This application was developped as a part of my Master's in Data Science with Openclassrooms. The code for the dashboard and the dockerfile can be found in this repository although the dashboard is deployed using streamlit cloud and can be found here.

• Click the link mentioned above to use the interactive dashboard. Type in a client ID to search the database for an existing client and retrieve the information and modify it if necessary.
• Sumit the information to get a prediction on whether the client can repay his loan back or not, with a probability percentage.
• Also explore which factors contributed to this decision and how the client compares to the rest of the clients in the database

The dashboard makes API calls to an API hosted on render which provides free hosting. The API is called from a docker image which is built using this code.

Since the API is hosted for free, if the API doesn't get any incoming requests for 15 minutes, it goes to sleep and and reactivates when a new request is made. Note that the reactivation might take upto a minute.