wfs-match, short for waveform-match, is a numerical relativity waveform parameter estimation script.
A customized algorithm, taylored for parameter estimation of numerical relativity waveforms. However, it can also be used on cleaned observational waveforms. It is implemented in python using MPI (Message Parsing Interface), and is thus capable of parallel execution and scaling to large systems.
This is a template based match algorithm that minimizes a least-squares figure of merit, between the input vector and every template in the bank. It then returns the best matching template, along with its parameters.
The bank construction can be from any source, which defaults to pycbc/lalsuite waveform generation from available approximants. Also offers functionality to store the generated templates in hdf5 format (in parallel if parallel hdf5 is available).
Outputs:
Best matching template, its parameters, errors (if true values are available).
Waveforms aligned in time and phase.
Ambiguity functions.
Color map of the match discribution.
The output of this routine can further be piped into an inference procedure..