Closed samuelm00 closed 2 years ago
Referring to this file https://github.com/elki-project/elki/blob/master/elki-clustering/src/main/java/elki/clustering/subspace/SUBCLU.java
form line 166 to 184 the algorithm generates all 1-dimensional clusters, but why do you run DBSCAN with the same parameters d times (the result should always be the same)?
166
184
Its running DBSCAN on d different subspaces, isn't it? If you enable the FINER debugging level, you should see that it is using different subspaces.
Referring to this file https://github.com/elki-project/elki/blob/master/elki-clustering/src/main/java/elki/clustering/subspace/SUBCLU.java
form line
166
to184
the algorithm generates all 1-dimensional clusters, but why do you run DBSCAN with the same parameters d times (the result should always be the same)?