Table 2 N L and N D for amine data set. The mean and standard deviation of the number of latent variables or degree of RR penalty (N L ) and the number of molecular descriptors (N D ) used to build the models for the the amine data set. The values of N L and N D are tuned by optimizing the Q2 value in an inner cross validation loop.
From: Unbiased descriptor and parameter selection confirms the potential of proteochemometric modelling
Filter | Regression | N L (mean ± std) | N D (mean ± std) |
|---|---|---|---|
no filter | PLS | 6.49 ± 1.48 | 12765 ± 0 |
no filter | RR | 1.13 ± 1.45 | 12765 ± 0 |
corrfilter | PLS | 7.00 ± 0.88 | 4748 ± 730 |
corrfilter | RR | 1.83 ± 1.18 | 4871 ± 640 |
PLSfilter | PLS | 6.18 ± 1.40 | 1933 ± 455 |
PLSfilter | RR | 2.10 ± 1.17 | 2136 ± 349 |