15 releases (4 breaking)

new 0.5.4 Oct 28, 2024
0.5.3 Oct 25, 2024
0.4.3 Oct 18, 2024
0.3.1 Jul 30, 2024
0.1.2 Feb 20, 2024

#9 in #forecasting

Download history 54/week @ 2024-07-08 4/week @ 2024-07-15 252/week @ 2024-07-29 1/week @ 2024-08-05 3/week @ 2024-08-12 4/week @ 2024-09-02 9/week @ 2024-09-09 1/week @ 2024-09-16 23/week @ 2024-09-23 4/week @ 2024-09-30 11/week @ 2024-10-07 743/week @ 2024-10-14 397/week @ 2024-10-21

1,158 downloads per month
Used in augurs

MIT/Apache

13KB
208 lines

Seasonality detection for time series

augurs-seasons contains methods for detecting seasonality or periodicity in time series.

It currently contains implementations to do so using periodograms, similar to the seasonal Python package.

Usage

use augurs::seasons::{Detector, PeriodogramDetector};

# fn main() {
let y = &[
    0.1, 0.3, 0.8, 0.5,
    0.1, 0.31, 0.79, 0.48,
    0.09, 0.29, 0.81, 0.49,
    0.11, 0.28, 0.78, 0.53,
    0.1, 0.3, 0.8, 0.5,
    0.1, 0.31, 0.79, 0.48,
    0.09, 0.29, 0.81, 0.49,
    0.11, 0.28, 0.78, 0.53,
];
// Use the detector with default parameters.
let periods = PeriodogramDetector::default().detect(y);
assert_eq!(periods[0], 4);

// Customise the detector using the builder.
let periods = PeriodogramDetector::builder()
    .min_period(4)
    .max_period(8)
    .threshold(0.8)
    .build()
    .detect(y);
assert_eq!(periods[0], 4);
# }

Credits

This implementation is based heavily on the seasonal Python package. It also makes heavy use of the welch-sde crate.

License

Dual-licensed to be compatible with the Rust project. Licensed under the Apache License, Version 2.0 <http://www.apache.org/licenses/LICENSE-2.0> or the MIT license <http://opensource.org/licenses/MIT>, at your option.

Dependencies

~4.5MB
~88K SLoC