This program features a graphical user interface that connects to a custom server to fetch a predicted value for the inputted stock. It uses a LSTM to make predictions and learns by using the mean squared error loss function. Predictions are graphed for the inputted stock upon prediction completion.
- Clone this repository
- Execute the
pip install -r requirements.txt
command in your terminal on your local machine(sudo or administrator privileges might be required)
- Windows:
python mainGUI.py
- macOS and Linux:
python3 mainGUI.py
- Double click on mainGUI.py to execute using the Python 3 interpreter
- Important: server.py file must be executed on weberkcudafac, since this machine has a static ip and that ip is the default for the server
- Note: after cloning the repo onto weberkcudafac, the following command needs executed
pip install -r requirements.txt
- To execute: enter the terminal and input
python3 server.py
- To kill: enter
ctrl+c
- Jordan Davis and Ryan Harris