Cluster 3.0 can also be run as a command line program. This may be useful if you want to run Cluster 3.0 on a remote server, and also allows automatic processing a large number of data files by running a batch script. Note, however, that the Python and Perl interfaces to the C Clustering Library may be better suited for this task, as they are more powerful than the command line program (see the manual for the C Clustering Library at http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/cluster/cluster.pdf).
The GUI version of Cluster 3.0 can be used as a command line program by applying the appropriate command line parameters. You can also compile Cluster 3.0 without GUI support (if you will be using it from the command line only) by downloading the source code from http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/cluster, and running
The executable is called
cluster. To run this program, execute
in which the options consist of the following command line parameters:
a: Subtract the mean of each row
m: Subtract the median of each row
a: Subtract the mean of each column
m: Subtract the median of each column
m: Pairwise complete- (maximum-) linkage (default)
s: Pairwise single-linkage
c: Pairwise centroid-linkage
a: Pairwise average-linkage
For the command line options -g, -e, the following integers can be used to specify the distance measure:
By default, no clustering is done, allowing you to use
cluster for normalizing a data set only.