Welcome to ML Ops Quickstart
Contents
Welcome to ML Ops Quickstart#
ML Ops Quickstart is a tool for initializing Machine Learning projects following ML Ops best practices.
Setting up new repositories is a time-consuming task that involves creating different files and
configuring tools such as linters, docker containers and continuous integration pipelines.
The goal of mloq
is to simplify that process, so you can start writing code as fast as possible.
mloq
generates customized templates for Python projects with focus on Maching Learning. An example of
the generated templates can be found in mloq-template.