Project no longer being maintained
Retrospection: Catboost heavily overfits just as magically as XGBoost despite lower RMSE,Higher Accuracy KFold and further testing is required, so you are better off using something like an LinearRegression or an LSVM/C Thanks
Refer report for details such as architecture,methdology,etc.
A Streamlit❤️ web app that predicts the chance of admission into masters program based on various factors using a Flask API with Catboost model running as a background process on Windows .
Refer demo video.
python -m pip install -r requirements.txt
To run the API as a background process on Windows follow the instructions mentioned here
or one could open another console using tmux or in VSCode.
Run this file which serves as the entry-point to the Streamlit frontend.
streamlit run streamlit-app/streamlit_app.py
Following code boldy assumes the use of Windows. Since a Windows process is being spawned for serving the REST api. Also all file path need to be changed to match your dir structure.