spm1d

spm1d.util

Utility module

This module contains a variety of convenience functions, including:

  • get_dataset
  • interp
  • p_corrected_bonf
  • p_critical_bonf
  • smooth

get_dataset

spm1d.util.get_dataset(*args)[source]

Warning

Deprecated

get_dataset is deprecated and will be removed from future versions of spm1d. Please access datasets using the “spm1d.data” interface.


interp

spm1d.util.interp(y, Q=101)[source]

Simple linear interpolation to n values.

Parameters:
  • y — a 1D array or list of J separate 1D arrays
  • Q — number of nodes in the interpolated continuum
Returns:
  • Q-component 1D array or a (J x Q) array
Example:
>>> y0 = np.random.rand(51)
>>> y1 = np.random.rand(87)
>>> y2 = np.random.rand(68)
>>> Y  = [y0, y1, y2]
>>> Y  = spm1d.util.interp(Y, Q=101)

p_corrected_bonf

spm1d.util.p_corrected_bonf(p, n)[source]

Bonferroni-corrected p value.

Warning

This correction assumes independence amongst multiple tests.

Parameters:
  • p — probability value computed from one of multiple tests
  • n — number of tests
Returns:
  • Bonferroni-corrected p value.
Example:
>>> p = spm1d.util.p_corrected_bonf(0.03, 8)    # yields p = 0.216

p_critical_bonf

spm1d.util.p_critical_bonf(alpha, n)[source]

Bonferroni-corrected critical Type I error rate.

Warning

This crticial threshold assumes independence amongst multiple tests.

Parameters:
  • alpha — original Type I error rate (usually 0.05)
  • n — number of tests
Returns:
  • Bonferroni-corrected critical p value; retains alpha across all tests.
Example:
>>> p = spm1d.util.p_critical_bonf(0.05, 20)    # yields p = 0.00256

smooth

spm1d.util.smooth(Y, fwhm=5.0)[source]

Smooth a set of 1D continua. This method uses scipy.ndimage.filters.gaussian_filter1d but uses the fwhm instead of the standard deviation.

Parameters:
  • Y — a (J x Q) numpy array
  • fwhm — Full-width at half-maximum of a Gaussian kernel used for smoothing.
Returns:
  • (J x Q) numpy array
Example:
>>> Y0  = np.random.rand(5, 101)
>>> Y   = spm1d.util.smooth(Y0, fwhm=10.0)

Note

A Gaussian kernel’s fwhm is related to its standard deviation (sd) as follows:

>>> fwhm = sd * sqrt(8*log(2))