deepextractor.utils.signal

Module Contents

deepextractor.utils.signal.whitened_snr_scaling(glitch, snr, srate=4096)

Scale a glitch signal to the target SNR in the whitened frame.

deepextractor.utils.signal.quality_factor_conversion(Q, f_0)

Convert quality factor Q and central frequency f_0 to decay time tau.

deepextractor.utils.signal.rescale(x)

Rescale each row of x to the range [-1, 1].

deepextractor.utils.signal.butter_lowpass(cutoff, fs, order=5)

deepextractor.utils.signal.butter_highpass(cutoff, fs, order=5)

deepextractor.utils.signal.butter_filter(data, fs, order=5)

Apply a bandpass (20–1024 Hz) Butterworth filter to data.

deepextractor.utils.signal.custom_whiten(self, psd, low_frequency_cutoff=None, return_psd=False, **kwds)

Return a whitened PyCBC TimeSeries.

This function is designed to be used with a PyCBC TimeSeries instance (as a monkey-patched method). Pass self as the TimeSeries object.

Parameters:

• psd (FrequencySeries) – The power spectral density used for whitening.

• low_frequency_cutoff (float, optional) – Low frequency cutoff for the inverse spectrum truncation.

• return_psd (bool, optional) – If True, return the PSD alongside the whitened data.

Returns:

• white (TimeSeries) – The whitened time series.

• psd (FrequencySeries, optional) – The PSD used (only returned if return_psd=True).

deepextractor.utils.signal.generate_gaussian_noise(mean, std_dev, num_samples, sample_shape)

Generate Gaussian noise samples as a numpy array.