def generate(self):
"""Produce the table assuming that a procedure state is now in effect if it has any rows."""
privateproc = False
if self.rowcount > 0:
import spss
try:
table = spss.BasePivotTable(self.tabletitle, self.omssubtype)
except:
StartProcedure(_("Messages"), self.procname)
privateproc = True
table = spss.BasePivotTable(self.tabletitle, self.omssubtype)
if self.caption:
table.Caption(self.caption)
# Note: Unicode strings do not work as cell values in 18.0.1 and probably back to 16
if self.columnlabels != []:
table.SimplePivotTable(self.rowdim, self.rowlabels, self.coldim, self.columnlabels, self.columnvalues)
else:
table.Append(spss.Dimension.Place.row,"rowdim",hideName=True,hideLabels=True)
table.Append(spss.Dimension.Place.column,"coldim",hideName=True,hideLabels=True)
colcat = spss.CellText.String("Message")
for r in self.rowlabels:
cellr = spss.CellText.String(r)
table[(cellr, colcat)] = cellr
if privateproc:
spss.EndProcedure()
def report(self, labelsyntax):
# write report
if not labelsyntax:
print(_("""No value labels were generated."""))
return
if len(self.labelvars) > 1:
labelvars = self.labelvars
else:
labelvars = len(self.varstolabel) * [self.labelvars][0]
spss.StartProcedure("Generate Value Labels", "STATSVALLBLSFROMDATA")
cells = [[labelvars[i],
spss.CellText.Number(len(self.conflicts[vname]), spss.FormatSpec.Count),
#spss.CellText.Number(self.duplabels[vname], spss.FormatSpec.Count)]\
spss.CellText.Number(len(self.duplabels[vname]), spss.FormatSpec.Count)]\
for i,vname in enumerate(self.varstolabel)]
caption = []
if self.syntax:
caption.append(_("""Generated label syntax: %s""" % self.syntax))
if self.execute:
caption.append(_("""Generated label syntax was applied"""))
else:
caption.append(_("""Generated label syntax was not applied"""))
caption.append(_("""A conflict means that different labels would be applied to the same value."""))
caption.append(_("""A duplicate means that the same label was used for different values."""))
tbl = spss.BasePivotTable(_("""Value Label Generation"""), "VALLBLSFROMDATA",
caption="\n".join(caption))
tbl.SimplePivotTable(rowdim= _("""Variable"""), rowlabels=self.varstolabel,
collabels=[_("""Label Source"""), _("""Label Conflicts"""), _("""Duplicate Labels""")],
cells=cells)
spss.EndProcedure()
def issue_warning(msg):
"""Issue a warning that is displayed in an SPSS pivot table"""
spss.StartProcedure("Warning!")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row, "rowdim", hideLabels=True)
rowLabel = spss.CellText.String("1")
table[(rowLabel, )] = spss.CellText.String(msg)
spss.EndProcedure()
def display(variables, withvars, stats, matnames, clevel, missing, inclusion):
"""Display pivot table output for regular or split files
variables is the main variable list
withvars is None or a list of variables to correlate with
stats is the result structure
clevel is the confidence level
missing is listwise or pairwise
include is include or exclude for user missing values"""
spss.StartProcedure(_("Correlations"), "CICORRELATIONS")
tbl = spss.BasePivotTable(_("Correlations"), "CICORRELATIONS")
tbl.Caption(
_("""Missing value handling: %s, %s. C.I. Level: %s""") %
(missing, inclusion, clevel))
rowsplits = []
for v in spss.GetSplitVariableNames():
rowsplits.append(tbl.Append(spss.Dimension.Place.row, v))
nsplitvars = len(rowsplits)
var1 = tbl.Append(spss.Dimension.Place.row, _("Variable"))
var2 = tbl.Append(spss.Dimension.Place.row, _("Variable2"))
vlist = withvars and withvars or variables
col1 = tbl.Append(spss.Dimension.Place.column, _("Statistic"))
tbl.SetCategories(col1, [
CellText.String(_("Correlation")),
CellText.String(_("Count")),
CellText.String(_("Lower C.I.")),
CellText.String(_("Upper C.I.")),
CellText.String(_("Notes"))
])
vardict = spssaux.VariableDict()
for vcount, s in enumerate(stats):
for i, vv in enumerate(vlist):
j = i + (withvars is not None and len(variables))
if nsplitvars > 0:
rows = copy.copy(s.splitvars)
else:
rows = []
###rows.append(s.variable)
rows.append(
CellText.VarName(
vardict[s.variable.toString().rstrip()].VariableIndex))
vvitem = vardict[vv].VariableIndex
rows.append(CellText.VarName(vvitem)) # 2/9/2022
if s.ns[j].toNumber() > 10:
note = ""
elif s.ns[j].toNumber() <= 3:
