template_bank_generator_HM.TemplateBank.gen_wfs_td_from_fd#

Back to Class page

Summary#

Generates conditioned time-domain waveforms given frequency domain waveforms.

Signature#

def gen_wfs_td_from_fd(self, wfs_fd_original, dt_extra = 0, whiten = False, highpass = True, truncate = False, target_snr = None, fs_in = None, log = False, **conditioning_kwargs)

Input variables#
Name	Type	Default	Description
`wfs_fd_original`			… x fs_in array with frequency domain waveforms (can be a vector for n_wf = 1)
`dt_extra`		0	Increase roll by this time for safety (s)
`whiten`		False	Flag whether to return the whitened waveform
`highpass`		True	Flag indicating whether to high-pass filter the waveform
`truncate`		False	Flag indicating whether to truncate the waveform. Ensure that the wf is linear-free before truncating.
`target_snr`		None	Target SNR to enforce, pass None to not apply this If this is a scalar, then all waveforms in wfs_fd have this SNR If this is an array of shape wfs_fd.shape[:-1], then the individual waveforms have the desired SNRs If whiten == True, this is the norm of the truncated and possibly highpassed whitened waveform If whiten == False and truncate == True or highpass == True, there are losses associated with conditioning
`fs_in`		None	Array of input frequencies. None indicates fs_in = self.fs_fft
`log`		False	Flag indicating whether we passed the log of the waveforms in
`\\conditioning_kwargs`			If known, pass conditioning parameters (useful when whitening different waveforms), defaults to those in the bank Can use multiple normfacs if needed

Output variables#

Return annotation	Docstring type	Description
`None`		… x fftsize array of TD waveforms (can be vector for n_wf=1)

Docstring#

Generates conditioned time-domain waveforms given frequency domain
waveforms.
:param wfs_fd_original:
    ... x fs_in array with frequency domain waveforms
    (can be a vector for n_wf = 1)
:param dt_extra: Increase roll by this time for safety (s)
:param whiten: Flag whether to return the whitened waveform
:param highpass:
    Flag indicating whether to high-pass filter the waveform
:param truncate:
    Flag indicating whether to truncate the waveform.
    Ensure that the wf is linear-free before truncating.
:param target_snr:
    Target SNR to enforce, pass None to not apply this
    If this is a scalar, then all waveforms in wfs_fd have this SNR
    If this is an array of shape wfs_fd.shape[:-1], then the individual
    waveforms have the desired SNRs
    If whiten == True, this is the norm of the truncated and possibly
    highpassed whitened waveform
    If whiten == False and truncate == True or highpass == True, there
    are losses associated with conditioning
:param fs_in:
    Array of input frequencies. None indicates fs_in = self.fs_fft
:param log:
    Flag indicating whether we passed the log of the waveforms in
:param conditioning_kwargs:
    If known, pass conditioning parameters (useful when whitening
    different waveforms), defaults to those in the bank
    Can use multiple normfacs if needed
:return:
    ... x fftsize array of TD waveforms (can be vector for n_wf=1)