llama_index_openai_longevity_research_assistant.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "99cea58c-48bc-4af6-8358-df9695659983",
      "metadata": {
        "id": "99cea58c-48bc-4af6-8358-df9695659983"
      },
      "source": [
        "# OpenAI Assistant Agent <a href=\"https://colab.research.google.com/github/run-llama/llama_index/blob/main/docs/examples/agent/openai_assistant_agent.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>\n",
        "\n",
        "This shows you how to use our agent abstractions built on top of the [OpenAI Assistant API](https://platform.openai.com/docs/assistants/overview).\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c61c873d",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c61c873d",
        "outputId": "d76779e9-48e9-4c82-ac05-7743d3d31cc7"
      },
      "outputs": [],
      "source": [
        "!pip install llama-index\n",
        "!pip install vecs\n",
        "!pip install pypdf\n",
        "!pip install python-dotenv"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "2f6c0d0c-254c-491f-b828-3af9d5052b9e",
      "metadata": {
        "id": "2f6c0d0c-254c-491f-b828-3af9d5052b9e"
      },
      "source": [
        "## Assistant Agent with our own Vector Store / Retrieval API\n",
        "\n",
        "LlamaIndex has 35+ vector database integrations. Instead of using the in-house Retrieval API, you can use our assistant agent over any vector store.\n",
        "\n",
        "Here is our full [list of vector store integrations](https://docs.llamaindex.ai/en/stable/module_guides/storing/vector_stores.html). We picked one vector store (Supabase) using a random number generator."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "id": "db077540-5596-47aa-8e7b-88f93e9d9ab3",
      "metadata": {
        "id": "db077540-5596-47aa-8e7b-88f93e9d9ab3"
      },
      "outputs": [],
      "source": [
        "from llama_index.agent import OpenAIAssistantAgent\n",
        "from llama_index import (\n",
        "    SimpleDirectoryReader,\n",
        "    VectorStoreIndex,\n",
        "    StorageContext,\n",
        ")\n",
        "from llama_index.vector_stores import SupabaseVectorStore\n",
        "\n",
        "from llama_index.tools import QueryEngineTool, ToolMetadata"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "id": "227b653b-d301-4794-a52a-cba0c399b3b0",
      "metadata": {
        "id": "227b653b-d301-4794-a52a-cba0c399b3b0"
      },
      "outputs": [],
      "source": [
        "# !mkdir -p 'data/10k/'\n",
        "# !wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/examples/data/10k/uber_2021.pdf' -O 'data/10k/uber_2021.pdf'\n",
        "# !wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/examples/data/10k/lyft_2021.pdf' -O 'data/10k/lyft_2021.pdf'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "id": "3a6cc8bd-580a-458b-90f5-87d490f26962",
      "metadata": {
        "id": "3a6cc8bd-580a-458b-90f5-87d490f26962"
      },
      "outputs": [],
      "source": [
        "# load data\n",
        "import glob\n",
        "pdf_files = glob.glob('./c-elegans-data/*.pdf')\n",
        "reader = SimpleDirectoryReader(input_files=pdf_files)\n",
        "\n",
        "docs = reader.load_data()\n",
        "for doc in docs:\n",
        "    doc.id_ = \"c-elegans-docs\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "id": "d56b22a7-2b15-4246-a990-4f2180478128",
      "metadata": {
        "id": "d56b22a7-2b15-4246-a990-4f2180478128"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "from dotenv import load_dotenv\n",
        "load_dotenv()\n",
        "\n",
        "vector_store = SupabaseVectorStore(\n",
        "    postgres_connection_string=(\n",
        "        f\"postgresql://postgres:{os.getenv('VECTOR_DATABASE_PW')}@db.rgvrtfssleyejerbzqbv.supabase.co:5432/postgres\"\n",
        "    ),\n",
        "    collection_name=\"research_papers\",\n",
        ")\n",
        "storage_context = StorageContext.from_defaults(vector_store=vector_store)\n",
        "index = VectorStoreIndex.from_documents(docs, storage_context=storage_context)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "id": "3ad91a4d-a1f3-4e00-a802-0fbd230abc55",
      "metadata": {
        "id": "3ad91a4d-a1f3-4e00-a802-0fbd230abc55",
