window-ops 0.0.13

Implementations of window operations such as rolling and expanding.

Window ops

Naive and fast implementations of common window operations.

This library is intended to be used as an alternative to pd.Series.rolling and pd.Series.expanding to gain a speedup by using numba optimized functions operating on numpy arrays. There are also online classes for more efficient updates of window statistics.

Install

pip install window-ops

How to use

Transformations

For a transformations n_samples -> n_samples you can use {[seasonal_](rolling|expanding)}_{(mean|max|min|std)} on an array.

Benchmarks

pd.__version__

'1.2.3'

n_samples = 1_000  # array size
window_size = 8  # for rolling operations
season_length = 7  # for seasonal operations
execute_times = 1_000 # number of times each function will be executed

Average times in milliseconds.

display_dataframe(times, fmt='{:.2f}')

	window_ops	pandas
rolling_mean	0	0.17
rolling_max	0.01	0.19
rolling_min	0.01	0.23
rolling_std	0.01	0.22
expanding_mean	0	0.13
expanding_max	0	0.13
expanding_min	0	0.13
expanding_std	0.01	0.14
seasonal_rolling_mean	0	2.62
seasonal_rolling_max	0.02	3.04
seasonal_rolling_min	0.02	2.85
seasonal_rolling_std	0.01	2.37
seasonal_expanding_mean	0	1.9
seasonal_expanding_max	0.01	1.79
seasonal_expanding_min	0.01	1.81
seasonal_expanding_std	0.01	2.45

display_dataframe(speedups, fmt='{:.0f}')

	times faster
rolling_mean	76
rolling_max	14
rolling_min	21
rolling_std	33
expanding_mean	44
expanding_max	32
expanding_min	32
expanding_std	19
seasonal_rolling_mean	632
seasonal_rolling_max	201
seasonal_rolling_min	173
seasonal_rolling_std	322
seasonal_expanding_mean	494
seasonal_expanding_max	353
seasonal_expanding_min	339
seasonal_expanding_std	238

Online

If you have an array for which you want to compute a window statistic and then keep updating it as more samples come in you can use the classes in the window_ops.online module. They all have a fit_transform method which take the array and return the transformations defined above but also have an update method that take a single value and return the new statistic.

Benchmarks

Average time in milliseconds it takes to transform the array and perform 100 updates.

display_dataframe(times.to_frame(), '{:.2f}')

	average time (ms)
RollingMean	0.07
RollingMax	0.09
RollingMin	0.09
RollingStd	0.24
ExpandingMean	0.08
ExpandingMax	0.03
ExpandingMin	0.02
ExpandingStd	0.08
SeasonalRollingMean	0.18
SeasonalRollingMax	0.14
SeasonalRollingMin	0.19
SeasonalRollingStd	0.25
SeasonalExpandingMean	0.09
SeasonalExpandingMax	0.06
SeasonalExpandingMin	0.06
SeasonalExpandingStd	0.09