note = _("Some items not computed")
else:
note = _("Normality assumption is not accurate")
statsi = [
s.corrs[j], s.ns[j], s.cis[j][0], s.cis[j][1],
CellText.String(note)
]
tbl.SetCellsByRow(rows, statsi)
###tbl.SetCellsByRow(rows, [Ctn(item) for item in statsi])
spss.EndProcedure()
def lsmon():
"""Create an SPSS pivot table of the license usage data returned by function lsmon_"""
features, unreserveds, reserveds, maxes, total_reserveds, total_unreserveds, hostname = lsmon_(
)
feature_labels = [feature_codes.get(f, "(Unknown)") for f in features]
asStr = spss.CellText.String
asInt = lambda value, fmt=spss.FormatSpec.Count: spss.CellText.Number(
value, fmt)
percentSpecs = [spss.FormatSpec.Percent] * len(features)
try:
stamp = time.strftime("%Y-%m-%d at %H:%M:%M")
spss.StartProcedure("SPSS license usage (%s)" % stamp)
table = spss.BasePivotTable("Table", "lsmon")
row = table.Append(spss.Dimension.Place.row, "feature #")
column = table.Append(spss.Dimension.Place.column, "Usage")
for feature in features:
table.SetCategories(row, asStr(feature))
table.SetCellsByColumn(asStr("label"), list(map(asStr,
feature_labels)))
table.SetCellsByColumn(asStr("unreserved\nn"),
list(map(asInt, unreserveds)))
percents = [
unreserveds[i] / float(total_unres + 10e-10) * 100
for i, total_unres in enumerate(total_unreserveds)
]
table.SetCellsByColumn(asStr("unreserved\n%"),
list(map(asInt, percents, percentSpecs)))
table.SetCellsByColumn(asStr("reserved\nn"),
list(map(asInt, reserveds)))
percents = [
reserveds[i] / float(total_res + 10e-10) * 100
for i, total_res in enumerate(total_reserveds)
]
table.SetCellsByColumn(asStr("reserved\n%"),
list(map(asInt, percents, percentSpecs)))
totals = [
unreserved + reserved
for unreserved, reserved in list(zip(unreserveds, reserveds))
]
table.SetCellsByColumn(asStr("total\nn"), list(map(asInt, totals)))
percents = [
(unreserveds[i] + reserveds[i]) / float(max_ + 10e-10) * 100
for i, max_ in enumerate(maxes)
]
table.SetCellsByColumn(asStr("total\n%"),
list(map(asInt, percents, percentSpecs)))
table.SetCellsByColumn(asStr("maximum\nn"), list(map(asInt, maxes)))
table.Caption("License server: %s." % hostname)
finally:
spss.EndProcedure()
def catvalues(mcset, varprefix, setname):
"""Convert MC set to MD set"""
#debugging
# makes debug apply only to the current thread
#try:
#import wingdbstub
#if wingdbstub.debugger != None:
#import time
#wingdbstub.debugger.StopDebug()
#time.sleep(1)
#wingdbstub.debugger.StartDebug()
#import thread
#wingdbstub.debugger.SetDebugThreads({thread.get_ident(): 1}, default_policy=0)
## for V19 use
### ###SpssClient._heartBeat(False)
#except:
#pass
if not mcset.startswith("$"):
raise ValueError(
_("""Only an MC set can be used with this procedure"""))
if not setname.startswith("$"):
setname = "$" + setname
resolver = Resolver()
allvars = resolver.resolve(mcset) # check existence and get all variables
resolver.close()
generatedvars, generatedvalues, generatedlabels = genSetsCategoryList(
mcset, allvars, resolver, setname, varprefix)
# list new variables and labels
StartProcedure(_("""Convert MC Set"""), "STATSMCSETCONVERT")
table = spss.BasePivotTable(title=_("""Variables Generated for Set %s""") %
setname,
templateName="STATSMCSETVARS")
table.SimplePivotTable(rowdim=_("Variable Name"),
rowlabels=generatedvars,
collabels=[_("Value Represented"),
_("Label")],
cells=list(zip(generatedvalues, generatedlabels)))
spss.EndProcedure()
def fleisskappaextension(variables, cilevel=95):
varnames = expandvarnames(variables)
vardict = spssaux.VariableDict(varnames)
if len(vardict) != len(varnames):
spss.StartProcedure(_("Fleiss Kappa"), "Fleiss Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row, "rowdim", hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""An invalid variable has been specified. This command is not executed."""
))
spss.EndProcedure()
elif len(varnames) < 2:
spss.StartProcedure(_("Fleiss Kappa"), "Fleiss Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row, "rowdim", hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""At least two variables must be specified. This command is not executed."""
))
spss.EndProcedure()
else:
try:
warntext = []
if cilevel < 50:
warntext.append(
_("CILEVEL cannot be less than 50%. It has been reset to 50%."
