scipy.ndimage.mean

scipy.ndimage.mean(input, labels=None, index=None)

Calculate the mean of the values of an array at labels.

Parameters:

input : array_like

Array on which to compute the mean of elements over distinct regions.

labels : array_like, optional

Array of labels of same shape, or broadcastable to the same shape as input. All elements sharing the same label form one region over which the mean of the elements is computed.

index : int or sequence of ints, optional

Labels of the objects over which the mean is to be computed. Default is None, in which case the mean for all values where label is greater than 0 is calculated.

Returns:

out : list

Sequence of same length as index, with the mean of the different regions labeled by the labels in index.

See also

ndimage.variance, ndimage.standard_deviation, ndimage.minimum, ndimage.maximum, ndimage.sum, ndimage.label

Examples

>>> from scipy import ndimage
>>> a = np.arange(25).reshape((5,5))
>>> labels = np.zeros_like(a)
>>> labels[3:5,3:5] = 1
>>> index = np.unique(labels)
>>> labels
array([[0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0],
       [0, 0, 0, 1, 1],
       [0, 0, 0, 1, 1]])
>>> index
array([0, 1])
>>> ndimage.mean(a, labels=labels, index=index)
[10.285714285714286, 21.0]