mkgrp | Create a new graphic structure |
<graphic> | name for the graphic structure |
[{-s <som> | | a TS-SOM structure |
-d <data>}] | a data frame |
setgdat | Connect a data frame to graphics |
<graphic> | name of the graphic structure |
-d <data> | a data frame |
setgcld | Connect a classified data frame to graphics |
<graphic> | name of the graphic structure |
-c <cldata> | a classified data frame |
Operation mkgrp creates a basic graphic structure. The graphic structure contains several subgraphs, which can be created manually with addsg, or according to a TS-SOM structure or a data frame.
The commands setgdat and setgcld connect a data frame and a classified data frame, respectively, to a graphic structure. Some graphical operations are able to use these as default parameters for <data> and <cldata>. In the case of TS-SOMs, a graphic structure can visualize any variables in neurons, and the classified data needs to be a SOM classification.
Example (ex7.1): The first command sequence creates a TS-SOM and a classified data and computes class statistics for the neurons.
... NDA> somtr -d boston -sout som1 -l 5 ... NDA> somcl -d boston -s som1 -cout cld1 NDA> clstat -d boston -c cld1 -dout clsta -avg -min -max ...
Then the graphic structure grp1 is created based on som1. Both the data frame containing statistics and classified data are connected to the graphic structure.
... NDA> mkgrp grp1 -s som1 NDA> setgdat grp1 -d clsta NDA> setgcld grp1 -c cld1
Example (ex7.2): In this example, a graphic structure is created for a data frame, in which case each data record in the frame will have its own subgraph.
NDA> load boston.dat NDA> selrec -d boston -dout highCrim -expr 'boston.crim' > 35; NDA> mkgrp grp1 -d highCrim ...