The purpose of this project is to create an dashboard environement for a bank to be able to use their data to decide whether to grant credit to its customers. The data come from kaggle : https://www.kaggle.com/c/home-credit-default-risk/data
Description of the files : The first 2 files will be deployed in an online VM with all the required data. The needed data and scripts on the VM are updated on VM reboot.
-api.py is for the api used for prediction
-dashapp.py is the dash application
-Projet7-featuring.ipynb is the feature ingeneering from the data from kaggle
-Projet7-classifier-choice.ipynb is some different classifier tests to choose the most adapted
-Projet7-gridsearchs-on-chosen-model.ipynb is the gridsearch on the best model previously found and saving model localy with mlflow -> final_model file is the so called model.