))
cilevel = 50
if cilevel > 99.999:
warntext.append(
_("CILEVEL cannot be greater than 99.999%. It has been reset to 99.999%."
))
cilevel = 99.999
if cilevel == int(cilevel):
cilevel = int(cilevel)
varlist = varnames[0]
for i in range(1, len(varnames)):
varlist = varlist + ' ' + varnames[i]
spss.Submit("PRESERVE.")
tempdir = tempfile.gettempdir()
spss.Submit("""CD "%s".""" % tempdir)
wtvar = spss.GetWeightVar()
if wtvar != None:
spss.Submit(r"""
COMPUTE %s=RND(%s).""" % (wtvar, wtvar))
spss.Submit(r"""
EXECUTE.""")
wtdn = GetWeightSum(varnames)
else:
wtdn = spss.GetCaseCount()
maxloops = wtdn + 1
spss.Submit(
"""SET PRINTBACK=OFF MPRINT=OFF OATTRS=ENG MXLOOPS=%s.""" %
maxloops)
activeds = spss.ActiveDataset()
if activeds == "*":
activeds = "D" + str(random.uniform(.1, 1))
spss.Submit("DATASET NAME %s" % activeds)
tmpvar1 = "V" + str(random.uniform(.1, 1))
tmpvar2 = "V" + str(random.uniform(.1, 1))
tmpvar3 = "V" + str(random.uniform(.1, 1))
tmpfile1 = "F" + str(random.uniform(.1, 1))
tmpfile2 = "F" + str(random.uniform(.1, 1))
tmpdata1 = "D" + str(random.uniform(.1, 1))
tmpdata2 = "D" + str(random.uniform(.1, 1))
tmpdata3 = "D" + str(random.uniform(.1, 1))
omstag1 = "T" + str(random.uniform(.1, 1))
omstag2 = "T" + str(random.uniform(.1, 1))
omstag3 = "T" + str(random.uniform(.1, 1))
omstag4 = "T" + str(random.uniform(.1, 1))
omstag5 = "T" + str(random.uniform(.1, 1))
omstag6 = "T" + str(random.uniform(.1, 1))
lowlabel = _("""Lower %s%% Asymptotic CI Bound""") % cilevel
upplabel = _("""Upper %s%% Asymptotic CI Bound""") % cilevel
spss.Submit(r"""
DATASET COPY %s WINDOW=HIDDEN.""" % tmpdata1)
spss.Submit(r"""
DATASET ACTIVATE %s WINDOW=ASIS.""" % tmpdata1)
filt = spssaux.GetSHOW("FILTER", olang="english")
if filt != "No case filter is in effect":
filtcond = filt.strip("(FILTER)")
select = "SELECT IF " + str(filtcond) + "."
spss.Submit("""%s""" % select)
spss.Submit("""EXECUTE.""")
spss.Submit("""USE ALL.""")
banana = spssaux.getDatasetInfo(Info="SplitFile")
if banana != "":
warntext.append(_("This command ignores split file status."))
spss.Submit(r"""SPLIT FILE OFF.""")
spss.Submit(r"""
COUNT %s=%s (MISSING).""" % (tmpvar1, varlist))
spss.Submit(r"""
SELECT IF %s=0.""" % tmpvar1)
spss.Submit(r"""
EXECUTE.
MISSING VALUES ALL ().""")
validn = spss.GetCaseCount()
if wtvar == None:
spss.Submit(r"""
SAVE OUTFILE=%s.""" % tmpfile1)
else:
spss.Submit(r"""
DO IF %s >= 1.""" % wtvar)
spss.Submit(r"""
+ LOOP #i=1 TO %s.""" % wtvar)
spss.Submit(r"""
XSAVE OUTFILE=%s
/KEEP=%s
/DROP=%s.""" % (tmpfile1, varlist, wtvar))
spss.Submit(r"""
+ END LOOP.
END IF.
EXECUTE.
""")
spss.Submit(r"""
OMS /SELECT ALL EXCEPT=WARNINGS
/IF COMMANDS=['Variables to Cases']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag1)
spss.Submit(r"""
VARSTOCASES
/MAKE %s FROM %s.""" % (tmpvar2, varlist))
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag1)
catdata = []
try:
cur = spss.Cursor(isBinary=False)
except:
cur = spss.Cursor()
while True:
datarow = cur.fetchone()
if datarow is None:
break
catdata.append(datarow[-1])
cur.close()
cats = list(set(catdata))
ncats = len(cats)
nraters = len(varnames)
neededn = max(ncats, nraters)
if validn < neededn:
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Fleiss Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag2)
spss.StartProcedure(_("Fleiss Kappa"), "Fleiss Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""There are too few complete cases. This command is not executed."""
