Metadata-Version: 2.1
Name: eventextreme
Version: 0.5.1
Summary: A package for extracting extreme events
Home-page: https://github.com/liuquan18/EventExtreme
Author: Quan Liu
Author-email: quan.liu@mpimet.mpg.de
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: pandas
Requires-Dist: numpy
Requires-Dist: scipy

![logo](logo.png)

# Overview
Extracting extreme events from daily time series data.


Such extreme events are defined as the values exceeding a certain threshold, and lasting for a certain duration.

Extremes events are identified by attributes such as:

    - extreme_start_time
    - extreme_end_time
    - extreme_duration
    - sign_start_time
    - sign_end_time
    - sign_duration
    - max
    - min
    - mean

![plot](plot.png)

The algorithm has some robust features, such as:

(P: positive sign, N: negative sign, E: extreme value)

    - single opposite valued data are ignored. e.g, P,P,P,N,P,P,P -> P,P,P,P,P,P
    - Multiple extreme events within one sign event are considered as one event. e.g, P,P,P,E,P,E,P,P,P is considered as one event.

# Installation
```bash
pip install eventextreme
```
