AI Stock Data Retriever

Overview

The AI Stock Data Retriever is a Streamlit application that integrates with the llama3.2:1b model (via the Ollama library) and Yahoo Finance API. This app processes natural language queries to retrieve stock data, including historical prices and the Price-to-Earnings (PE) ratio.

Features

• Natural Language Query Handling: Users can input queries like "Get 1 year stock prices of TCS," and the AI processes the query to determine the stock symbol and required data.
• Stock Price Retrieval: Fetch historical stock prices for a specified duration using the Yahoo Finance API.
• PE Ratio Retrieval: Get the current Price-to-Earnings ratio of a stock.
• Streamlit Integration: A user-friendly interface for query input and displaying results.

Snaps:

Setup Instructions

Prerequisites

1. Python 3.7+
2. Required libraries:
   • ollama
   • yfinance
   • pandas
   • streamlit

Installation

1. Clone the repository or save the script to your local machine.
2. Install the required dependencies:

   pip install ollama yfinance pandas streamlit

Starting the App

1. Run the Streamlit app:

   streamlit run src/trading_agent.py

2. Open the provided URL in your browser to interact with the app.

---

Code Details

Key Components

1. get_stock_prices(symbol: str, duration: str) -> pd.DataFrame:

   • Fetches historical stock prices for a given symbol and duration.
   • Uses Yahoo Finance API to query data.

2. current_pe_ratio(symbol: str) -> int:

   • Retrieves the current Price-to-Earnings ratio of a stock using Yahoo Finance.

3. AI Integration:

   • The llama3.2:1b model processes the user's query and determines the required data.
   • The model calls the appropriate Python functions (get_stock_prices or current_pe_ratio) using function calling capabilities provided by the Ollama library.

4. Streamlit Interface:

   • Accepts user input and displays the results interactively.

---

How It Works

1. User Query: The app accepts a natural language query about a stock.
2. AI Processing: The query is passed to the llama3.2:1b model via the Ollama library.
3. Function Execution:
   • Based on the AI’s response, the appropriate function (get_stock_prices or current_pe_ratio) is executed.
4. Result Display:
   • If the function returns a DataFrame (e.g., stock prices), it is displayed in a table.
   • For single values (e.g., PE ratio), the result is shown as text.

---

Example Queries

• "Get 1 year stock prices of TCS."
• "What is the current PE ratio of Infosys?"
• "Fetch 1 month stock prices for Reliance."

---

Troubleshooting

1. Ollama Connection Issues:
   • Ensure Ollama is running locally.
   • Verify the llama3.2:1b model is available and loaded in Ollama.
2. Yahoo Finance API Errors:
   • Confirm the stock symbol and duration are valid.
   • Check your internet connection for API requests.

---

Future Enhancements

• Add support for multiple stock exchanges and suffixes.
• Implement error handling for invalid queries or symbols.
• Extend functionality to include other stock metrics, such as dividends or EPS.

Let me know if you need further customization or clarification!