Use Rust’s Speed to Install Python Libraries Up to 100 Times Faster | by Eryk Lewinson | Mar, 2024

For data scientists and Python programmers, pip desires no introduction. As a bundle supervisor, it’s each the go-to reply or the place to start in our look for the proper reply available on the market.

Trivia alert 🚨: Whereas pip doesn’t require an introduction, I’ve solely merely found that it actually stands for “Pip Installs Packages” or “Preferred Installer Program”.

pip shouldn’t be the one bundle supervisor on the market. Beneath, you’ll discover essentially the most well-liked devices within the market:

• pip — the same old bundle supervisor for Python, which moreover comes preinstalled with Python.
• conda — a bundle and environment administration system designed notably for data science. Aside from Python, it might presumably moreover arrange packages written in several languages (for example, R).
• poetry — it targets to simplify Python dependency administration by providing choices akin to atmosphere pleasant dependency determination and digital environments.
• pipenv — this machine combines pip for bundle arrange with virtualenv for creating isolated environments.

And as this XKCD comic illustrates, there could also be on a regular basis room for a model new joiner!

At current, we’ll look into uv, which boasts being over 100 situations faster than pip. uv capabilities as a Python bundle installer, digital environment creator and a resolver. To realize its blazing tempo, it was inbuilt Rust. Furthermore, it was designed as a drop-in various for pip and pip-tools workflows.

That sounds promising! Sooner than putting it to the check out, let’s briefly briefly contact upon the excellence between an installer and resolver:

• bundle installers (e.g., pip) are used to place in, substitute, and take away Python packages from the atmosphere.
• bundle resolvers (e.g., pip-tools) make certain the consistency and reproducibility of dependencies by producing precise lists and resolving mannequin conflicts.

Sooner than we dive into testing, it’s value mentioning that I’m conducting the assessments on an M1 Mac Mini. And a typical disclaimer…