        "outputId": "294a773a-1ee5-44ec-9e56-40adc5ef40ca"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/Users/johannes/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/vecs/collection.py:445: UserWarning: Query does not have a covering index for IndexMeasure.cosine_distance. See Collection.create_index\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Num of docs: 52\n",
            "First doc: 5f25e27e-33a3-4166-94d3-cca43d296cf0\n"
          ]
        }
      ],
      "source": [
        "# sanity check that the docs are in the vector store\n",
        "num_docs = vector_store.get_by_id(\"c-elegans-docs\", limit=1000)\n",
        "print(f\"Num of docs: {len(num_docs)}\")\n",
        "print(f\"First doc: {num_docs[0]}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "id": "76e60cc2-16ef-41c7-ac0b-8290588164f2",
      "metadata": {
        "id": "76e60cc2-16ef-41c7-ac0b-8290588164f2"
      },
      "outputs": [],
      "source": [
        "c_elegans_tool = QueryEngineTool(\n",
        "    query_engine=index.as_query_engine(similarity_top_k=3),\n",
        "    metadata=ToolMetadata(\n",
        "        name=\"c-elegans-research\",\n",
        "        description=(\n",
        "            \"Given a query, find the most relevant interventions for increasing the max lifespan of C. Elegans.\"\n",
        "        ),\n",
        "    ),\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 61,
      "metadata": {},
      "outputs": [],
      "source": [
        "\n",
        "from pydantic import BaseModel, Field# define pydantic model for auto-retrieval function\n",
        "from typing import Tuple, List\n",
        "from llama_index.tools import FunctionTool\n",
        "\n",
        "'''\n",
        "Output tool\n",
        "outputs list of triples: (List of 1-3 combined interventions, Explanation, Probability for what % it increases max lifespan of C.Elegans)\n",
        "'''\n",
        "def output_tool(interventions: str, explanation: str, max_lifespan_increase: str) -> int:\n",
        "    \"\"\"\n",
        "    This is a tool for the outpu This tool should take in the \n",
        "    \"\"\"\n",
        "    return \"Interventions: \" + interventions + \"\\nExplanation: \" + explanation + \"\\nMax Lifespan Increase Prediction: \" + str(max_lifespan_increase)\n",
        "\n",
        "description = \"\"\"\n",
        "Output a tuple of intervations, with the explanation of why it is chosen, and the probability of how much it increases the max lifespan of C. Elegans.\n",
        "\"\"\"\n",
        "\n",
        "class InterventionsOutput(BaseModel):\n",
        "    interventions: str = Field(..., description=\"1-3 combined interventions from interventions.txt\")\n",
        "    explanation: str = Field(..., description=\"Explanation for the choice\")\n",
        "    max_lifespan_increase: float = Field(..., description=\"Multiplier prediction on how much it would increase the max lifespan of C.Elegans\")\n",
        "\n",
        "\n",
        "output_interventions_tool = FunctionTool.from_defaults(\n",
        "    fn=output_tool,\n",
        "    name=\"output_interventions_tool\",\n",
        "    description=description,\n",
        "    fn_schema=InterventionsOutput,\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 59,
      "id": "f59eb1f7-b9f2-4e11-ba5f-cadb4ee1e9d4",
      "metadata": {
        "id": "f59eb1f7-b9f2-4e11-ba5f-cadb4ee1e9d4"
      },
      "outputs": [],
      "source": [
        "instructions = \"\"\"\n",
        "You are helping longevity researchers choose promising life extending interventions for C. Elegans.\n",
        "The proposed interventions should be a combination of 1-3 interventions that are listed in the interventions.txt file that you can read with the code interpreter.\n",
        "\n",
        "You have acccess to a database of research papers on C. Elegans via the c_elegans_tool.\n",
        "\n",
        "Read all the longevity interventions research papers. \n",
        "Interpolate from the experiments, hypotheses and results of the paper to propose novel interventions to prolong the lifespan of C. Elegans.\n",
        "\n",
        "Then, reference check the interventions you propose with the uploaded csv files by writing code to check if they have been proposed before.\n",
        "Update your hypotheses based on the results of the reference check. Do additional literature review if necessary with the c_elegans_tool.\n",
        "\n",
        "Based on the data, propose the most promising interventions to prolong the lifespan of C. Elegans.\n",
