langchain-weather-assistant

This project demonstrates how to create a weather agent using LangChain, OpenAI's GPT-4, and external APIs to fetch weather data based on a user's query. It integrates with OpenWeatherMap to retrieve weather details and geolocation information (latitude and longitude) for a given city. The app is built using Streamlit for the user interface.

Features

• Weather Information Retrieval: Fetch current weather details (e.g., temperature, humidity, etc.) based on latitude and longitude.
• Geolocation Lookup: Convert a city name to latitude and longitude using OpenWeatherMap's API.
• Streamlit Interface: A simple web interface where users can input a city name and get weather details.
• LangChain Agent: Utilizes LangChain to build a natural language processing agent powered by GPT-4 to process and invoke the weather and geolocation tools.

Requirements

• requests
• streamlit
• langchain
• langchain_openai
• python-dotenv

To install the required libraries, run:

pip install -r requirements.txt

Setup

1. Sign up for an API Key from OpenWeatherMap:

   • Go to OpenWeatherMap and sign up to get an API key.
   • Save your API key in the .env file as WEATHER_API_KEY (example .env content):

     WEATHER_API_KEY=your_api_key_here
     

2. Clone this Repository:

   git clone https://github.com/your-username/weather-agent-langchain.git
   cd weather-agent-langchain

3. Run the Streamlit App:

   To start the app, run:

   streamlit run weather_assistant.py

   This will open the app in your browser.

How It Works

1. LangChain Agent: The agent is initialized with two structured tools:

   • get_latitude_longitude: Fetches latitude and longitude for a city name using OpenWeatherMap's geolocation API.
   • get_weather: Fetches weather details (like temperature and humidity) for the given latitude and longitude using OpenWeatherMap's weather API.

2. Interaction Flow:

   • The user inputs a city name in the Streamlit app.
   • The LangChain agent processes the query, first converting the city name to latitude and longitude, then retrieving the corresponding weather details.
   • The weather details are then displayed in the app's UI.

3. User Query: The user can ask any weather-related query, and the agent will interpret the input, call the appropriate tool(s), and return the relevant data.

Example Usage

1. Enter a city name like "New York" in the text input field.
2. Click "Get Weather" to retrieve weather details such as temperature, humidity, and weather conditions for that city.

Code Explanation

• WeatherModel and LatitudeLongitudeModel: These are Pydantic models used to define the input schema for the weather and geolocation tools.
• get_weather and get_latitude_longitude: Functions that make HTTP requests to the OpenWeatherMap API to fetch weather details and geolocation data.
• LangChain Tools: The tools are wrapped using StructuredTool to match the required function signature and description for LangChain's agent.
• Streamlit UI: A simple UI to allow users to input a query and view the results.

Weather Assistant Images

Screen as soon as application starts

Screen as soon as you enter your question and press Get Weather

Screen with LLM response

Screen comparing weather in Pune and Bangalore

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

For questions or issues, feel free to open an issue on the GitHub repository or contact [your-email@example.com].

---

You can modify the repository URL and email as necessary!