pykooh 0.3.2

Efficient implementatins of the Konno Ohmachi filter in Python

pykooh

Konno Ohmachi filter implemented in Numba.

This code implements Konno-Ohmachi spectral smoothing as defined in:

Konno, K. and Ohmachi, T., 1998. Ground-motion characteristics estimated
from spectral ratio between horizontal and vertical components of
microtremor. Bulletin of the Seismological Society of America, 88(1),
pp.228-241.

This code was originally written for smoothing sub-module in gmprocess by Bruce Worden. Dave Boore has provided notes on this topic, which also may be of interest. Notes regarding the characteristics of the Konno-Ohmachi filter and the implementation are provided in the implementation Jupyter Notebook.

Installation

pykooh is available via pip and can be installed with:

pip install pykooh

By default, pykooh uses numba for the fast implementation of the filter. Performance can be increased by using cython, but this requires a C complier. If a C compiler is available, install cython required dependencies with:

pip install pykooh[cython]

Usage

Smooth a signal using a bandwith of 30.

import pykooh
signal_smooth = pykooh.smooth(freqs, freqs_raw, signal_raw, 30)

Additional examples and comparison with obspy are provided in example.

Citation

Please cite this software using the following DOI.

Revision History

v0.3.2

• Change setup.py to install numpy prior to import.

v0.3.1

• Rename to pykooh

v0.3.0

• Rename to pykoom

• Add support for numba

• Make cython an optonal dependency

v0.2.5

• Packaging is hard. MANIFEST is fixed now.

v0.2.4

• Added History to MANIFEST.

v0.2.3

• Updated badges.

• Added tests for example and implemenation notebooks.

v0.2.2

• Moved Cython to a setup_requires

v0.2.1

• Fixed packaging issue

v0.2

• Added calculation of effective amplitude spectrum

v0.1.2

• Initial release