-
Notifications
You must be signed in to change notification settings - Fork 3
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
1 parent
9d695eb
commit f149560
Showing
2 changed files
with
264 additions
and
264 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,264 @@ | ||
| { | ||
| "cells": [ | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 5, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "#@title Install and import necessary libraries\n", | ||
| "%pip install -qU \"tool-parse[langchain]\" langchain-groq langgraph duckduckgo-search\n", | ||
| "\n", | ||
| "import typing as t\n", | ||
| "\n", | ||
| "from duckduckgo_search import DDGS\n", | ||
| "\n", | ||
| "from langchain_groq.chat_models import ChatGroq\n", | ||
| "from langchain_core.messages import AIMessage, HumanMessage, ToolMessage\n", | ||
| "\n", | ||
| "from langgraph.checkpoint.memory import MemorySaver\n", | ||
| "from langgraph.prebuilt import create_react_agent\n", | ||
| "\n", | ||
| "from tool_parse.integrations.langchain import ExtendedStructuredTool, patch_chat_model" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "#### Define tools" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 17, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "data": { | ||
| "text/plain": [ | ||
| "[ExtendedStructuredTool(name='search_text', description='Search for text in the web.', func=<function search_text at 0x0000012F2A321120>),\n", | ||
| " ExtendedStructuredTool(name='product_info', description='Information about the product.', func=<class '__main__.ProductInfo'>)]" | ||
| ] | ||
| }, | ||
| "execution_count": 17, | ||
| "metadata": {}, | ||
| "output_type": "execute_result" | ||
| } | ||
| ], | ||
| "source": [ | ||
| "def search_text(\n", | ||
| " text: str,\n", | ||
| " *,\n", | ||
| " safe_search: bool = True,\n", | ||
| " backend: t.Literal[\"api\", \"html\", \"lite\"] = \"api\",\n", | ||
| " max_results: int = 1,\n", | ||
| "):\n", | ||
| " \"\"\"\n", | ||
| " Search for text in the web.\n", | ||
| " :param text: Text to search for.\n", | ||
| " :param safe_search: If True, enable safe search.\n", | ||
| " :param backend: Backend to use for retrieving results.\n", | ||
| " :param max_results: Max results to return.\n", | ||
| " \"\"\"\n", | ||
| " return DDGS().text(\n", | ||
| " keywords=text,\n", | ||
| " safesearch=\"on\" if safe_search else \"off\",\n", | ||
| " backend=backend,\n", | ||
| " max_results=max_results,\n", | ||
| " )\n", | ||
| "\n", | ||
| "class ProductInfo(t.NamedTuple): # Can be t.TypedDict or pydantic.BaseModel\n", | ||
| " \"\"\"\n", | ||
| " Information about the product.\n", | ||
| " :param name: Name of the product.\n", | ||
| " :param price: Price of the product.\n", | ||
| " :param in_stock: If the product is in stock.\n", | ||
| " \"\"\"\n", | ||
| "\n", | ||
| " name: str\n", | ||
| " price: float\n", | ||
| " in_stock: bool = False\n", | ||
| "\n", | ||
| "tools = [\n", | ||
| " ExtendedStructuredTool(func=search_text),\n", | ||
| " ExtendedStructuredTool(func=ProductInfo, name=\"product_info\"),\n", | ||
| "]\n", | ||
| "# OR # tools = ExtendedStructuredTool.from_objects(search_text, ProductInfo)\n", | ||
| "tools" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "#### Patch the langchain chat model" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 4, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "# Set groq api key (https://console.groq.com/keys)\n", | ||
| "\n", | ||
| "import os, getpass\n", | ||
| "\n", | ||
| "def set_env(var: str):\n", | ||
| " if not os.environ.get(var):\n", | ||
| " os.environ[var] = getpass.getpass(f\"{var}: \")\n", | ||
| "\n", | ||
| "set_env(\"GROQ_API_KEY\")" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 18, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "# Patch the instance\n", | ||
| "model = patch_chat_model(ChatGroq(api_key=os.getenv(\"GROQ_API_KEY\"), model=\"llama3-70b-8192\"))\n", | ||
| "# OR Patch the class and then instantiate it\n", | ||
| "model = patch_chat_model(ChatGroq)(api_key=os.getenv(\"GROQ_API_KEY\"), model=\"llama3-70b-8192\")" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "#### Create a react agent" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 19, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "agent_executor = create_react_agent(model, tools, checkpointer=MemorySaver())\n", | ||
| "config = {\"configurable\": {\"thread_id\": \"id123\"}}\n", | ||
| "\n", | ||
| "def query_agent(__query: str):\n", | ||
| " for chunk in agent_executor.stream({\"messages\": [HumanMessage(content=__query)]}, config):\n", | ||
| " message = list(chunk.values())[0][\"messages\"][0]\n", | ||
| " if isinstance(message, AIMessage):\n", | ||
| " print(\"[AGENT]\")\n", | ||
| " if message.tool_calls:\n", | ||
| " print(\"Calling tool:\")\n", | ||
| " metadata = message.tool_calls[0]\n", | ||
| " print(f\"name={metadata['name']!r}\")\n", | ||
| " print(f\"arguments={metadata['args']}\")\n", | ||
| " else:\n", | ||
| " print(message.content)\n", | ||
| " elif isinstance(message, ToolMessage):\n", | ||
| " print(f\"[{message.name!r} TOOL OUTPUT]\")\n", | ||
| " print(message.content)\n", | ||
| "\n", | ||
| " print(\"---\" * 20)" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 20, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "[AGENT]\n", | ||
