Model implementation in production with python

Question

I built a machine learning model of binary classification in python.

It works on my laptop (e.g. command line tool). Now I want to deploy it in production on a separate server in my company. It has to take inputs from another server (C# application), make some calculations and return outputs back to it.

My question is what are the best practices of doing such thing in production? As I know it can be done through TCP/IP connection.

I am new in this field and I don't know the terms used here. So can anybody guide me?

Thanks.

Answer 1

I would say it depends on your infrastructure and how can the other application (C#) can communicate.

The easiest way in my opinion would be through a REST API (http request). There are some tools in different languages to create REST endpoints easily and request REST endpoints.

For example, in python, you can request the content of a URL like this: What is the quickest way to HTTP GET in Python?

But it depends on what you have on the C# side. Can you update the C# code?

Here are a range of solutions:

• REST API: need to expose REST endpoints on the communicating "service".
  • in C#: https://learn.microsoft.com/en-us/aspnet/web-api/overview/older-versions/build-restful-apis-with-aspnet-web-api
  • in python, I would recommend django framework if you need to create a server (but if the python only request things and don't serve as a server, you may not need it)
• message queue like rabbitmq or zeromq, but it requires an external service to manage queues and messages
• TCP/IP socket like you suggested, but it requires to manage yourself those connections