template_bank_generator_HM.TemplateBank.set_waveform_conditioning

Back to Class page

Summary

Finds normalization factor, support, and shift for whitened waveforms in bank if they aren’t already defined \*NOTE\* for PSD estimation, need: fftsize \* dt > chunktime

Signature

def set_waveform_conditioning(self, fftsize, dt, wt_filter_fd = None, min_support = None)

Input variables

Name	Type	Default	Description
fftsize			(int) Number of samples in FFT for the template
dt			(float) Sampling interval of template (in seconds)
wt_filter_fd		None	Frequency domain whitening filter (~irfft of 1/ASD). Lives on rfftfreq(fftsize, dt). If None, defaults to version in bank
min_support		None

Output variables

Return annotation	Docstring type	Description
None		Updates 1. self.fftsize 2. self.dt 3. self.wt_filter_fd 4. self.support_whitened_wf: TD support of waveform (in units of dt). Hardcoded to params.DEF_MAX_WFDURATION 5. self.shift_wf: Shift applied for weight on the right of filter (in units of dt) 6. self.normfac: Normalization factor to divide waveform \* whitening filter by for convolution with whitened data and filter parameters

Docstring

Finds normalization factor, support, and shift for whitened waveforms
in bank if they aren't already defined
*NOTE* for PSD estimation, need:  fftsize * dt > chunktime
:param fftsize: (int) Number of samples in FFT for the template
:param dt: (float) Sampling interval of template (in seconds)
:param wt_filter_fd:
    Frequency domain whitening filter (~irfft of 1/ASD). Lives on
    rfftfreq(fftsize, dt). If None, defaults to version in bank
:return: Updates
         1. self.fftsize
         2. self.dt
         3. self.wt_filter_fd
         4. self.support_whitened_wf:
                TD support of waveform (in units of dt). Hardcoded to
                params.DEF_MAX_WFDURATION
         5. self.shift_wf:
                Shift applied for weight on the right of filter
                (in units of dt)
         6. self.normfac:
                Normalization factor to divide waveform *
                whitening filter by for convolution with whitened data
         and filter parameters