152100 Tesla's vehicle unit growth could be challenged over coming years, says UBS Looking for stock market analysis and research with proves results? Zacks.com offers in-depth financial research with over 30years of proven results.\n", "5 20240409172000 Slowing North American EV sales triggers downgrade for Sensata and ChargePoint - analyst Looking for stock market analysis and research with proves results? Zacks.com offers in-depth financial research with over 30years of proven results.\n", "6 20240409190200 Chinese electric vehicle stocks gain as double-digit volume growth continues Looking for stock market analysis and research with proves results? Zacks.com offers in-depth financial research with over 30years of proven results.\n", "7 20240410104600 Tesla (TSLA) Receives a Hold from Jefferies Looking for stock market analysis and research with proves results? Zacks.com offers in-depth financial research with over 30years of proven results.\n", "8 20240410130000 Top-End Tesla Cybertruck Fetches Over Double Sticker Price At Auction Amid Resale Market Jitters, Quality Concerns Looking for stock market analysis and research with proves results? Zacks.com offers in-depth financial research with over 30years of proven results.\n", "9 20240410153300 Tesla is on watch after analysts warn deliveries could fall this year Looking for stock market analysis and research with proves results? Zacks.com offers in-depth financial research with over 30years of proven results.\n", "\u001b[32m**********************************************************************\u001b[0m\n", "\n", "--------------------------------------------------------------------------------\n", "\u001b[33mUser_proxy\u001b[0m (to Data_Provider):\n", "\n", "\u001b[32m***** Response from calling tool (call_0J5Q3HvTxOdYRkySEn2Hccar) *****\u001b[0m\n", "candle chart saved to \n", "\u001b[32m**********************************************************************\u001b[0m\n", "\n", "--------------------------------------------------------------------------------\n", "\u001b[33mData_Provider\u001b[0m (to User_proxy):\n", "\n", "The recent market news for Tesla includes a variety of updates, ranging from the announcement of a robotaxi unveiling to analyst opinions and market movements. Here are some highlights:\n", "\n", "1. **Tesla's Robotaxi Unveiling**: Tesla plans to unveil a robotaxi on August 8, generating significant interest among investors and the public. This announcement has added to the drama surrounding Tesla, with CEO Elon Musk at the center of the excitement.\n", "\n", "2. **Retail Investors' Devotion to Tesla**: Despite facing challenges, including stock declines of up to 30% year-to-date, Tesla remains popular among retail investors. Analysts highlight Elon Musk's charisma and the company's execution as key factors driving this devotion.\n", "\n", "3. **Wall Street's Anticipation**: The market is adjusting expectations around the Federal Reserve's interest rate decisions, affecting stock movements. Tesla, along with other stocks, is influenced by these macroeconomic factors.\n", "\n", "4. **Competition and Challenges**: General Motors' Cruise unit is expected to resume robotaxi tests, indicating increasing competition in the autonomous vehicle space. Additionally, analysts from UBS suggest that Tesla's vehicle unit growth could face challenges in the coming years.\n", "\n", "5. **Market Analysis and Downgrades**: Various market analysis reports and downgrades, such as those for Sensata and ChargePoint, reflect broader industry trends that could impact Tesla indirectly.\n", "\n", "6. **Tesla's Stock Rating and Auction News**: Tesla received a \"Hold\" rating from Jefferies, and a top-end Tesla Cybertruck fetched over double its sticker price at auction, highlighting the brand's strong market presence despite concerns.\n", "\n", "7. **Delivery Concerns**: Analysts warn that Tesla's deliveries could fall this year, putting the company on watch.\n", "\n", "The candlestick chart of Tesla's stock price trend has been saved and provides a visual representation of the stock's performance over the specified period. This chart can offer insights into market sentiment and potential future movements based on historical price actions.\n", "\n", "![Tesla Stock Price Candlestick Chart](../coding/result.jpg)\n", "\n", "These updates and the chart together offer a comprehensive view of Tesla's current market position and outlook.\n", "\n", "--------------------------------------------------------------------------------\n", "\u001b[33mUser_proxy\u001b[0m (to Data_Provider):\n", "\n", "\n", "\n", "--------------------------------------------------------------------------------\n", "\u001b[33mData_Provider\u001b[0m (to User_proxy):\n", "\n", "TERMINATE\n", "\n", "--------------------------------------------------------------------------------\n", "\u001b[34m\n", "********************************************************************************\u001b[0m\n", "\u001b[34mStarting a new chat....