That’s how I use it, but it includes some very nice package publishing capabilities that I may explore one of these days.

Poetry is a dependency and package manager, and a virtual environment manager.

Jump to the tldr for the steps and not the explanation. I’ve been skeptical of migrating since I haven’t had issues with pipenv yet, but the momentum of Poetry seems to be gaining and I’ve recently updated my tooling to use Poetry. I’ve been a very happy pipenv user for a while now (technologically speaking), and have had very few issues with it, however, it’s pretty much a defunct package and Poetry is the latest solution on the scene. There’s Conda, pip, pipenv, virtualenv, etc. If you’ve been in the Data Science and Python world for more than a day, you’re probably very familiar with the plethora of ways to manage Python environments. These are my current set of tools that I use routinely for Machine Learning, Data Analysis, and Software Engineering. Tools change often, and if I decide there is a better way, I’ll update this post. I’m writing this blog post to myself, mostly to remind myself what I did last time to configure my environment. remembering how the hell you set it up last time.static checking, run-time error catching.

multiple Python environments, maintaining independency between different projects and versions.reproducibility, keeping track of installed packages.

Visual studio code auto format python