I’m going to say something that’ll upset a lot of people: pandas had its run. Polars is just better. I don’t mean that lightly. I spent years writing pandas code. I taught pandas to junior developers. I built production systems on pandas. But after migrating several data pipelines to Polars and DuckDB over the past year, I can’t go back. The performance difference isn’t incremental — it’s a different universe.
No comments yet. Log in to reply on the Fediverse. Comments will appear here.