))
spss.EndProcedure()
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag2)
elif ncats < 2:
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Fleiss Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag2)
spss.StartProcedure(_("Fleiss Kappa"), "Fleiss Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""All ratings are the same. This command is not executed."""
))
spss.EndProcedure()
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag2)
else:
if len(warntext) > 0:
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Fleiss Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag2)
if len(warntext) == 1:
spss.StartProcedure(_("Fleiss Kappa"), "Fleiss Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String("%s" %
warntext[0])
spss.EndProcedure()
if len(warntext) == 2:
spss.StartProcedure(_("Fleiss Kappa"), "Fleiss Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
"%s \n"
"%s" % (warntext[0], warntext[1]))
spss.EndProcedure()
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag2)
spss.Submit(r"""
AGGREGATE
/OUTFILE=%s
/BREAK=%s
/%s=N.""" % (tmpfile2, tmpvar2, tmpvar3))
spss.Submit(r"""
DATASET DECLARE %s WINDOW=HIDDEN.""" % tmpdata2)
spss.Submit(r"""
DATASET DECLARE %s WINDOW=HIDDEN.""" % tmpdata3)
spss.Submit(r"""
OMS /SELECT ALL EXCEPT=WARNINGS
/IF COMMANDS=['Matrix']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag3)
spss.Submit(r"""
MATRIX.
GET x
/FILE=%s
/VARIABLES=%s.
GET ratecats
/FILE=%s
/VARIABLES=%s.
COMPUTE n=NROW(x).
COMPUTE c=NROW(ratecats).
COMPUTE y=MAKE(n,c,0).
LOOP i=1 to n.
+ LOOP j=1 to NCOL(x).
+ LOOP k=1 to c.
+ DO IF x(i,j)=ratecats(k).
+ COMPUTE y(i,k)=y(i,k)+1.
+ END IF.
+ END LOOP.
+ END LOOP.
END LOOP.
COMPUTE k=NCOL(x).
COMPUTE pe=MSUM((CSUM(y)/MSUM(y))&**2).
COMPUTE pa=MSSQ(y)/(NROW(y)*k*(k-1))-(1/(k-1)).
COMPUTE kstat=(pa-pe)/(1-pe).
COMPUTE cp=(CSSQ(y)-CSUM(y))&/((k-1)&*CSUM(y)).
COMPUTE pj=CSUM(y)/MSUM(y).
COMPUTE one=MAKE(1,NCOL(pj),1).
COMPUTE qj=one-pj.
COMPUTE kj=(cp-pj)&/qj.
COMPUTE num=2*((pj*t(qj))**2-MSUM(pj&*qj&*(qj-pj))).
COMPUTE den=n*k*(k-1)*((pj*t(qj))**2).
COMPUTE ase=SQRT(num/den).
COMPUTE z=kstat/ase.
COMPUTE sig=1-CHICDF(z**2,1).
SAVE {kstat,ase,z,sig}
/OUTFILE=%s
/VARIABLES=kstat,ase,z,sig.
COMPUTE asej=MAKE(1,c,SQRT(2/(n*k*(k-1)))).
COMPUTE zj=kj&/asej.
COMPUTE sigj=one-CHICDF(zj&**2,1).
SAVE {ratecats,t(cp),t(kj),t(asej),t(zj),t(sigj)}
/OUTFILE=%s
/VARIABLES=category,cp,kstat,ase,z,sig.
END MATRIX.""" % (tmpfile1, varlist, tmpfile2, tmpvar2, tmpdata2, tmpdata3))
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag3)
spss.Submit(r"""
DATASET ACTIVATE %s WINDOW=ASIS.""" % tmpdata2)
spss.Submit(r"""
COMPUTE lower=kstat-SQRT(IDF.CHISQUARE(%s/100,1))*ase.""" % cilevel)
spss.Submit(r"""
COMPUTE upper=kstat+SQRT(IDF.CHISQUARE(%s/100,1))*ase.""" % cilevel)
spss.Submit(r"""
FORMATS kstat ase z sig lower upper (F11.3).
VARIABLE LABELS kstat %s. """ % _smartquote(_("""Kappa""")))
spss.Submit(r"""
VARIABLE LABELS ase %s. """ % _smartquote(_("""Asymptotic Standard Error""")))
spss.Submit(r"""
VARIABLE LABELS z %s. """ % _smartquote(_("""Z""")))
spss.Submit(r"""
VARIABLE LABELS sig %s. """ % _smartquote(_("""P Value""")))
spss.Submit(r"""
VARIABLE LABELS lower %s. """ % _smartquote(_(lowlabel)))
spss.Submit(r"""
VARIABLE LABELS upper %s. """ % _smartquote(_(upplabel)))
spss.Submit(r"""
EXECUTE.
