OPENCLASSROOMS - Data Scientist - Project 7
This repository contains files to build, run and deploy a machine learning web application which predicts the probabiliy for a customer to repay a loan.
The dataset used for this project is the Home Credit Default Risk, which includes information on hundreds of thousands loans.
├── LICENSE
├── README.md
│
├── api <- FastAPI files
│
├── dashboard <- Streamlit application
│
├── data
│ ├── external <- Data from third party sources (not uploaded on git)
│ ├── processed <- Data after analysis and feature engineering
│
├── docs <- Documentation and reports
│
├── mlruns <- MLFlow runs
│
├── models <- Trained and serialized models
│
├── notebooks <- Jupyter notebooks
│
├── tests <- Pytest unit tests (local)
- Python 3.x
- Jupyter Notebook
- NumPy
- Pandas
- Matplotlib / Seaborn
- Scikit-learn
- Imbalanced-learn
- XGBoost
- LightGBM
- Hyperopt
- SHAP
- Evidently
- MLFlow
- Streamlit
- FastAPI
- Docker
- GitHub Actions
- Azure Web App
- Application : https://scoringapp-dashboard.azurewebsites.net/ (stopped)