QUICK CLUSTER: Implement the /SAVE sub-command.

[pspp] / tests / language / stats / quick-cluster.at

diff --git a/tests/language/stats/quick-cluster.at b/tests/language/stats/quick-cluster.at
index 75dd52cb19f89a0a584ab3281bc9c24334ae7953..34294468c62679269e7e3cdea9fc7440774938ce 100644 (file)
--- a/tests/language/stats/quick-cluster.at
+++ b/tests/language/stats/quick-cluster.at
@@ -1,3 +1,19 @@
+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])
@@ -29,14 +45,13 @@ z,F8.0
 
 Table: Final Cluster Centers
 ,Cluster,
-,,
 ,1,2
-,,
 x,6.50,19.20
 y,7.50,3250.00
 z,6.50,5058.00
 
 Table: Number of Cases in each Cluster
+,,Count
 Cluster,1,6
 ,2,5
 Valid,,11
@@ -44,31 +59,31 @@ Valid,,11
 AT_CLEANUP
 
 AT_SETUP([QUICK CLUSTER with large data set])
+AT_KEYWORDS([slow])
 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
 Table: Final Cluster Centers
 ,Cluster,,,
-,,,,
 ,1,2,3,4
-,,,,
-x,.00,.00,.00,3.00
+x,NaN,NaN,NaN,3.00
 
 Table: Number of Cases in each Cluster
+,,Count
 Cluster,1,0
 ,2,0
 ,3,0
-,4,500000
-Valid,,500000
+,4,50000
+Valid,,50000
 ])
 AT_CLEANUP
 
@@ -91,7 +106,7 @@ end input program.
 
 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])
@@ -106,7 +121,7 @@ end loop.
 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])
@@ -128,7 +143,7 @@ begin data.
 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])
@@ -144,7 +159,7 @@ begin data.
 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])
@@ -155,6 +170,12 @@ AT_CLEANUP
 
 
 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. Because 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.
@@ -169,16 +190,15 @@ 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 *.
@@ -194,19 +214,28 @@ begin data.
 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 13  pspp-s.csv > top-s.csv])
+AT_CHECK([head -n 13  pspp-pw.csv > top-pw.csv])
+AT_CHECK([diff top-s.csv top-pw.csv])
+
 
-AT_CHECK([diff pspp-s.csv pspp-pw.csv], [0])
+AT_CHECK([grep Valid pspp-s.csv], [0], [Valid,,11
+])
+
+AT_CHECK([grep Valid pspp-pw.csv], [0], [Valid,,12
+])
 
 
 AT_CLEANUP
@@ -249,7 +278,7 @@ AT_CLEANUP
 
 AT_SETUP([QUICK CLUSTER /PRINT subcommand])
 AT_DATA([quick-cluster.sps], [dnl
-data list notable list /cluster (A8) x (F) y (F).
+data list notable list /cluster (A8) x y (F8.0).
 begin data.
 A 10.45 9.38
 A 10.67 9.17
@@ -318,21 +347,18 @@ QUICK CLUSTER x y
 AT_CHECK([pspp -O format=csv quick-cluster.sps], [0], [dnl
 Table: Initial Cluster Centers
 ,Cluster,,
-,,,
 ,1,2,3
-,,,
-x,0,0,1
-y,0,1,0
+x,-11,-12,11
+y,-12,11,11
 
 Table: Final Cluster Centers
 ,Cluster,,
-,,,
 ,1,2,3
-,,,
 x,-10,-10,10
 y,-10,10,10
 
 Table: Number of Cases in each Cluster
+,,Count
 Cluster,1,20
 ,2,19
 ,3,18
@@ -400,3 +426,162 @@ Case Number,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
+
+
+
+AT_SETUP([QUICK CLUSTER with save])
+AT_DATA([quick-cluster.sps], [dnl
+DATA LIST notable LIST /x y z.
+BEGIN DATA.
+22,2930,4099
+17,3350,4749
+22,2640,3799
+20, 3250,4816
+15,4080,7827
+4,5,4
+5,6,5
+6,7,6
+7,8,7
+8,9,8
+9,10,9
+END DATA.
+QUICK CLUSTER x y z
+  /CRITERIA=CLUSTER(2) MXITER(20)
+  /SAVE = CLUSTER (cluster) DISTANCE (distance).
+
+list.
+])
+
+AT_CHECK([pspp -O format=csv quick-cluster.sps], [0], [dnl
+Table: Final Cluster Centers
+,Cluster,
+,1,2
+x,6.50,19.20
+y,7.50,3250.00
+z,6.50,5058.00
+
+Table: Number of Cases in each Cluster
+,,Count
+Cluster,1,6
+,2,5
+Valid,,11
+
+Table: Data List
+x,y,z,cluster,distance
+22.00,2930.00,4099.00,2.00,1010.98
+17.00,3350.00,4749.00,2.00,324.79
+22.00,2640.00,3799.00,2.00,1399.00
+20.00,3250.00,4816.00,2.00,242.00
+15.00,4080.00,7827.00,2.00,2890.72
+4.00,5.00,4.00,1.00,4.33
+5.00,6.00,5.00,1.00,2.60
+6.00,7.00,6.00,1.00,.87
+7.00,8.00,7.00,1.00,.87
+8.00,9.00,8.00,1.00,2.60
+9.00,10.00,9.00,1.00,4.33
+])
+AT_CLEANUP
+
+
+AT_SETUP([QUICK CLUSTER with single save])
+AT_DATA([quick-cluster.sps], [dnl
+DATA LIST notable LIST /x y z.
+BEGIN DATA.
+22,2930,4099
+17,3350,4749
+22,2640,3799
+20, 3250,4816
+15,4080,7827
+4,5,4
+5,6,5
+6,7,6
+7,8,7
+8,9,8
+9,10,9
+END DATA.
+QUICK CLUSTER x y z
+  /CRITERIA=CLUSTER(2) MXITER(20)
+  /SAVE = DISTANCE.
+
+list.
+])
+
+AT_CHECK([pspp -O format=csv quick-cluster.sps], [0], [dnl
+Table: Final Cluster Centers
+,Cluster,
+,1,2
+x,6.50,19.20
+y,7.50,3250.00
+z,6.50,5058.00
+
+Table: Number of Cases in each Cluster
+,,Count
+Cluster,1,6
+,2,5
+Valid,,11
+
+Table: Data List
+x,y,z,QCL_0
+22.00,2930.00,4099.00,1010.98
+17.00,3350.00,4749.00,324.79
+22.00,2640.00,3799.00,1399.00
+20.00,3250.00,4816.00,242.00
+15.00,4080.00,7827.00,2890.72
+4.00,5.00,4.00,4.33
+5.00,6.00,5.00,2.60
+6.00,7.00,6.00,.87
+7.00,8.00,7.00,.87
+8.00,9.00,8.00,2.60
+9.00,10.00,9.00,4.33
+])
+AT_CLEANUP
+
+
+dnl This one was noticed to crash at one point.
+AT_SETUP([QUICK CLUSTER crash on bizarre input])
+AT_DATA([badn.sps], [dnl
+data list notable list /x.
+begin da\a*
+22
+17
+22
+22
+15
+4,
+5,
+6,
+7,j8,
+9,
+end data.
+
+quick cluster x
+" /criteria=cluster(2) mxiter(20)
+  /save = distance 
+  .
+
+list.
+])
+
+AT_CHECK([pspp -O format=csv badn.sps], [1], [ignore])
+
+AT_CLEANUP