dash.math.decompose
Decompose an ensemble into mean and deviations
Syntax
Description
Calculates the mean across an ensemble, and also returns the deviation of each ensemble member from the mean. Acts along the second dimension of the input array. Includes NaN values in all means.
… = dash.match.decompose(X, dim) Indicate which dimension is the ensemble dimension. The mean and deviations will be calculated across this dimension. If not specified, acts along the second dimension.
Input Arguments
X
numeric array, [? x nMembers x ?]
The ensemble that should be decomposed. An N-dimensional numeric array.
Output Arguments
Xmean
numeric array [? x 1 x ?]
The mean of the ensemble. A numeric array. The ensemble dimension of the input X will have a length of 1 in the output Xmean.
Xdeviations
numeric array [? x nMembers x ?]
The deviations of each ensemble member from the ensemble mean. Has the same size as the input X. The sum of the deviations across the ensemble dimension will be 0.