Hands-on Tutorial
Part 1 — End to End Machine Learning Model Deployment Using Flask
How to prepare the data and develop a machine learning model for the loan approval prediction app
It is the first part of the series of end-to-end machine learning model deployments. In this part, we have a lot of work to do in preparing the data set, creating a robust machine-learning model, and how to store the model for the production app. Some tricks and tricks in machine learning processes are included as you follow the tutorial.
Without further ado, let’s jump into the tutorial!
Our motivation
Suppose we are working at ABC bank and being asked by the data insight manager to create an automation tool to determine whether a customer who applies for a loan is granted or not. This project becomes a priority because the result is expected to make the loan approval verification becomes more efficient.
As a data scientist, you have a projection about this project and make a to-do list, starting from data acquisition, and machine learning model development to deployment. The end product must be an app because it will be implemented by non-technical users in your company.
