-
Cryptocurrency Explorer: Real-Time Prices tab
🚀 About the Project: Cryptocurrency Explorer
Inspiration:
As a data scientist with a passion for finance and technology, I wanted to create a project that merges both interests. Cryptocurrencies are rapidly evolving, and their volatile nature makes them perfect candidates for data-driven analysis and forecasting. Inspired by the growing need for easy access to crypto insights, I set out to build Cryptocurrency Explorer, a tool that offers users the ability to track, analyze, and predict the prices of the top cryptocurrencies.
What I Learned:
This project helped deepen my understanding of time-series forecasting using the ARIMA model, as well as the intricacies of working with APIs. Throughout the process, I learned about the limitations of free API services (like rate limits), the importance of caching API results to enhance performance, and the challenges of presenting real-time data in a user-friendly interface. Moreover, I gained more experience with Python libraries like pandas, Streamlit, and statsmodels to power the forecasting model and the overall app functionality.
How I Built It:
The app is built using Streamlit, a framework that simplifies the creation of interactive web apps using Python. It pulls data from the CoinGecko API to fetch real-time prices, historical data, and market trends for the top 100 cryptocurrencies.
I chose ARIMA for the price prediction feature due to its simplicity and effectiveness in short-term forecasting. Given the unpredictable and volatile nature of cryptocurrencies, I limited predictions to shorter horizons, as longer predictions tend to lose accuracy.
Features:
- Real-time Prices: Track current prices of popular cryptos, with metrics like 24-hour price change and market cap.
- Historical Analysis: View price trends for up to 365 days.
- Forecasting with ARIMA: Use ARIMA to predict future prices. (Best for short-term predictions.)
- Top Cryptocurrencies: Insights on top gainers and most expensive cryptos.
Challenges I Faced:
One of the main challenges was balancing the API’s rate limit with the functionality I wanted to offer. Since I used the free version of the CoinGecko API, I had to carefully optimize requests by caching responses and focusing on the top 100 cryptocurrencies to ensure the app performs efficiently without hitting the rate limit. Additionally, ensuring the ARIMA model produced meaningful forecasts, especially in such a volatile market, required tweaking and testing different configurations to find the right balance between complexity and performance.
Feel free to explore the app here: Cryptocurrency Explorer
Log in or sign up for Devpost to join the conversation.