""")
try:
cur = spss.Cursor(isBinary=False)
except:
cur = spss.Cursor()
data1 = cur.fetchone()
cur.close()
collabels1=[spss.GetVariableLabel(0),spss.GetVariableLabel(1),spss.GetVariableLabel(2),spss.GetVariableLabel(3), \
spss.GetVariableLabel(4),spss.GetVariableLabel(5)]
celldata1 = [
data1[0], data1[1], data1[2], data1[3], data1[4], data1[5]
]
spss.Submit(r"""
DATASET ACTIVATE %s WINDOW=ASIS.""" % tmpdata3)
spss.Submit(r"""
COMPUTE lower=kstat-SQRT(IDF.CHISQUARE(%s/100,1))*ase.""" % cilevel)
spss.Submit(r"""
COMPUTE upper=kstat+SQRT(IDF.CHISQUARE(%s/100,1))*ase.""" % cilevel)
spss.Submit(r"""
FORMATS category (F10.0) cp kstat ase z sig lower upper (F11.3).
VARIABLE LABELS category %s. """ % _smartquote(_("""Rating Category""")))
spss.Submit(r"""
VARIABLE LABELS cp %s. """ % _smartquote(_("""Conditional Probability""")))
spss.Submit(r"""
VARIABLE LABELS kstat %s. """ % _smartquote(_("""Kappa""")))
spss.Submit(r"""
VARIABLE LABELS ase %s. """ % _smartquote(_("""Asymptotic Standard Error""")))
spss.Submit(r"""
VARIABLE LABELS z %s. """ % _smartquote(_("""Z""")))
spss.Submit(r"""
VARIABLE LABELS sig %s. """ % _smartquote(_("""P Value""")))
spss.Submit(r"""
VARIABLE LABELS lower %s. """ % _smartquote(_(lowlabel)))
spss.Submit(r"""
VARIABLE LABELS upper %s. """ % _smartquote(_(upplabel)))
spss.Submit(r"""
EXECUTE.""")
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Fleiss Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag4)
spss.Submit(r"""
OMS
/SELECT TEXTS
/IF COMMANDS=['Fleiss Kappa'] LABELS=['Active Dataset']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag5)
if len(warntext) > 0:
spss.Submit(r"""
OMS
/SELECT HEADINGS
/IF COMMANDS=['Fleiss Kappa']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag6)
n = spss.GetCaseCount
rlabels = []
data2 = []
try:
cur = spss.Cursor(isBinary=False)
except:
cur = spss.Cursor()
for i in range(0, spss.GetCaseCount()):
datarow = cur.fetchone()
data2.append(datarow[1:])
rlabels.append(datarow[0])
cur.close()
def _flatten(seq):
for item in seq:
if spssaux._isseq(item):
for subitem in _flatten(item):
yield subitem
else:
yield item
data2 = [item for item in _flatten(data2)]
spss.StartProcedure(_("Fleiss Kappa"), "Fleiss Kappa")
table1 = spss.BasePivotTable(_("Overall Kappa"),
"Overall Kappa")
table1.SimplePivotTable(rowdim=_(""),
rowlabels=[CellText.String("Overall")],
coldim="",
collabels=collabels1,
cells=celldata1)
if any(item != round(item) for item in rlabels):
caption = (_(
"Non-integer rating category values are truncated for presentation."
))
else:
caption = ("")
table2 = spss.BasePivotTable(
_("Kappas for Individual Categories"),
_("Individual Category Kappa Statistics"),
caption=caption)
rowlabels = [(CellText.String("{:>9.0f}".format(rlabels[i])))
for i in range(len(rlabels))]
collabels=[spss.GetVariableLabel(1),spss.GetVariableLabel(2),spss.GetVariableLabel(3), \
spss.GetVariableLabel(4),spss.GetVariableLabel(5),spss.GetVariableLabel(6), \
spss.GetVariableLabel(7)]
table2.SimplePivotTable(rowdim=_(" Rating Category"),
rowlabels=rowlabels,
coldim="",
collabels=collabels,
cells=data2)
spss.EndProcedure()
if len(warntext) > 0:
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag6)
finally:
try:
spss.Submit("""
DATASET CLOSE %s.""" % tmpdata1)
spss.Submit(r"""
DATASET ACTIVATE %s WINDOW=ASIS.""" % activeds)
if validn >= neededn:
if ncats >= 2:
spss.Submit("""
OMSEND TAG=['"%s"' '"%s"'].""" % (omstag4, omstag5))
spss.Submit("""
DATASET CLOSE %s.""" % tmpdata2)
spss.Submit("""
DATASET CLOSE %s.""" % tmpdata3)
spss.Submit("""
ERASE FILE=%s.""" % tmpfile1)
spss.Submit(r"""
ERASE FILE=%s.""" % tmpfile2)
except:
pass
spss.Submit("""
RESTORE.