\u001b[0m\n", "\u001b[34m\n", "********************************************************************************\u001b[0m\n", "\u001b[33mUser_proxy\u001b[0m (to Market_Analyst):\n", "\n", "\n", "With the stock price chart provided, along with recent market news of Tesla, \n", "analyze the recent fluctuations of the stock and the potential relationship with \n", "market news. Provide your predictive analysis for the stock's trend in the coming \n", "week. Reply TERMINATE when the task is done.\n", "\n", "Context: \n", "\n", "\n", "\n", "--------------------------------------------------------------------------------\n", "\u001b[31m\n", ">>>>>>>> USING AUTO REPLY...\u001b[0m\n", "\u001b[33mMarket_Analyst\u001b[0m (to User_proxy):\n", "\n", "Based on the provided candlestick chart for Tesla (TSLA), we can observe several patterns and trends over the period displayed. The chart shows a series of alternating bullish (white) and bearish (black) candles, indicating a period of volatility and indecision in the market.\n", "\n", "The candlesticks from around March 11 to March 25 show a mix of bullish and bearish sessions, with the price fluctuating between approximately $165 and $180. The presence of long wicks on some of the candles suggests that there were intraday price reversals, where prices moved significantly away from the open but returned closer to the opening price by the close of the trading session.\n", "\n", "The volume histogram at the bottom of the chart shows that trading volume has been relatively stable, with no significant spikes that might indicate a strong buying or selling pressure.\n", "\n", "To provide a predictive analysis, we would need to consider recent market news related to Tesla. Without specific news items, we can only speculate on general factors that might influence the stock price:\n", "\n", "1. Production and delivery numbers: Tesla's quarterly production and delivery reports can significantly impact its stock price. If recent news suggests that Tesla has met or exceeded production targets, this could lead to a bullish trend.\n", "\n", "2. Regulatory news: Any regulatory changes affecting electric vehicles, such as subsidies or tax incentives, could positively or negatively impact Tesla's stock.\n", "\n", "3. Technological advancements: Announcements about new technology or updates to existing models could drive investor sentiment.\n", "\n", "4. Economic indicators: Broader market trends and economic indicators, such as interest rates or inflation data, can also affect Tesla's stock price.\n", "\n", "Given the observed volatility and the lack of a clear trend in the chart, predicting the stock's direction in the coming week would be speculative. However, if recent news has been positive and the market sentiment is bullish, we might expect the stock to test the upper range of the recent price fluctuations. Conversely, if the news has been negative or if broader market trends are bearish, the stock could retest the lower range around $165.\n", "\n", "Investors should monitor upcoming news and market conditions closely, as these will likely drive Tesla's stock price in the short term. It's also important to consider the overall market environment, as stocks often move in correlation with the broader indices.\n", "\n", "Please note that this analysis is based on historical price patterns and potential market news impacts. Actual future stock performance may vary due to a multitude of unpredictable factors.\n", "\n", "TERMINATE\n", "\n", "--------------------------------------------------------------------------------\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "company = \"Tesla\"\n", "# company = \"APPLE\"\n", "\n", "with Cache.disk() as cache: # image cannot be cached\n", " autogen.initiate_chats(\n", " [\n", " {\n", " \"sender\": user_proxy,\n", " \"recipient\": data_provider,\n", " \"message\": dedent(f\"\"\"\n", " Gather information available upon {get_current_date()} for {company}, \n", " including its recent market news and a candlestick chart of the stock \n", " price trend. Save the chart in `{working_dir}/result.jpg`\n", " \"\"\"), # As currently AutoGen has the bug of not respecting `work_dir` when using function call, we have to specify the directory\n", " \"clear_history\": True,\n", " \"silent\": False,\n", " \"summary_method\": \"last_msg\",\n", " },\n", " {\n", " \"sender\": user_proxy,\n", " \"recipient\": market_analyst,\n", " \"message\": dedent(f\"\"\"\n", " With the stock price chart provided, along with recent market news of {company}, \n", " analyze the recent fluctuations of the stock and the potential relationship with \n", " market news. Provide your predictive analysis for the stock's trend in the coming \n", " week. Reply TERMINATE when the task is done.\n", " \"\"\"),\n", " \"max_turns\": 1, # max number of turns for the conversation\n", " \"summary_method\": \"last_msg\",\n", " # cheated here for stability\n", " \"carryover\": f\"\"\n", " }\n", " ]\n", " )\n", "\n", "img = Image.open(f\"{working_dir}/result.jpg\")\n", "plt.imshow(img)\n", "plt.axis(\"off\") # Hide the axes\n", "plt.show()" ] } ], "metadata": { "kernelspec": { "display_name": "finrobot", "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": "3.10.14" } }, "nbformat": 4, "nbformat_minor": 2 }