Skip to content

machine learning web app game where the user competes against the AI in picking stocks

License

Notifications You must be signed in to change notification settings

austintackaberry/stocks

Repository files navigation

StockIT

StockIT is a machine learning web app game that allows the user to compete agiainst the AI in buying/selling stocks. Random stocks are chosen and a random date in the past is chosen. The stock price starts trending from that date and continues for 365 days. While the stock price trends, the user competes against the AI in buying/selling stocks.

Pandas and Scikit Learn was used to develop a linear regression model to predict stock price. D3 and React were used to plot the data and update the leaderboard. Optimized for desktop and mobile.

Getting Started

In order to run a dev server on your local host, you will need to first install all the required npm packages for the frontend.

cd client

npm install

Now that you have installed all the required frontend packages, you can start the frontend server on port 3000:

npm start

Next you will need to install the required python packages:

cd ..

Install the virtual environment directory:

python3 -m venv venv

Activate the virtual environment

. venv/bin/activate

Install the requirements

pip install -r requirements.txt

Now you can open up an additional terminal and start up a server for the backend on port 5000:

export FLASK_APP=app.py

export FLASK_ENV=development

flask run

Now you are all set up!

Built With

  • React
  • D3
  • Python
  • Flask
  • Pandas
  • Scikit Learn

Authors

Austin Tackaberry

License

This project is licensed under the MIT License - see the LICENSE.md file for details

About

machine learning web app game where the user competes against the AI in picking stocks

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published