""")
def dopmml(files):
"""Display pmml files"""
# debugging
#makes debug apply only to the current thread
# The global namespace qualifier depends on the PMML and is set per file
global ns
#try:
#import wingdbstub
#if wingdbstub.debugger != None:
#import time
#wingdbstub.debugger.StopDebug()
#time.sleep(2)
#wingdbstub.debugger.StartDebug()
#import thread
#wingdbstub.debugger.SetDebugThreads({thread.get_ident(): 1}, default_policy=0)
## for V19 use
#SpssClient._heartBeat(False)
#except:
#pass
fh = FileHandles()
files = fh.resolve(
files) # note that files might be a wildcard file expression
spss.StartProcedure("STATS PMML DISPLAY", "STATSPMMLINFO")
didone = False
try:
for fnumber, f in enumerate(glob.glob(files)):
text = Text("""STATS_PMML_DISPLAY-""" + str(fnumber))
didone = True
try:
try:
tree = ElementTree.parse(f)
root = tree.getroot()
first = True
except:
raise ValueError(
_("""Invalid xml. Error: %s File: %s""") %
(sys.exc_info()[1].message, f))
# Discriminant xml is not tagged as pmml
if not (root.tag.endswith("PMML")
or root.tag.endswith("spss-ml")):
text.addtext("""%f is not a PMML file""" % f)
continue
ftime = time.ctime(os.stat(f).st_mtime)
text.title = _("""Summary Information: %s""") % f
text.addtext(
_("""Summary Model Information for File: %s\nmodified: %s\n"""
) % (f, ftime))
for ch in root.getchildren():
if first:
first = False
# find the namespace - there ought to be a better way :-)
thetag = re.match(r"\{.*\}", ch.tag)
if thetag:
ns = thetag.group()
else:
ns = ""
chtag = re.sub(r"""\{.*\}""", "", ch.tag)
if chtag == "Header":
doHeader(chtag, ch)
elif chtag == "TransformationDictionary":
displayTransformations(ch, text)
elif chtag == "MiningModel":
displayMining(ch, text)
elif chtag in [
'DataDictionary', 'Extension', 'dictionary'
]:
continue
elif chtag == "model" and ch.find(
"./" + ns + 'discriminant-model') is not None:
disc = ch.find("./" + ns + 'discriminant-model')
prints['discriminant-model'](disc, text)
elif chtag not in prints:
text.addtext(
_("""Cannot display element: %s""") % chtag)
else:
prints[chtag](ch, text)
finally:
text.printit()
del (text) # should not be necessary
finally:
spss.EndProcedure()
if not didone:
raise ValueError(
_("""No files were found to process. File Specification: %s""") %
files)
def rake(info, variables, marginals,finalweight, visible=False, showweights=True,
poptotal=None, delta=0, iter=20, conv=.0001,checkempty=True,
yvar=None, xvar=None, paneldownvar=None, panelacrossvar=None, autoheatmap=None, histogram=True):
"""Calculate a weight variable such that for each controlled dimension, the (weighted) count in each category matches a specified total or fraction.
e variables is a list of the variables for which control totals or proportions are provided. It can be a sequence or
a white-space separated string
marginals is a list of dictionaries where the key is the value of a variable and the value is the target control total or fraction.
fractional marginals should normally add to 1 and counts should total the same in each dimension, but this is not enforced.
If there are no cases for a given value, the total or fraction will be less than expected.
Negative and zero marginals are tolerated but are generally inappropriate.
If a control value is not in the appropriate marginals dictionary or is zero, the resulting weight will be SYSMIS.
finalweight is a string naming the new weight variable. The variable must not already exist in the active dataset.
visible indicates whether or not the procedure output and auxiliary dataset are displayed.
poptotal is a total to which the final weights will be scaled. By default, they are scaled to sum to the existing weight total, if any, or the number of cases.
delta, iter, and conv are iteration parameters corresponding to the GENLOG parameters and can be used if there are convergence problems.
checkempty adjusts for empty cells. If it is known that there are none, this adjustment, which can use considerable memory, can be bypassed.
If the active dataset does not have a name, one is assigned automatically."""
variables = _buildvarlist(variables)
if len(variables) == 1: # delta should always be 0 if this is a 1-d problem
delta = 0.
wtvar = spss.GetWeightVar()
if wtvar: #aggrweight will be the GENLOG CSTRUCTURE variable
aggrweight = "W_" + rname()
else:
aggrweight = "N_" + rname()
activeds = spssaux.GetActiveDatasetName()
if not activeds:
activeds = "D_" + rname()
spss.Submit("DATASET NAME " + activeds)
nbreakvars = len(variables)
if nbreakvars != len(marginals):
raise ValueError(_("The number of control variables does not match number of sets of control totals"))
# aggregate the data according to the list of control variables
aggrdsname = "D_" + rname()
countname = "N_" + rname()
spss.Submit("WEIGHT OFF.")
