Skip to main content

Python interface to harminv

Project description

This is a python interface to harminv, a library for performing harmonic inversion on a signal.



  • libharminv-dev

  • numpy


Make sure that you have the requirements above, then

pip install pharminv

See below for help with installing harminv.


harminv.invert contains the basic functionality of the harminv tool. This function should remain stable but the API elsewhere is subject to change.

import numpy as np
import harminv

time = np.linspace(0, 1, 1000)

signal = np.cos(2 * np.pi * time * 10) + np.cos(2 * np.pi * time * 20)

inversion = harminv.invert(signal, fmin=1, fmax=100, dt=0.001)

inversion.freq        # mode frequencies
inversion.decay       # decay rates
inversion.Q           # Q factor
inversion.amplitudes  # absolute amplitudes
inversion.phase       # phase shift


harminv was written by Steven G. Johnson and is licensed under the GNU GPLv2+, copyright 2005 by the Massachusetts Institute of Technology.

This python interface to harminv is licensed under the standard BSD 3-clause license.

Installing libharminv

On Debian / Ubuntu this is available with

sudo apt-get install libharminv-dev

Otherwise you’ll need to build it yourself. Download the sources from here.

You need to install libharminv somewhere that python can find it at runtime. This should happen by default (installs to /usr/local), but you may need to change this to /usr:


Now install with the following:

./configure --with-pic --enable-shared --prefix=$PREFIX
make install

User installation

If you don’t have admin rights on your system you’ll need to install to a directory that you have access to. Here I’ll assume that is $HOME/.local.

Here is how I do it (using Anaconda python with the mkl libraries on CentOS 5):

./configure --with-pic --enable-shared --prefix=$HOME/.local --with-blas=$HOME/.local/lib/
make install

Note that I’ve linked to openblas. I installed this separately. If you have BLAS/LAPACK installed on your system elsewhere you might not need to do this. There might be a way to link against Anaconda mkl libraries but I don’t know what it is.

Now you need to set LD_LIBRARY_PATH to include your directory. I have this in my bashrc:

export LD_LIBRARY_PATH=$HOME/.local

Finally, when you install pharminv, you need to tell pip where to look for your installation of libharminv:

CFLAGS="-I$HOME/.local/include -L$HOME/.local/lib" pip install pharminv

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pharminv-0.4.tar.gz (55.0 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page