pyextremes

library2.5.0pypypi✓ verified 86d ago

A Python library for Extreme Value Analysis (EVA) of time series data. It implements methods for block maxima (BM) and peak-over-threshold (POT) approaches, including fitting, diagnostic plots, and return value estimation. Current stable version is 2.5.0; release cadence is irregular with occasional major/minor updates.

analytics

pip install pyextremes

INSTALL

IMPORT

SIG · PYEXTREMES

pyextremes

analyticspythonv2.5.0

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

get_extremes

✓ from pyextremes import get_extremes

✗ from pyextremes.extremes import get_extremes

get_extremes is a top-level function in v2.x; was previously in a submodule.

fit_extreme

✓ from pyextremes import fit_extreme

✗ from pyextremes.models import fit_extreme

fit_extreme is now top-level; submodule path removed in v2.1.0.

plot

✓ from pyextremes import plot

Basic extreme value analysis using block maxima on random Gumbel data.

import numpy as np from pyextremes import get_extremes, fit_extreme, plot # Generate sample data np.random.seed(42) data = np.random.gumbel(loc=10, scale=2, size=1000) # Extract extremes using block maxima data_block_maxima = get_extremes(data, method='BM', block_size=100) # Fit model results = fit_extreme(data_block_maxima, method='BM', distribution='gumbel') # Diagnostic plot plot(results, plot_type='return_value', return_period=[10, 50, 100])

Debug

Known issues

breakingIn v2.0.0, the API was redesigned: get_extremes and fit_extreme moved to top-level. Old imports from submodules will break.

fix

Update imports to top-level: from pyextremes import get_extremes, fit_extreme.

affects: <2.0.0

breakingIn v2.4.0, the 'pot' method argument in get_extremes changed name from 'threshold' to 'r' for return values. This may cause silent failure if using keyword arguments.

fix

Use named parameter 'r' instead of 'threshold' when calling get_extremes with method='POT'.

affects: >=2.4.0

gotchaThe default method for get_extremes is 'BM' (block maxima) with block_size=365. Users expecting 'POT' (peak-over-threshold) must specify method='POT' and provide a threshold/r value, otherwise results may be misleading.

fix

Explicitly set method and required parameters: get_extremes(data, method='POT', r=threshold_value).

affects: all

Errors

Common errors & fixes

ImportError: cannot import name 'get_extremes' from 'pyextremes'

Older installed version of pyextremes (<2.0.0) does not have get_extremes as top-level function.

fix

Upgrade pyextremes to latest version: pip install --upgrade pyextremes.

TypeError: get_extremes() got an unexpected keyword argument 'threshold'

In pyextremes >=2.4.0, the parameter for POT method is 'r', not 'threshold'.

fix

Replace keyword 'threshold' with 'r' in get_extremes call.

Upgrade

Version history

2.5.0latest on PyPI · released Feb 19, 2026

Audit

Dependencies

numpyrequiredCore dependency for numerical operations

scipyrequiredUsed for optimization and statistical distributions

matplotlibrequiredFor plotting diagnostic plots

Agent activity

25 hits · last 30 days

node

OpenAI (training)

Resources

No resource links recorded.