gwadama.tat#
tat.py
Time analysis toolkit.
- gwadama.tat.find_merger(h: ndarray) int[source]#
Estimate the index position of the merger in the given strain.
This function provides a rough estimate of the merger index position by locating the maximum of the absolute amplitude of the gravitational wave signal in the time domain. It assumes that the merger roughly corresponds to this peak, which holds for certain clean or high-SNR simulated CBC gravitational waves.
- Warning:
This function may be replaced in the near future by a more formal estimator with a better model, such as a Gaussian fit for binary mergers.
- Caution:
This is a very ad-hoc method and may not be accurate for all datasets, especially depending on the sensitivity of the detector. This method assumes that the peak amplitude in the time domain corresponds closely to the merger, which may not hold for lower-SNR signals or noisy data.
- Parameters:
- hnp.ndarray
The gravitational wave strain data.
- Returns:
- int
The index of the estimated merger position in the strain data.
- gwadama.tat.find_time_origin(times: ndarray) int[source]#
Find the index position of the origin of a time array.
It is just a shortcut for np.argmin(np.abs(times)).
- Parameters:
- timesNDArray
Time array.
- Returns:
- _int
Index position of the time origin (0).
- gwadama.tat.gen_time_array(t0, t1, sr)[source]#
Generate a time array with constant sampling rate.
Extension of numpy.arange which takes care of the case when an extra sample is produced due to round-off errors. When this happens, the extra sample is cut off.
- Parameters:
- t0, t1: float
Initial and final times of the array: [t0, t1).
- sr: int
Sample rate.
- length: int
Length of the final time array in samples. If due to round-off errors the length of the array is longer, it will be adjusted.
- Returns:
- times: NDArray
Time array.
- gwadama.tat.pad_time_array(times: ndarray, pad: int | tuple) ndarray[source]#
Extend a time array by ‘pad’ number of samples.
- Parameters:
- times: NDArray
Time array.
- pad: int | tuple
If int, number of samples to add on both sides. If tuple, number of samples to add on each side.
- Returns:
- NDArray
Padded time array.
Notes
Computes again the entire time array.
Due to round-off errors some intermediate time values might be slightly different.
- gwadama.tat.resample(strain: ndarray, time: ndarray | int, sample_rate: int, full_output=True) tuple[ndarray, int, int][source]#
Resample a single strain in time domain.
Resample strain’s sampling rate using an interpolation in the time domain for upscalling to a constant rate, and then decimate it to the target rate.
The upscaled sample rate is chosen as the minimum common multiple between the next integer value of the maximum sampling rate found in the original strain, and the target sample rate.
- Parameters:
- strain: 1d-array
Only one strain.
- time: 1d-array | int | float
Time points. If an Int or Float is given, it is interpreted as the former sampling rate, and assumed to be constant.
- sample_rate: int
Target sample rate. NOTE: It cannot be fractional.
- full_output: bool, optional
If True, also returns the new time points, the upscaled sampling rate, and the factor down.
- Returns:
- strain: 1d-array
Strain at the new sampling rate.
- time: 1d-array, optional
New time points.
- sr_up: int, optional
Upscaled sample rate.
- factor_down: int, optional
Factor at which the signal is decimated after the upscalling.