API Template

This project provides a starting point for developing an API using FastAPI. Intended to be simultaneously scalable and quick to implement, this code allows for rapidly developing an API with minimal boilerplate, built-in validation, and easy deployment. It provides a structured yet flexible foundation for building robust, production-ready services with FastAPI's powerful asynchronous capabilities.

The intention is to allow data scientists / analysts to quickly make available API endpoints for more effective code modularization.

Server Initialization and Test Request

Functionalities are provided to easily launch the server and then make a request to verify that the boilerplate code is working .

To launch the server, simply run python main.py:

(zero) [main] m $  python main.py 
INFO:     Started server process [2765591]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)

...and file run.py includes a test case:

import requests

url = 'http://localhost:8000/main'
payload = {
    'param1': 'param1',
    'param2': 'param2',
}

response = requests.post(url, json=payload)
print(response.status_code)
print(response.json())

...after running run.py, the server log should look like:

(zero) [main] m $  python main.py 
INFO:     Started server process [2765591]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)
param1=param1 param2=param2
INFO:     127.0.0.1:36828 - "POST /main HTTP/1.1" 200 OK

Integrating Custom Functionality

In the given example, params param1 and param2 are sent to the server in run.py, and the printed to the terminal in endpoints.py.

Custom functionality can be implemented by modifying endpoints.py, where imports can be added and functionality implemented in endpoint functions like main_endpoint.

In the current state, the API functionality is intended to be provided locally (i.e. only programs on the same computer will be able to make requests). Exposing endpoints via a remote server will require relatively simple development steps, however the details of these steps are dependent on many factors, therefore more development work is required to launch the API on a remote server.