All projects

Stock Forecasting App

Machine learning

GOOGpricemapredictedtrain 80%test

A small Streamlit app for exploring stock forecasts. Enter a ticker and it downloads the price history with yfinance, then charts the closing price against its 50-, 100- and 200-day moving averages.

The model is a Keras network trained in the accompanying notebook, using the previous 100 days of prices to predict the next. The app plots its predictions against the actual prices to show how well the model performs.

Built with

  • Python
  • TensorFlow / Keras
  • Streamlit
  • yfinance
  • scikit-learn
  • Matplotlib
  • Jupyter