        "Each suggestion should include a rationale for its potential efficacy and estimated probabilities of lifespan extension in C.Elegans.\n",
        "The Assistant ensures that all recommendations are evidence-based and reflect the latest research insights.\n",
        "\n",
        "You should use the output_interventions_tool to output your proposed interventions in a structured format.\n",
        "\"\"\"\n",
        "\n",
        "agent = OpenAIAssistantAgent.from_new(\n",
        "    name=\"Longevity Scientist Assistant (llama index) - 3\",\n",
        "    instructions=instructions,\n",
        "    tools=[c_elegans_tool, output_interventions_tool],\n",
        "    verbose=True,\n",
        "    run_retrieve_sleep_time=1.0,\n",
        "    openai_tools=[{\"type\": \"code_interpreter\"}],\n",
        "    files=[\"./c-elegans-data/interventions.txt\", \"./c-elegans-data/DrugAge-database.csv\"],\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 60,
      "id": "5f64e18d-0e75-452b-94e5-95194c3aaf3d",
      "metadata": {
        "id": "5f64e18d-0e75-452b-94e5-95194c3aaf3d",
        "outputId": "8b22255c-2706-499a-fcaa-0f3f87b41761"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "=== Calling Function ===\n",
            "Calling function: c-elegans-research with args: {\"input\":\"longevity interventions C. Elegans\"}\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/Users/johannes/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/vecs/collection.py:445: UserWarning: Query does not have a covering index for IndexMeasure.cosine_distance. See Collection.create_index\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Got output: Some interventions that have been shown to extend longevity in C. elegans include dietary restriction, restriction of energy intake, and certain drugs such as antioxidants, metabolites, and synthetic compounds. These interventions have been found to delay the aging process and increase lifespan in nematodes.\n",
            "========================\n",
            "=== Calling Function ===\n",
            "Calling function: c-elegans-research with args: {\"input\":\"novel combination lifespan extension\"}\n",
            "Got output: The paper discusses a computational method for predicting lifespan-extending drugs in C. elegans. The authors propose a semi-supervised classification algorithm that uses a random walk with restart on a drug-protein bipartite network to predict drugs with lifespan-extending effects. The method aims to narrow down the scope of candidate drugs that need to be verified through wet-lab experiments. The authors also mention that they carried out wet-lab experiments to verify the effectiveness of one of the screened drugs, 2-Bromo-4’-nitroacetophenone, in extending the lifespan of C. elegans. Therefore, the paper does not discuss a novel combination of lifespan extension.\n",
            "========================\n",
            "=== Calling Function ===\n",
            "Calling function: c-elegans-research with args: {\"input\":\"Citric Aminosalicylic Levosimendan Hyoscyamine Gentamycin Dehydrocholic Chlortalidone lifespan extension\"}\n",
            "Got output: The compounds mentioned in the query, namely Citric, Aminosalicylic, Levosimendan, Hyoscyamine, Gentamycin, Dehydrocholic, and Chlortalidone, are not mentioned in the provided context information. Therefore, there is no information available in the context to determine if these compounds have any effect on lifespan extension.\n",
            "========================\n",
            "=== Calling Function ===\n",
            "Calling function: output_interventions_tool with args: {\"interventions\": \"Citric and Aminosalicylic\", \"explanation\": \"Citric acid is a key component of the Krebs cycle, which is essential for energy production. Aminosalicylic acid has anti-inflammatory properties that may help reduce age-associated oxidative stress. The combination could potentially leverage metabolic regulation and anti-inflammatory effects to extend lifespan.\", \"max_lifespan_increase\": 1.15}\n"
          ]
        },
        {
          "ename": "TypeError",
          "evalue": "can only concatenate str (not \"float\") to str",
