pelt 0.3.1

Changepoint detection with Pruned Exact Linear Time

Changepoint detection with Pruned Exact Linear Time.

Usage

Python

from pelt import predict

predict(signal, penalty=20, segment_cost_function="l1", jump=10, minimum_segment_length=2)

Rust

use std::num::NonZero;
use pelt::{Pelt, SegmentCostFunction};

// Setup the structure for calculating changepoints
let pelt = Pelt::new()
  .with_jump(NonZero::new(5).expect("Invalid number"))
  .with_minimum_segment_length(NonZero::new(2).expect("Invalid number"))
  .with_segment_cost_function(SegmentCostFunction::L1);

// Do the calculation on a data set
let penalty = 10.0;
let result = pelt.predict(&signal[..], penalty)?;

Run locally

# Install maturin inside a Python environment
python3 -m venv .env
source .env/bin/activate
pip install maturin numpy

# Create a Python package from the Rust code
maturin develop --features python

# Open an interpreter
python

>>> from pelt import predict
>>> import numpy as np
>>> signal = np.array([np.sin(np.arange(0, 1000, 10))]).transpose()
>>> predict(signal, penalty=20)

Benchmarks

[!WARNING] Like all benchmarks, take these with a grain of salt.

Python

Comparison with ruptures:

Cost Function	Data Points	Data Dimension	Mean pelt	Mean ruptures	Times Faster
L2	100	1D	2.197 μs	3.125 ms	1422.4x
L2	100	2D	2.478 μs	3.160 ms	1275.0x
L2	1000	1D	126.394 μs	195.964 ms	1550.4x
L2	1000	2D	57.435 μs	95.860 ms	1669.0x
L2	10000	1D	21.401 ms	12.571 s	587.4x
L2	10000	2D	2.397 ms	1.699 s	708.8x
L1	100	1D	10.613 μs	4.804 ms	452.7x
L1	100	2D	21.625 μs	5.105 ms	236.1x
L1	1000	1D	321.511 μs	184.912 ms	575.1x
L1	1000	2D	2.152 ms	636.542 ms	295.9x
L1	10000	1D	13.598 ms	15.572 s	1145.2x
L1	10000	2D	85.147 ms	30.706 s	360.6x

Command

maturin develop --features python --release
python benches/bench_compare.py

Profile

Command

cargo build --example simple --profile profiling \
 && samply record target/profiling/examples/simple tests/signals-large.csv

Credits

• fastpelt
• ruptures