Idiomatic Pandas

on YouTube ↗28m

" summary

Uses Pandas 1.2.3 on a ~40,000-row, 83-column auto dataset and shows concrete wins: convert integer/float columns (cylinders from float to int), parse dates and turn low-cardinality strings into categorical to cut memory from 19 MB to 7 MB, create derived columns (automatic, speed, engine flags), build a reusable tweak function pipeline, prefer chained operations + assign to avoid SettingWithCopy and in-place myths, avoid apply for numeric work because vectorized ops are about 50x faster, and use groupby/agg/unstack for clear aggregations and plots.

AI-written summary · about

" chapters (6)

1. 0:00Idiomatic Pandas
2. 1:58Correct data types
3. 14:23Chaining operations
4. 19:13Avoid apply
5. 22:41Group by and pivot tables
6. 27:31Summary

" related talks

• 35mIdiomatic PandasMatt Harrison · Global Artificial Intelligence Virtual Conference 2022
• 32mPandas May Be Slow, but Pandas Doesn't Have to BeAndrew Vaccaro · PyColorado 2019
• 29mMaking Pandas FlyIan Ozsvald · PyData Fest Amsterdam 2020
• 40mBeyond Pandas: The Great Python Dataframe ShowdownJuan Luis Cano Rodríguez · PyData London 2022
• 44mHigh Perfomance PandasLuis David Camacho Gonzalez · PyCon Latam 2019
• 35mFast Pandas Performance Tricks You Wish You Knew EarlierAllen Y, SOORAJ ITS · PythonAsia 2026
• 25mPandas on Jupyter: Improving Your Data Analysis WorkflowNick Macro · PyOhio 2018
• 21mHow to use Pandas EfficientlyHossein Mortazavi · PyData Yerevan 2022