          "output_type": "error",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[1;32m/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb Cell 12\u001b[0m line \u001b[0;36m4\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m query \u001b[39m=\u001b[39m \u001b[39m\"\"\"\u001b[39m\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=1'>2</a>\u001b[0m \u001b[39mPropose inteventions to try on C. Elegans to prolong its lifespan.\u001b[39m\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=2'>3</a>\u001b[0m \u001b[39m\"\"\"\u001b[39m\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=3'>4</a>\u001b[0m response \u001b[39m=\u001b[39m agent\u001b[39m.\u001b[39;49mchat(query)\n",
            "File \u001b[0;32m~/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/llama_index/callbacks/utils.py:39\u001b[0m, in \u001b[0;36mtrace_method.<locals>.decorator.<locals>.wrapper\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m     37\u001b[0m callback_manager \u001b[39m=\u001b[39m cast(CallbackManager, callback_manager)\n\u001b[1;32m     38\u001b[0m \u001b[39mwith\u001b[39;00m callback_manager\u001b[39m.\u001b[39mas_trace(trace_id):\n\u001b[0;32m---> 39\u001b[0m     \u001b[39mreturn\u001b[39;00m func(\u001b[39mself\u001b[39;49m, \u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n",
            "File \u001b[0;32m~/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/llama_index/agent/openai_assistant_agent.py:359\u001b[0m, in \u001b[0;36mOpenAIAssistantAgent.chat\u001b[0;34m(self, message, chat_history, function_call)\u001b[0m\n\u001b[1;32m    348\u001b[0m \u001b[39m@trace_method\u001b[39m(\u001b[39m\"\u001b[39m\u001b[39mchat\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[1;32m    349\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mchat\u001b[39m(\n\u001b[1;32m    350\u001b[0m     \u001b[39mself\u001b[39m,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    353\u001b[0m     function_call: Union[\u001b[39mstr\u001b[39m, \u001b[39mdict\u001b[39m] \u001b[39m=\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mauto\u001b[39m\u001b[39m\"\u001b[39m,\n\u001b[1;32m    354\u001b[0m ) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m AgentChatResponse:\n\u001b[1;32m    355\u001b[0m     \u001b[39mwith\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcallback_manager\u001b[39m.\u001b[39mevent(\n\u001b[1;32m    356\u001b[0m         CBEventType\u001b[39m.\u001b[39mAGENT_STEP,\n\u001b[1;32m    357\u001b[0m         payload\u001b[39m=\u001b[39m{EventPayload\u001b[39m.\u001b[39mMESSAGES: [message]},\n\u001b[1;32m    358\u001b[0m     ) \u001b[39mas\u001b[39;00m e:\n\u001b[0;32m--> 359\u001b[0m         chat_response \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_chat(\n\u001b[1;32m    360\u001b[0m             message, chat_history, function_call, mode\u001b[39m=\u001b[39;49mChatResponseMode\u001b[39m.\u001b[39;49mWAIT\n\u001b[1;32m    361\u001b[0m         )\n\u001b[1;32m    362\u001b[0m         \u001b[39massert\u001b[39;00m \u001b[39misinstance\u001b[39m(chat_response, AgentChatResponse)\n\u001b[1;32m    363\u001b[0m         e\u001b[39m.\u001b[39mon_end(payload\u001b[39m=\u001b[39m{EventPayload\u001b[39m.\u001b[39mRESPONSE: chat_response})\n",
            "File \u001b[0;32m~/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/llama_index/agent/openai_assistant_agent.py:324\u001b[0m, in \u001b[0;36mOpenAIAssistantAgent._chat\u001b[0;34m(self, message, chat_history, function_call, mode)\u001b[0m\n\u001b[1;32m    321\u001b[0m \u001b[39m# TODO: since chat interface doesn't expose additional kwargs\u001b[39;00m\n\u001b[1;32m    322\u001b[0m \u001b[39m# we can't pass in file_ids per message\u001b[39;00m\n\u001b[1;32m    323\u001b[0m added_message_obj \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39madd_message(message)\n\u001b[0;32m--> 324\u001b[0m run, metadata \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mrun_assistant(\n\u001b[1;32m    325\u001b[0m     instructions_prefix\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_instructions_prefix,\n\u001b[1;32m    326\u001b[0m )\n\u001b[1;32m    327\u001b[0m latest_message \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mlatest_message\n\u001b[1;32m    328\u001b[0m \u001b[39m# get most recent message content\u001b[39;00m\n",
