AT_BANNER([QUICK CLUSTER])

AT_SETUP([QUICK CLUSTER with small data set])
AT_DATA([quick-cluster.sps], [dnl
DATA LIST 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).
])
AT_CHECK([pspp -o pspp.csv quick-cluster.sps])
AT_CHECK([cat pspp.csv], [0], [dnl
Table: Reading free-form data from INLINE.
Variable,Format
x,F8.0
y,F8.0
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
Cluster,1,6
,2,5
Valid,,11
])
AT_CLEANUP

AT_SETUP([QUICK CLUSTER with large data set])
AT_DATA([quick-cluster.sps], [dnl
input program.
loop #i = 1 to 50000.
compute x = 3.
end case.
end loop.
end file.
end input program.
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,NaN,NaN,NaN,3.00

Table: Number of Cases in each Cluster
Cluster,1,0
,2,0
,3,0
,4,50000
Valid,,50000
])
AT_CLEANUP


AT_SETUP([QUICK CLUSTER with weights])
AT_DATA([qc-weighted.sps], [dnl
input program.
loop #i = 1 to 400.
 compute x = mod (#i, 4).
 compute w = 5.
 end case.
end loop.
loop #i = 1 to 400.
 compute x = mod (#i, 4).
 compute w = 3.
 end case.
end loop.
end file.
end input program.

weight by w.

QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (10).
])

AT_CHECK([pspp -o pspp-w.csv qc-weighted.sps])


AT_DATA([qc-unweighted.sps], [dnl
input program.
loop #i = 1 to 3200.
 compute x = mod (#i, 4).
 end case.
end loop.
end file.
end input program.

QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (10).
])

AT_CHECK([pspp -o pspp-unw.csv qc-unweighted.sps])

AT_CHECK([diff pspp-w.csv pspp-unw.csv], [0])

AT_CLEANUP

AT_SETUP([QUICK CLUSTER with listwise missing])
AT_DATA([quick-miss.sps], [dnl
data list notable list /x *.
begin data.
1
1
2
3
4
.
2
end data.

QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (10).
])

AT_CHECK([pspp -o pspp-m.csv quick-miss.sps])

AT_DATA([quick-nmiss.sps], [dnl
data list notable list /x *.
begin data.
1
1
2
3
4
2
end data.

QUICK CLUSTER x /CRITERIA = CLUSTER(4) MXITER (10).
])

AT_CHECK([pspp -o pspp-nm.csv quick-nmiss.sps])

AT_CHECK([diff pspp-m.csv pspp-nm.csv], [0])

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. 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.
1   2
1   2.2
1.1 1.9
1   9
1   10
1.3 9.5
0.9 8.9
3.5 2
3.4 3
3.5 2.5
3.1 2.0
end data.

QUICK CLUSTER x y 
	/PRINT = INITIAL
	/CRITERIA = CLUSTER(3) NOINITIAL NOUPDATE
	.
])

AT_CHECK([pspp -O format=csv quick-s.sps  > pspp-s.csv])

AT_DATA([quick-pw.sps], [dnl
data list notable list /x * y *.
begin data.
1   2
1   2.2
1.1 1.9
1   9
1   10
1.3 9.5
0.9 8.9
3.5 2
3.4 3
3.5 2.5
3.1 2.0
.   2.3
end data.

QUICK CLUSTER x y 
	/CRITERIA = CLUSTER(3) NOINITIAL NOUPDATE
	/PRINT = INITIAL
	/MISSING = PAIRWISE
	.
])

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([grep Valid pspp-pw.csv], [0], [Valid,,12
])


AT_CLEANUP



AT_SETUP([QUICK CLUSTER crash on bad cluster quantity])
AT_DATA([badn.sps], [dnl
data list notable list /x * y *.
begin data.
1   2
1   2.2
end data.

QUICK CLUSTER x y 
	/CRITERIA = CLUSTER(0)
	.
])

AT_CHECK([pspp -O format=csv badn.sps], [1], [ignore])

AT_CLEANUP

AT_SETUP([QUICK CLUSTER infinite loop on bad command name])
AT_DATA([quick-cluster.sps], [dnl
data list notable list /x y.
begin data.
1   2
1   2.2
end data.

QUICK CLUSTER x y /UNSUPPORTED.
])
AT_CHECK([pspp -O format=csv quick-cluster.sps], [1], [dnl
quick-cluster.sps:7.20-7.30: error: QUICK CLUSTER: Syntax error at `UNSUPPORTED'.
])
AT_CLEANUP



AT_SETUP([QUICK CLUSTER /PRINT subcommand])
AT_DATA([quick-cluster.sps], [dnl
data list notable list /cluster (A8) x (F) y (F).
begin data.
A 10.45 9.38
A 10.67 9.17
A 10.86 9.63
A 8.77 8.45
A 8.04 11.77
A 10.34 9.83
A 10.37 10.54
A 11.49 8.18
A 10.17 11.10
A 11.37 9.16
A 10.25 8.83
A 8.69 9.92
A 10.36 10.39
A 10.89 10.51
A 9.9 11.39
A 11.1 10.91
A 11.77 8.47
A 9.5 10.46
B -11.01 -9.21
B -10.8 -11.76
B -10.03 -10.29
B -9.54 -9.17
B -10.16 -9.82
B -10.01 -8.63
B -9.6 -10.22
B -11.36 -10.93
B -10.63 -10.97
B -9.53 -10.78
B -9.40 -10.26
B -10.76 -9.76
B -9.9 -10.11
B -10.16 -9.75
B -8.65 -11.31
B -10.10 -10.90
B -11.67 -9.89
B -11.11 -9.23
B -8.7 -8.43
B -11.35 -8.68
C -10.20 9.00
C -10.12 9.92
C -10.41 10.16
C -9.86 10.12
C -10.31 10.12
C -9.57 10.16
C -9.69 9.93
C -9.14 10.84
C -9.8 10.19
C -9.97 10.22
C -11.65 10.81
C -9.80 11.39
C -10.31 10.74
C -10.26 10.38
C -11.57 10.02
C -10.50 9.75
C -9.06 9.63
C -10.17 10.82
C -10.22 9.99
end data.

QUICK CLUSTER x y
  /CRITERIA=CLUSTERS(3)
  /PRINT=INITIAL CLUSTER.
])

AT_CHECK([pspp -O format=csv quick-cluster.sps], [0], [dnl
Table: Initial Cluster Centers
,Cluster,,
,,,
,1,2,3
,,,
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
Cluster,1,20
,2,19
,3,18
Valid,,57

Table: Cluster Membership
Case Number,Cluster
1,3
2,3
3,3
4,3
5,3
6,3
7,3
8,3
9,3
10,3
11,3
12,3
13,3
14,3
15,3
16,3
17,3
18,3
19,1
20,1
21,1
22,1
23,1
24,1
25,1
26,1
27,1
28,1
29,1
30,1
31,1
32,1
33,1
34,1
35,1
36,1
37,1
38,1
39,2
40,2
41,2
42,2
43,2
44,2
45,2
46,2
47,2
48,2
49,2
50,2
51,2
52,2
53,2
54,2
55,2
56,2
57,2
])

AT_CLEANUP