Practical Optimisations for Pandas

Discover techniques for writing efficient pandas code in this 46-minute EuroPython 2020 conference talk. Learn to identify performance bottlenecks, implement computational efficiency strategies, and optimize memory usage in pandas. Explore various optimizations including data processing, reading data, using NumExpr, subselecting, concatenating, appending, sorting, filtering, compiling with Cython, leveraging pure Python and vectorized methods, managing memory footprint, implementing concurrency and parallelism, and utilizing other frameworks. Gain insights into additional resources and techniques for enhancing pandas performance in data science applications.

Intro
Overview
Optimizations
Processing Data
Reading Data
Type
NumExpression
Subselect
Concatenate
Append
Sorting
Filtering
Compiling
cyton
cyto
Pure Python
Vectorized Methods
Memory Footprint
Memory Footprint Example
Concurrency Parabolism
Other Techniques
Other Frameworks
Techniques
Additional Resources
Questions
Ending