# aggregate to a new dataset and activate it. Sum original weight variable if any
cmd=\
"""DATASET DECLARE %(aggrdsname)s %(vis)s.
AGGREGATE
/OUTFILE= %(aggrdsname)s
/BREAK=%(breakvars)s
%(wtspec)s
/%(countname)s=N.
dataset activate %(aggrdsname)s.""" % \
{'aggrdsname': aggrdsname,
'breakvars' : " ".join(variables),
'wtspec' : ("/" + aggrweight + (wtvar and ("= SUM(" + wtvar + ")") or "= N")),
'countname' : countname,
'vis' : visible and " " or " WINDOW=HIDDEN"
}
spss.Submit(cmd)
# if weighting, make a dictionary of the mean weight in each cell for use in final adjustment pass
if wtvar:
allaggrdata = spssdata.Spssdata(variables + [aggrweight] + [countname]).fetchall()
meaninputwts = {}
for row in allaggrdata:
try:
themean = row[nbreakvars] / row[nbreakvars+1]
except:
themean = None
meaninputwts[row[:nbreakvars]] = themean
# get a cursor and add control totals to this dataset
# The control total is the product of all the variable value control totals or proportions
# Track the cases to see if there are any empty cells, i.e., some combination of the marginals across
# all the control variables that does not occur in the aggregated dataset.
spss.StartProcedure("SPSSINC RAKE - I")
if checkempty:
cellset = _setprod(_dictlisttotupledsets(marginals)) # build set of tuples of all sets
curs = spssdata.Spssdata(accessType='r', indexes=variables, names=False)
for case in curs:
cellset.discard(tuple(case[:nbreakvars])) # remove found cells
curs.CClose()
spss.EndProcedure()
#add cases for any empty cells
if cellset:
spss.StartProcedure("SPSSINC RAKE - I")
curs = spssdata.Spssdata(accessType='a', names=False)
for s in cellset:
for cv in range(nbreakvars):
curs.appendvalue(cv, s[cv])
for cv in range(2): #was nbreakvars
curs.appendvalue(nbreakvars+cv, 1e-12) # 1e-8
curs.CommitCase()
curs.CClose()
spss.EndProcedure()
# compute expected count from marginals
try:
curs = spssdata.Spssdata(accessType='w')
ctrlwt = "W_"+ rname() # this will be the SPSS case weight
curs.append(ctrlwt)
curs.commitdict()
novalues = []
for case in curs:
w = 1.
for i, v in enumerate(marginals):
w = w* v.get(case[i], 0)
if w == 0:
spec = (variables[i], case[i])
if not spec in novalues:
uspec = spec[0]
if not isinstance(uspec, str):
uspec = str(uspec, locale.getlocale()[1])
info.addrow(_("Variable: %s, value: %s. No control value supplied: weight will be SYSMIS.") % (uspec, spec[1]))
novalues.append(spec)
break
curs.casevalues([w])
finally:
curs.CClose()
spss.EndProcedure()
info.generate()
# run GENLOG
expectedname = 'expected_' + rname()
newwt = "W_"+ rname()
#dbg
###spss.Submit("""save outfile="c:/temp/genloginput.sav".""")
if not visible:
omstag = "O_" + rname()
spss.Submit("OMS /SELECT ALL EXCEPT =WARNINGS /DESTINATION VIEWER=NO /TAG=" + omstag)
#debugcmd = """WEIGHT BY %(ctrlwt)s.
#GENLOG
#%(breakvars)s /CSTRUCTURE = %(aggrweight)s
#/MODEL = POISSON
#/PRINT = FREQ ESTIM
#/plot none
#/CRITERIA = CIN(95) ITERATE(%(iter)s) CONVERGE(%(conv)s) DELTA(%(delta)s)
#/save= pred(%(expectedname)s)
#/DESIGN %(breakvars)s . """ %\
#{'ctrlwt' : ctrlwt,
#'breakvars': " ".join(variables),
#'aggrweight' : aggrweight ,
#'expectedname' : expectedname,
#'iter' : iter,
#'conv' : conv,
#'delta' : delta
#}
try:
spss.Submit("""WEIGHT BY %(ctrlwt)s.
GENLOG
%(breakvars)s /CSTRUCTURE = %(aggrweight)s
/MODEL = POISSON
/PRINT = FREQ ESTIM
/plot none
/CRITERIA = CIN(95) ITERATE(%(iter)s) CONVERGE(%(conv)s) DELTA(%(delta)s)
/save= pred(%(expectedname)s)
/DESIGN %(breakvars)s . """ %\
{'ctrlwt' : ctrlwt,
'breakvars': " ".join(variables),
'aggrweight' : aggrweight ,
'expectedname' : expectedname,
'iter' : iter,
'conv' : conv,
'delta' : delta
})
except:
spss.Submit("DATASET ACTIVATE " + activeds)
raise ValueError(_("""Failure in GENLOG procedure. Processing stopped.
