pygsti.extras.drift.signal.bartlett_spectrum

pygsti.extras.drift.signal.bartlett_spectrum#

bartlett_spectrum(x, numspectra, counts=1, null_hypothesis=None, transform='dct')#

Calculates the Bartlett power spectrum. This involves splitting the data into disjoint sections of the same length, and generating a power spectrum for each such section, and then averaging all these power spectra.

Parameters:

  • x (array) – The data to calculate the spectrum for.

  • numspectra (int) – The number of “chunks” to split the data into, with a spectra calculated for each chunk. Note that if len(x) / num_spectra is not an integer, then not all of the data will be used.

  • counts (int, optional) – The number of “clicks” per time-step in x, used for standarizing the data.

  • null_hypothesis (array, optional) – The null hypothesis that we’re looking for violations of. If left as None then this is the no-drift null hypothesis, with the static probability set to the mean of the data.

  • transform (str, optional) – The transform to use the generate the power spectra.

Returns:

The Bartlett power spectrum

Return type:

array