Polars

Polars is the fast successor to Pandas. Written in Rust, query-optimized, and parallelized. v1.0 (2024) signaled production readiness.

When to Use

• Performance: 10-100x faster than Pandas on large data.
• Lazy Evaluation: Define a query plan and execute it efficiently (.collect()).
• Streaming: Process datasets larger than RAM.

Core Concepts

Lazy API (.lazy())

Builds a query plan. Polars optimizes it (predicate pushdown) before reducing it.

Expressions

pl.col("a").sum() is an expression, not a direct calculation.