Metrohm Vision Manual - Manual (Page 99)

▪▪▪▪▪▪▪ 97
4. From the lower part of the Edit Sample Selection Parameters, dialog box choose the filter
options. Only the spectra fitting these criteria (detector cell type and wavelength range) will
be let through sample selection. Click on OK button.
5. If the Random Selection method was chosen, define the distribution of spectra between sets
and click OK. Skip the next two points.
6. If selection other than random was chosen, from the Edit Math Properties box, choose
appropriate parameters for the selection method and math treatment, and then click OK.
7. In the Sample Selection Options dialog box, optionally check the Find Redundant and Find
Outliers boxes, configure the appropriate parameters of outlier and redundant sample
detection, and click OK.
Checking the Reset Product Mean box will reject outliers from the sample set before
redundant samples are found.
8. The sample selection algorithm will proceed with calculations and display results.
Manual Sample Selection
Manual Sample Selection allows users to move samples between library product sets. To move
samples, select samples to be moved and drag and dropped them to appropriate set on the library
tree.
Note: Manual Sample Selection has to be done after the sample selection algorithm is applied, but
before results are saved.
Visual Display of Results
The right side of the screen contains a series of graphs, the number and kind depending on your
choice of sample selection method. The results utilize a uniform color code to denote spectra. The
red color designates outliers, and light blue redundant samples. The dark blue samples are the
training set spectra. The left hand side of the screen shows the library tree and the distribution of
samples between the training, acceptance, and rejection sets.