NBA Chatbot using Anthropic API

This project integrates the Anthropic API to create a chatbot for querying NBA statistics. It uses the nba_api library to fetch player and team statistics.

Features

• Fetch player information, including team and player ID, based on the player's name.
• Retrieve player statistics, like points scored in a season, free throw percentage, and current team.
• Get the number of league titles won by a team.

Setup

1. Install the required Python libraries:

pip install streamlit dotenv anthropic nba_api

2. Create a .env file in the project root and add your Anthropic API key:

ANTHROPIC_API_KEY=your_api_key_here

Usage

Run the Streamlit application:

streamlit run app.py

In the app, you can enter a player's name or team ID to fetch the respective data and statistics.

Code Overview

• AnthropicFunctionCalling: Class to handle API interaction and processing of NBA data.
• get_player_info, get_player_statistics, get_league_titles: Methods to fetch specific data from the NBA API.
• process_tool_call: Processes the function calls based on the tool name and input.
• calculate_cost: Calculates the cost of using the Anthropic API based on input and output tokens.
• main: The main function of the Streamlit app, handling the UI and session state.

Deployment

The application can be deployed on a web server or run locally to interact with the NBA data through a chat interface.

Notes

• Ensure you have a valid Anthropic API key set in your .env file.
• The cost of using the Anthropic API will depend on the number of tokens processed during the conversation.