ReAct Framework Implementation

This repository contains an implementation of the "ReAct" framework using OpenAI's GPT-3.5 Turbo model and LangChain, enabling dynamic reasoning and interaction with external information sources.

Prerequisites

Before running the code, ensure that you have the necessary libraries installed and API keys set up:

• OpenAI API key: Obtain one from the OpenAI platform.
• Serper API key: For external search functionality.

Set these keys as environment variables in your local environment or your Colab notebook, as demonstrated in the code.

Installation

To set up the required libraries and dependencies, run the following commands:

!pip install --upgrade openai
!pip install --upgrade langchain
!pip install --upgrade python-dotenv
!pip install google-search-results

Usage

1. Import the necessary libraries:

import os
import openai
from langchain.llms import OpenAI
from langchain.agents import load_tools
from langchain.agents import initialize_agent
from dotenv import load_dotenv

2. Set your API keys as environment variables:

os.environ["OPENAI_API_KEY"] = "YOUR_OPENAI_API_KEY"
os.environ["SERPER_API_KEY"] = "YOUR_SERPER_API_KEY"

3. Configure the LLM and tools:

llm = OpenAI(model_name="gpt-3.5-turbo", temperature=0)
tools = load_tools(["google-serper", "llm-math"], llm=llm)
agent = initialize_agent(tools, llm, agent="zero-shot-react-description", verbose=True)

4. Run the agent with your desired queries:

# Example 1: Query about Ronaldo
agent.run("What is Ronaldo's current team? How much is his salary in this team?")

# Example 2: Query about Gemini in Google
agent.run("What is Gemini in Google?")

Output

The code will execute the queries and provide responses using the ReAct framework, combining reasoning and action to generate informative answers.

Attribution

This code example was adapted from the LangChain documentation , and credit goes to them for the original implementation.