| "Calling tool:\n", | ||
| "name='search_text'\n", | ||
| "arguments={'backend': 'lite', 'max_results': 5, 'text': 'langgraph docs'}\n", | ||
| "------------------------------------------------------------\n" | ||
| ] | ||
| }, | ||
| { | ||
| "name": "stderr", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "C:\\Users\\synac\\AppData\\Local\\Temp\\ipykernel_7024\\857887115.py:15: UserWarning: lxml is not installed. Using backend='api'.\n", | ||
| " return DDGS().text(\n" | ||
| ] | ||
| }, | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "['search_text' TOOL OUTPUT]\n", | ||
| "[{\"title\": \"LangGraph - LangChain\", \"href\": \"https://www.langchain.com/langgraph\", \"body\": \"LangGraph is a framework for building stateful, multi-agent systems with LLMs. Learn how to use LangGraph with Python or JavaScript, see conceptual guides, and access LangGraph Cloud for deployment and scaling.\"}, {\"title\": \"️LangGraph - GitHub Pages\", \"href\": \"https://langchain-ai.github.io/langgraph/\", \"body\": \"LangGraph is a framework for creating stateful, multi-actor applications with LLMs, using cycles, controllability, and persistence. Learn how to use LangGraph with examples, features, and integration with LangChain and LangSmith.\"}, {\"title\": \"️LangGraph.js - GitHub Pages\", \"href\": \"https://langchain-ai.github.io/langgraphjs/\", \"body\": \"LangGraph.js is a library for building stateful, multi-actor applications with LLMs, used to create agent and multi-agent workflows. Compared to other LLM frameworks, it offers these core benefits: cycles, controllability, and persistence. LangGraph allows you to define flows that involve cycles, essential for most agentic architectures ...\"}, {\"title\": \"Introduction to LangGraph\", \"href\": \"https://academy.langchain.com/courses/intro-to-langgraph\", \"body\": \"No. LangGraph is an orchestration framework for complex agentic systems and is more low-level and controllable than LangChain agents. On the other hand, LangChain provides a standard interface to interact with models and other components, useful for straight-forward chains and retrieval flows. How is LangGraph different from other agent frameworks?\"}, {\"title\": \"langgraph - PyPI\", \"href\": \"https://pypi.org/project/langgraph/\", \"body\": \"LangGraph is a library for building stateful, multi-actor applications with LLMs, used to create agent and multi-agent workflows. Compared to other LLM frameworks, it offers these core benefits: cycles, controllability, and persistence. LangGraph allows you to define flows that involve cycles, essential for most agentic architectures ...\"}]\n", | ||
| "------------------------------------------------------------\n", | ||
| "[AGENT]\n", | ||
| "Here are the top 5 sources for LangGraph docs using the lite backend: \n", | ||
| "\n", | ||
| "1. LangGraph - LangChain (https://www.langchain.com/langgraph) - Learn how to use LangGraph with Python or JavaScript, see conceptual guides, and access LangGraph Cloud for deployment and scaling.\n", | ||
| "\n", | ||
| "2. ️LangGraph - GitHub Pages (https://langchain-ai.github.io/langgraph/) - LangGraph is a framework for creating stateful, multi-actor applications with LLMs, using cycles, controllability, and persistence.\n", | ||
| "\n", | ||
| "3. ️LangGraph.js - GitHub Pages (https://langchain-ai.github.io/langgraphjs/) - LangGraph.js is a library for building stateful, multi-actor applications with LLMs, used to create agent and multi-agent workflows.\n", | ||
| "\n", | ||
| "4. Introduction to LangGraph (https://academy.langchain.com/courses/intro-to-langgraph) - LangGraph is an orchestration framework for complex agentic systems and is more low-level and controllable than LangChain agents.\n", | ||
| "\n", | ||
| "5. langgraph - PyPI (https://pypi.org/project/langgraph/) - LangGraph is a library for building stateful, multi-actor applications with LLMs, used to create agent and multi-agent workflows.\n", | ||
| "------------------------------------------------------------\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "query_agent(\"Search 5 sources for langgraph docs using lite backend\")" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 21, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "[AGENT]\n", | ||
| "Calling tool:\n", | ||
| "name='product_info'\n", | ||
| "arguments={'in_stock': True, 'name': 'RTX 4900', 'price': 3500}\n", | ||
| "------------------------------------------------------------\n", | ||
| "['product_info' TOOL OUTPUT]\n", | ||
| "[\"RTX 4900\", 3500.0, true]\n", | ||
| "------------------------------------------------------------\n", | ||
| "[AGENT]\n", | ||
| "Here is the product information:\n", | ||
| "\n", | ||
| "The product is RTX 4900, priced at $3500.00, and it is in stock.\n", | ||
| "------------------------------------------------------------\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "query_agent(\"Parse the product: RTX 4900, priced at $3.5k, is in stock.\")" | ||
| ] | ||
| } | ||
| ], | ||
| "metadata": { | ||
| "kernelspec": { | ||
| "display_name": "Python 3", | ||
| "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" | ||
| } | ||
| }, | ||
| "nbformat": 4, | ||
| "nbformat_minor": 2 | ||
| } |
Oops, something went wrong.