Learning the Polars DataFrame Library!

[@KeithGalli]

Quick correction! The group_by example at 52:40 has an error with how we sort the resulting dataframe. The sort should happen outside the select so that the columns are sorted together. Proper results will give you "Water" as the most popular Type 1 with a count of 31. The corrected code can be found in exploring-polars.ipynb file (added to the github repo, linked in description) Thanks for tuning in today! Let me know if you have additional questions or topics that you would like to know about Polars :)

[@ici6308]

I am happy to be the first person to comment on this tutorial. Although I have not yet viewed the entire video yet, let me tell you that your tutorials are among the best on youtube in the field of data science. You are a great guy !

[@random-drops]

Can't agree more. Thanks, Keith!

[@KeithGalli]

Thank you both for the support!

[@oenriqueg]

X3

[@oenriqueg]

Thanks, Keith! You are amazing. I always see your videos with a great feeling. Greetings from Mexico.

[@KeithGalli]

¡Gracias por tus amables palabras!

[@drumpf4all]

Very cool! I haven’t heard of this library. I’ve heard of pandas for sure but not polars. Definitely interesting. Thank you.

[@Codechannel-i8x]

I love your videos, and was wondering whats your spec's?

[@KeithGalli]

Computer specs? Macbook Pro 14" with M2 chip & 16gb of ram 🙂

[@andybecker5001]

Polars is definitely faster with large files. You can switch back and forth by assigning your polars dataframe to pandas dataframe if you already have functions you built and like to use in pandas.

[@KeithGalli]

Cool to know about the dataframe switching, that's very helpful. I should have had a larger set of files prepared for this livestream, but I will make sure to properly benchmark performance in future tutorials that I prepare with the Polars library

[@KumR]

30