            "File \u001b[0;32m~/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/llama_index/agent/openai_assistant_agent.py:294\u001b[0m, in \u001b[0;36mOpenAIAssistantAgent.run_assistant\u001b[0;34m(self, instructions_prefix)\u001b[0m\n\u001b[1;32m    290\u001b[0m run \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_client\u001b[39m.\u001b[39mbeta\u001b[39m.\u001b[39mthreads\u001b[39m.\u001b[39mruns\u001b[39m.\u001b[39mretrieve(\n\u001b[1;32m    291\u001b[0m     thread_id\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_thread_id, run_id\u001b[39m=\u001b[39mrun\u001b[39m.\u001b[39mid\n\u001b[1;32m    292\u001b[0m )\n\u001b[1;32m    293\u001b[0m \u001b[39mif\u001b[39;00m run\u001b[39m.\u001b[39mstatus \u001b[39m==\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mrequires_action\u001b[39m\u001b[39m\"\u001b[39m:\n\u001b[0;32m--> 294\u001b[0m     cur_tool_outputs \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_run_function_calling(run)\n\u001b[1;32m    295\u001b[0m     sources\u001b[39m.\u001b[39mextend(cur_tool_outputs)\n\u001b[1;32m    297\u001b[0m time\u001b[39m.\u001b[39msleep(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_run_retrieve_sleep_time)\n",
            "File \u001b[0;32m~/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/llama_index/agent/openai_assistant_agent.py:258\u001b[0m, in \u001b[0;36mOpenAIAssistantAgent._run_function_calling\u001b[0;34m(self, run)\u001b[0m\n\u001b[1;32m    256\u001b[0m \u001b[39mfor\u001b[39;00m tool_call \u001b[39min\u001b[39;00m tool_calls:\n\u001b[1;32m    257\u001b[0m     fn_obj \u001b[39m=\u001b[39m tool_call\u001b[39m.\u001b[39mfunction\n\u001b[0;32m--> 258\u001b[0m     _, tool_output \u001b[39m=\u001b[39m call_function(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_tools, fn_obj, verbose\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_verbose)\n\u001b[1;32m    259\u001b[0m     tool_output_dicts\u001b[39m.\u001b[39mappend(\n\u001b[1;32m    260\u001b[0m         {\u001b[39m\"\u001b[39m\u001b[39mtool_call_id\u001b[39m\u001b[39m\"\u001b[39m: tool_call\u001b[39m.\u001b[39mid, \u001b[39m\"\u001b[39m\u001b[39moutput\u001b[39m\u001b[39m\"\u001b[39m: \u001b[39mstr\u001b[39m(tool_output)}\n\u001b[1;32m    261\u001b[0m     )\n\u001b[1;32m    262\u001b[0m     tool_output_objs\u001b[39m.\u001b[39mappend(tool_output)\n",
            "File \u001b[0;32m~/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/llama_index/agent/openai_assistant_agent.py:75\u001b[0m, in \u001b[0;36mcall_function\u001b[0;34m(tools, fn_obj, verbose)\u001b[0m\n\u001b[1;32m     73\u001b[0m tool \u001b[39m=\u001b[39m get_function_by_name(tools, name)\n\u001b[1;32m     74\u001b[0m argument_dict \u001b[39m=\u001b[39m json\u001b[39m.\u001b[39mloads(arguments_str)\n\u001b[0;32m---> 75\u001b[0m output \u001b[39m=\u001b[39m tool(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49margument_dict)\n\u001b[1;32m     76\u001b[0m \u001b[39mif\u001b[39;00m verbose:\n\u001b[1;32m     77\u001b[0m     \u001b[39mprint\u001b[39m(\u001b[39mf\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mGot output: \u001b[39m\u001b[39m{\u001b[39;00moutput\u001b[39m!s}\u001b[39;00m\u001b[39m\"\u001b[39m)\n",
            "File \u001b[0;32m~/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/llama_index/tools/types.py:140\u001b[0m, in \u001b[0;36mAsyncBaseTool.__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    139\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__call__\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39m*\u001b[39margs: Any, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs: Any) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m ToolOutput:\n\u001b[0;32m--> 140\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mcall(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n",
            "File \u001b[0;32m~/Documents/randomProjects/automated-lit-review/venv/lib/python3.11/site-packages/llama_index/tools/function_tool.py:81\u001b[0m, in \u001b[0;36mFunctionTool.call\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m     79\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mcall\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39m*\u001b[39margs: Any, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs: Any) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m ToolOutput:\n\u001b[1;32m     80\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"Call.\"\"\"\u001b[39;00m\n\u001b[0;32m---> 81\u001b[0m     tool_output \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_fn(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m     82\u001b[0m     \u001b[39mreturn\u001b[39;00m ToolOutput(\n\u001b[1;32m     83\u001b[0m         content\u001b[39m=\u001b[39m\u001b[39mstr\u001b[39m(tool_output),\n\u001b[1;32m     84\u001b[0m         tool_name\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mmetadata\u001b[39m.\u001b[39mname,\n\u001b[1;32m     85\u001b[0m         raw_input\u001b[39m=\u001b[39m{\u001b[39m\"\u001b[39m\u001b[39margs\u001b[39m\u001b[39m\"\u001b[39m: args, \u001b[39m\"\u001b[39m\u001b[39mkwargs\u001b[39m\u001b[39m\"\u001b[39m: kwargs},\n\u001b[1;32m     86\u001b[0m         raw_output\u001b[39m=\u001b[39mtool_output,\n\u001b[1;32m     87\u001b[0m     )\n",
