flask-to-fastapi-ml

A simple machine learning model deployed to Flask and FastAPI

Description

• The purpose of this repo is to assess Flask to FastAPI migration
• Model accepts 11 features and determines loan eligility

Disclaimer:

• The classification model used below should not be used in any real world scenario.
• The model itself is a toy example and is flawed. There is one heavily weighted feature.

Usage

cd /PATH/TO/FILES

python flask_app.py

or 

python fastapi_app.py

Interacting with Model

You have multiple options:

• Open interact_with_model.ipynb and run the cells
• For FastAPI, navigate to http://127.0.0.1:8000/docs and "Try it out"
• Try it with Postman
• Command line

Contents

├── data.csv #  Contains labelled training data 
├── fastapi_app.py #  FastAPI App - Imports * from functions.py
├── flask_app.py #  Flask App - Imports * from functions.py
├── functions.py #  Custom functions and variables.
├── grid_rfc.sav #  Random Forest Classifier
└── interact_with_model.ipynb #  Notebook for interacting with both models