+dnl PSPP - a program for statistical analysis.
+dnl Copyright (C) 2017 Free Software Foundation, Inc.
+dnl
+dnl This program is free software: you can redistribute it and/or modify
+dnl it under the terms of the GNU General Public License as published by
+dnl the Free Software Foundation, either version 3 of the License, or
+dnl (at your option) any later version.
+dnl
+dnl This program is distributed in the hope that it will be useful,
+dnl but WITHOUT ANY WARRANTY; without even the implied warranty of
+dnl MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+dnl GNU General Public License for more details.
+dnl
+dnl You should have received a copy of the GNU General Public License
+dnl along with this program. If not, see <http://www.gnu.org/licenses/>.
+dnl
AT_BANNER([QUICK CLUSTER])
AT_SETUP([QUICK CLUSTER with small data set])
AT_SETUP([QUICK CLUSTER with large data set])
AT_DATA([quick-cluster.sps], [dnl
input program.
-loop #i = 1 to 500000.
+loop #i = 1 to 50000.
compute x = 3.
end case.
end loop.
end file.
end input program.
-QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (100).
+QUICK CLUSTER x /CRITERIA = CLUSTER(4) NOINITIAL.
])
AT_CHECK([pspp -o pspp.csv quick-cluster.sps])
AT_CHECK([cat pspp.csv], [0], [dnl
,,,,
,1,2,3,4
,,,,
-x,.00,.00,.00,3.00
+x,NaN,NaN,NaN,3.00
Table: Number of Cases in each Cluster
Cluster,1,0
,2,0
,3,0
-,4,500000
-Valid,,500000
+,4,50000
+Valid,,50000
])
AT_CLEANUP
weight by w.
-QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (100).
+QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (10).
])
AT_CHECK([pspp -o pspp-w.csv qc-weighted.sps])
end file.
end input program.
-QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (100).
+QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (10).
])
AT_CHECK([pspp -o pspp-unw.csv qc-unweighted.sps])
2
end data.
-QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (100).
+QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (10).
])
AT_CHECK([pspp -o pspp-m.csv quick-miss.sps])
2
end data.
-QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (100).
+QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (10).
])
AT_CHECK([pspp -o pspp-nm.csv quick-nmiss.sps])
AT_SETUP([QUICK CLUSTER with pairwise missing])
+
+dnl This test runs two programs, which are identical except that one
+dnl has an extra case with one missing value. Becuase the syntax uses
+dnl NOINITIAL and NOUPDATE, the results should be identical except for
+dnl the final classification.
+
AT_DATA([quick-s.sps], [dnl
data list notable list /x * y *.
begin data.
3.4 3
3.5 2.5
3.1 2.0
-3.9 2.5
-3.8 2.0
end data.
QUICK CLUSTER x y
- /CRITERIA = CLUSTER(3) MXITER (100)
+ /PRINT = INITIAL
+ /CRITERIA = CLUSTER(3) NOINITIAL NOUPDATE
.
])
-AT_CHECK([pspp -O format=csv quick-s.sps | tail -5 > pspp-s.csv])
+AT_CHECK([pspp -O format=csv quick-s.sps > pspp-s.csv])
AT_DATA([quick-pw.sps], [dnl
data list notable list /x * y *.
3.4 3
3.5 2.5
3.1 2.0
-3.9 .
-3.8 .
+. 2.3
end data.
QUICK CLUSTER x y
- /CRITERIA = CLUSTER(3) MXITER (100)
+ /CRITERIA = CLUSTER(3) NOINITIAL NOUPDATE
+ /PRINT = INITIAL
/MISSING = PAIRWISE
.
])
-AT_CHECK([pspp -O format=csv quick-pw.sps | tail -5 > pspp-pw.csv])
+AT_CHECK([pspp -O format=csv quick-pw.sps > pspp-pw.csv])
+
+AT_CHECK([head -n 18 pspp-s.csv > top-s.csv])
+AT_CHECK([head -n 18 pspp-pw.csv > top-pw.csv])
+AT_CHECK([diff top-s.csv top-pw.csv])
+
+
+AT_CHECK([grep Valid pspp-s.csv], [0], [Valid,,11
+])
-AT_CHECK([diff pspp-s.csv pspp-pw.csv], [0])
+AT_CHECK([grep Valid pspp-pw.csv], [0], [Valid,,12
+])
AT_CLEANUP
,,,
,1,2,3
,,,
-x,0,0,1
-y,0,1,0
+x,-11,-12,11
+y,-12,11,11
Table: Final Cluster Centers
,Cluster,,
])
AT_CLEANUP
+
+
+dnl Test for a crash which happened on bad input syntax
+AT_SETUP([QUICK CLUSTER -- Empty Parentheses])
+
+AT_DATA([empty-parens.sps], [dnl
+data list notable list /x * y *.
+begin data.
+1 2
+1 2.2
+end data.
+
+QUICK CLUSTER x y
+ /CRITERIA = CONVERGE()
+ .
+])
+
+AT_CHECK([pspp -o pspp.csv empty-parens.sps], [1], [ignore])
+
+AT_CLEANUP
+