The repository contains a solution for the price prediction problem for stock data. Using deep neural networks, a model was created that quite nicely predicts stock price data.
In this mini project, Tata stock data was used (link). The data is in csv format, contains Opening, Closing, High and Low prices for trading days between 2014 and 2018.
You can check the workflow in notebook.ipynb file. This notebook starts from loading and visualization, then moves to preprocessing/preparation, next it describes how to build the model and finally shows some results. In the notebook, some pickled objects will be generated which may be used later on.
In app.py there is a Dash app that visualizes predictions and ground truth values of the Tata stock prices. You may launch it by typing:
python app.py
and then open the app in the browser.
You may check some of my other projects here.