Inspiration

• Inspiration:
  • Our inspiration stems from the desire to empower retail traders by providing them with tools to confidently invest in cryptocurrencies and other financial assets. We aimed to democratize access to market insights and trends, enabling individuals to make informed decisions in volatile markets.

What it does

• Functionality:
  • Our project allows users to visualize potential trends in the cryptocurrency market, equipping them to anticipate and navigate both bull and bear markets effectively. By leveraging machine learning techniques, we aim to provide users with actionable insights for their investment strategies.

How we built it

• Development Process:
  • We utilized machine learning algorithms, particularly LSTM RNNs, to analyze historical market data and predict future price movements. Data was sourced from Yahoo Finance, and we extensively studied documentation on LSTM RNNs to implement our predictive model effectively.

Challenges we ran into

• Challenges Faced:
  • Training the LSTM RNN model proved to be computationally intensive and time-consuming, requiring significant resources. Additionally, we encountered difficulties in scaling the graphs accurately to reflect the data's magnitude and complexity.

Accomplishments that we're proud of

• Achievements:
  • We are proud to have developed a functional model that accurately predicts future cryptocurrency prices based on historical data. Our success in creating a reliable prediction tool demonstrates the potential of machine learning in financial analysis and decision-making.

What we learned

• Key Takeaways:
  • Through this project, we gained valuable insights into the computational demands of LSTM RNNs and the challenges of working with large-scale financial data. We also recognized the versatility of machine learning algorithms across various sectors, from finance to technology and beyond.

What's next for LSTM RNN

• Future Steps:
  • Moving forward, we aim to address the issue of data scaling to ensure accurate representation on graphs, particularly concerning the y-axis. Additionally, we plan to explore methods for undoing data scaling to facilitate more meaningful analysis. Furthermore, we aspire to enhance our model's capabilities by enabling it to predict future prices on specific dates with greater precision and accuracy.

Built With

• keras
• matplotlib
• numpy
• python
• sklearn
• tensorflow
• yfinance