With the release of Pandas v3 and several recommendations for Polars from Marc Garcia's blog, I'm curious about diving into Polars. As a new dad, my time is limited, so I'm hoping to find the best resource that gives me a solid foundation in Polars without requiring a lot of heavy reading. Any suggestions?
5 Answers
I made the switch to Polars last year. I had to rework an entire Pandas pipeline, which involved getting the hang of Polars' syntax and main methods.
First off, what projects are you working on? Understanding your needs might help in deciding if you should really switch to Polars. If your current setup in Pandas works fine and you're not facing performance issues, it might be better to stick with it for now and consider Polars for new projects. If performance is a concern, maybe profile your code and see if converting a specific slow part to Polars could work.
Honestly, just dive in and check out the documentation! They even have a section dedicated to migrating from Pandas, which might be helpful for you.
If you're familiar with Pandas already, you might find using LLMs like ChatGPT useful for translating your work into Polars syntax.
The docs are solid. They provide enough information to get you started with Polars without feeling overwhelmed.
I'm mainly analyzing data for my EV charging setup, which involves billing and system load analysis. I have around 5 million rows. I'm also considering using Polars at work for my epidemiology projects, which deal with much larger datasets, so lazy frames would be key there. It's just that I'm currently doing everything in R, so switching over will be a bigger change.