Parameters | Predictive Modeling | Display Frobenius Measure

Display Frobenius Measure
Specify this option to compute the Frobenius measure (displayed in terms of percentage error) between the raw additive and dominance matrices with their respectively low-rank matrices obtained either via randomized or truncated singular value decomposition (SVD).
The smaller the percentage error, the better is the approximation between the low-rank SVD and the raw matrix.
Note: This option is available only when the Compute near optimal low-rank randomized SVD for the additive and dominance matrices check box is checked.
To Specify This Option:
*
Make sure that the Compute near optimal low-rank randomized SVD for the additive and dominance matrices check box is checked.
*