Benchmark Results

These plots collect the largest benchmark runs that directly compare fastmob with scikit-mobility (skmob). They are useful as a quick visual snapshot of fastmob behavior on 4M-row trajectory workloads.

The numbers are environment-specific, so they should be read as benchmark evidence for this run rather than as a universal performance guarantee.

Preprocessing

Preprocessing - Pandas (Speed)

4M-row pandas preprocessing speed benchmark

Preprocessing - Pandas (Memory) (Still needs adjusts)

4M-row pandas preprocessing memory benchmark

Preprocessing - Polars (Speed)

4M-row polars preprocessing speed benchmark

Preprocessing - Polars (Memory) (Still needs adjusts)

4M-row polars preprocessing memory benchmark

Measures - Individual

Measures - Individual - Pandas (Speed)

4M-row pandas measures individual speed benchmark

Measures - Individual - Pandas (Memory) (Still needs adjusts)

4M-row pandas measures individual memory benchmark

Measures - Individual - Polars (Speed)

4M-row polars measures individual speed benchmark

Measures - Individual - Polars (Memory) (Still needs adjusts)

4M-row polars measures individual memory benchmark

Measures - Collective

Measures - Collective - Pandas (Speed)

4M-row pandas measures collective speed benchmark

Measures - Collective - Pandas (Memory) (Still needs adjusts)

4M-row pandas measures collective memory benchmark

Measures - Collective - Polars (Speed)

4M-row polars measures collective speed benchmark

Measures - Evaluation

Measures - Evaluation - (Speed) - Mostly numpy/scipy wrappers

measures evaluation speed benchmark

Models

Models Agents - Speed

Simple models speed benchmark

Models Agents - Speed (No rust core yet...)

Simple models speed benchmark

Privacy

Privacy - Pandas (Speed)

4M-row pandas privacy speed benchmark

Privacy - Polars (Speed)

4M-row polars privacy speed benchmark

Pre-Sorted Benchmark Results

These refer to special metrics which benefit from sorted paths.