3 ms·

Python Pandas Ditches NumPy for Speedier PyArrow

• culebron21 1y ago
  Why do they call Pandas a legacy tech in the end of the article? What has come to replace it, that I don't know of?
  • jononor 1y ago
    Some are using polars for new code, where they might have used pandas before. I would not call pandas "legacy" though. It it still as ubiquitous and viable as before.
    • culebron21 1y ago
      Oh, yes, I tried it, it's faster, but ergonomics is worse. Every interaction is one notch harder.
  • aradox66 1y ago
    SQL lol
• bicepjai 1y ago
  Polars is amazing, but I would argue pandas is definitely not legacy :)