NDDataStats¶

class astroimtools.stats.NDDataStats(nddata, sigma_clip=None, lower_bound=None, upper_bound=None, mask_value=None, mask_invalid=True)[source]¶

Bases: object

Class to calculate (sigma-clipped) image statistics on NDData objects.

Set the sigma_clip keyword to perform sigma clipping.

Parameters:

nddataNDData: NDData object containing the data array (and an optional mask) on which to calculate statistics. Masked pixels are excluded when computing the image statistics.
sigma_clipastropy.stats.SigmaClip instance, optional: A SigmaClip object that defines the sigma clipping parameters. If None then no sigma clipping will be performed (default).
lower_boundfloat, optional: The minimum data value to include in the statistics. All pixel values less than lower_bound will be ignored. None means that no lower bound is applied (default).
upper_boundfloat, optional: The maximum data value to include in the statistics. All pixel values greater than upper_bound will be ignored. None means that no upper bound is applied (default).
mask_valuefloat, optional: A data value (e.g., 0.0) to be masked. mask_value will be masked in addition to any input mask.
mask_invalidbool, optional: If True (the default), then any unmasked invalid values (e.g. NaN, inf) will be masked.

Examples

>>> import numpy as np
>>> from astropy.nddata import NDData
>>> from astroimtools import NDDataStats
>>> data = np.arange(10)
>>> data[0] = 100.
>>> nddata = NDData(data)
>>> stats = NDDataStats(nddata)
>>> stats.mean
14.5
>>> stats.std  
28.605069480775605
>>> stats.mad_std  
3.706505546264005
>>> from astropy.stats import SigmaClip
>>> sigclip = SigmaClip(sigma=2.5)
>>> stats = NDDataStats(nddata, sigma_clip=sigclip)
>>> stats.mean
5.0
>>> stats.std  
2.581988897471611
>>> stats.mad_std  
2.965204437011204

Attributes Summary

`biweight_location`	The biweight location of the pixel values.
`biweight_midvariance`	The biweight midvariance of the pixel values.
`kurtosis`	The kurtosis of the pixel values.
`mad_std`	A robust standard deviation using the median absolute deviation (MAD).
`max`	The maximum pixel value.
`mean`	The mean of pixel values.
`median`	The median of the pixel values.
`min`	The minimum pixel value.
`mode`	The mode of the pixel values.
`npixels`	The number of good (unmasked/unclipped) pixels.
`nrejected`	The number of rejected (masked/clipped) pixels.
`skew`	The skew of the pixel values.
`std`	The standard deviation of the pixel values.

Attributes Documentation

biweight_location¶: The biweight location of the pixel values.

biweight_midvariance¶: The biweight midvariance of the pixel values.

kurtosis¶: The kurtosis of the pixel values.

mad_std¶

A robust standard deviation using the median absolute deviation (MAD). The MAD is defined as median(abs(a - median(a))).

The standard deviation estimator is given by:

\[\sigma \approx \frac{\textrm{MAD}}{\Phi^{-1}(3/4)} \approx 1.4826 \ \textrm{MAD}\]

where \(\Phi^{-1}(P)\) is the normal inverse cumulative distribution function evaluated at probability \(P = 3/4\).

max¶: The maximum pixel value.

mean¶: The mean of pixel values.

median¶: The median of the pixel values.

min¶: The minimum pixel value.

mode¶

The mode of the pixel values.

The mode is estimated simply as (3 * median) - (2 * mean).

npixels¶: The number of good (unmasked/unclipped) pixels.

nrejected¶: The number of rejected (masked/clipped) pixels.

skew¶: The skew of the pixel values.

std¶: The standard deviation of the pixel values.

Navigation

NDDataStats¶