Z-score normalization is an algorithm that produces data with each feature (column) having zero mean and unit variance.
Given a set X of n feature vectors x 1 = (x 11 , … , x 1p ), ... , x n = (x n1 , … , x np ) of dimension p, the problem is to compute the matrix Y = (y i j ) of dimension n x p as following:
Intel DAAL provides two modes for computing the result matrix. You can enable the mode by setting the flag doScale to a certain position (for details, see Batch Processing > Algorithm Parameters). The mode may include:
Centering only. In this case, and no scaling is performed. After normalization, the mean of j-th component of result set (Y) j will be zero.
Centering and scaling. In this case, , where is the standard deviation of j-th component of set (X) j . After normalization, the mean of j-th component of result set (Y) j will be zero and its variance will get a value of one.
Some algorithms require normalization parameters (mean and variance) as an input. The implementation of Z-score algorithm in Intel DAAL does not return these values by default. Enable this option by setting the resultsToCompute flag. For details, see Batch Processing > Algorithm Parameters.