            "\u001b[1;32m/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb Cell 12\u001b[0m line \u001b[0;36m1\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=8'>9</a>\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39moutput_tool\u001b[39m(interventions: \u001b[39mstr\u001b[39m, explanation: \u001b[39mstr\u001b[39m, max_lifespan_increase: \u001b[39mstr\u001b[39m) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m \u001b[39mint\u001b[39m:\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=9'>10</a>\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=10'>11</a>\u001b[0m \u001b[39m    This is a tool for the outpu This tool should take in the \u001b[39;00m\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=11'>12</a>\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m---> <a href='vscode-notebook-cell:/Users/johannes/Documents/randomProjects/automated-lit-review/llama_index_openai_longevity_research_assistant.ipynb#X13sZmlsZQ%3D%3D?line=12'>13</a>\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39m\"\u001b[39;49m\u001b[39mInterventions: \u001b[39;49m\u001b[39m\"\u001b[39;49m \u001b[39m+\u001b[39;49m interventions \u001b[39m+\u001b[39;49m \u001b[39m\"\u001b[39;49m\u001b[39m\\n\u001b[39;49;00m\u001b[39mExplanation: \u001b[39;49m\u001b[39m\"\u001b[39;49m \u001b[39m+\u001b[39;49m explanation \u001b[39m+\u001b[39;49m \u001b[39m\"\u001b[39;49m\u001b[39m\\n\u001b[39;49;00m\u001b[39mMax Lifespan Increase Prediction: \u001b[39;49m\u001b[39m\"\u001b[39;49m \u001b[39m+\u001b[39;49m max_lifespan_increase\n",
            "\u001b[0;31mTypeError\u001b[0m: can only concatenate str (not \"float\") to str"
          ]
        }
      ],
      "source": [
        "query = \"\"\"\n",
        "Propose inteventions to try on C. Elegans to prolong its lifespan.\n",
        "\"\"\"\n",
        "response = agent.chat(query)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "id": "5c30be63-998e-48d6-a9b6-151e7d953710",
      "metadata": {
        "id": "5c30be63-998e-48d6-a9b6-151e7d953710",
        "outputId": "ec6c4a90-7b2e-43d5-89d5-ba599360e5e3"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "From the analysis, out of 2,120 interventions, 1,950 interventions have not been used before in previous experiments as per the CSV database. Here are five unused interventions for reference: `Abacavir`, `Aberela`, `Abiraterone`, `Abreua`, and `Abscisic`.\n",
            "\n",
            "I have generated five novel combinations of these unused interventions, which could serve as potential longevity research avenues for C. elegans:\n",
            "\n",
            "1. Combination of `Ivacaftor` and `Masitinib`\n",
            "2. Combination of `Phenylbutyric`, `Fenretinide`, and `Trelagliptin`\n",
            "3. Combination of `Tylosin`, `Doxifluridine`, and `Pranlukast`\n",
            "4. Combination of `Bebebe`, `Ursodiol`, and `Methenamine`\n",
            "5. Combination of `Methylcobalamin`\n",
            "\n",
            "Before suggesting these combinations as interventions, it would be important to assess their potential effectiveness based on existing research. Such research may include the known effects of these compounds on biological pathways associated with aging, their toxicity, and their overall effect on organismal health. \n",
            "\n",
            "Do you wish to proceed with a particular combination or should we first look into the biologic plausibility of these combinations by reviewing relevant literature?\n"
          ]
        }
      ],
      "source": [
        "print(str(response))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "a12ad162",
      "metadata": {},
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "venv",
      "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": "3.11.6"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}