The error could be either a failure to compute the result
or not having a license for the Advanced Statistics option"""))
finally:
if not visible:
spss.Submit("OMSEND TAG=" + omstag)
# get the expected counts, normalized by the cell N in order to distribute
expkts = {}
spss.StartProcedure("SPSSINC RAKE - II")
try:
curs = spssdata.Spssdata(accessType='r', indexes = variables + [expectedname] + [aggrweight]+ [countname])
weightsum = 0.
wsum = 0.
for case in curs:
weightsum += case[nbreakvars] or 0 # allow for missing values j.i.c
if case[-3]:
wsum += case[-2]
w = case[-3]/case[-1]
else:
w = None
expkts[tuple(case[:nbreakvars])] = w
finally:
curs.CClose()
spss.EndProcedure()
# normalize weights to user total or sum of sample weights
poptotal = poptotal or wsum
for key in expkts:
if expkts[key]:
expkts[key] *= poptotal/weightsum
# return to the first dataset and apply weights.
spss.Submit("DATASET ACTIVATE " + activeds)
if not visible:
spss.Submit("DATASET CLOSE " + aggrdsname)
spss.StartProcedure("SPSSINC RAKE - III")
if wtvar:
indexes = variables+ [wtvar]
else:
indexes = variables
curs = spssdata.Spssdata(accessType='w', indexes = indexes)
try:
failed = False
curs.append(spssdata.vdef(finalweight, vlabel=_("Raked Weight")))
curs.commitdict()
wirisum = 0.
wirisumNewwt = 0.
wiri2sum = 0.
wiri2sumNewwt = 0.
wisum = 0.
actuals = {} # dictionary for weights and counts actually used
actualsNewwt = {}
for case in curs:
index = case[:nbreakvars]
rwt = expkts.get(index, None) # raked weight
if wtvar:
wt = case[-1]
else:
wt = 1.
# for weighted data, adjust cell weights by input case weight normalized by cell mean weight
try:
if wtvar:
newwt = rwt * wt / meaninputwts[index]
else:
newwt = rwt
except:
newwt = None
curs.casevalues([newwt])
#curs.casevalues([rwt])
try:
kt = actuals.get(index, (0,0))[1] + wt
actuals[index] = [rwt, kt]
except:
pass
if wtvar:
try:
cumwt, cumkt = actualsNewwt.get(index, [0,0])
cumwt += newwt
cumkt += wt
#cumkt += 1
actualsNewwt[index] = [cumwt, cumkt]
except:
pass
if not rwt is None:
wisum += wt
wirisum += wt * newwt
wiri2sum += wt * newwt * newwt
except:
curs.CClose()
curs = None
spss.EndProcedure()
raise
finally:
if not curs is None:
curs.CClose()
denom = wisum * wiri2sum
if denom != 0:
sampleeff = 100. * wirisum * wirisum / denom
else:
sampleeff = None
cells = [sampleeff]
rowlabels = [_("""Sample Balance""")]
if wtvar:
#denom = wisum * wiri2sumNewwt
#if denom != 0:
#sampleeffNewwt = 100. * wirisumNewwt * wirisumNewwt / denom
#else:
#sampleeffNewwt = None
#cells.append(sampleeffNewwt)
rowlabels = [_("""Sample Balance Including Final Weight Adjustment""")]
tbl = spss.BasePivotTable(_("""Sample Balance Based on Variables: %s""") % ", ".join(variables),
"RAKEBALANCE")
tbl.SimplePivotTable(rowlabels=rowlabels, collabels=[_("""Balance""")], cells=cells)
# table of weights
if showweights:
collabels = [_("""Category Rake Weight""")]
if wtvar:
for k in actuals:
cumwt, cumkt = actualsNewwt[k]
act = actuals[k]
act.append(cumwt / cumkt)
actuals[k] = act
collabels.append(_("Case Count Weighted by Input Weight"))
collabels.append(_("Mean Adjusted Raked Weight"))
else:
collabels.append(_("Unweighted Case Count"))
#items = sorted(expkts.items())
items = sorted(actuals.items())
rowlabels = [", ".join([str(v) for v in item[0]]) for item in items]
cells = [item[-1] for item in items]
tbl2 = spss.BasePivotTable(_("""Raked Weights"""), "RAKEDWEIGHTS")
tbl2.SimplePivotTable(rowdim=", ".join(variables), rowlabels = rowlabels,
collabels=collabels, cells=cells)
spss.EndProcedure()
if not failed:
if histogram:
dohistogram(finalweight)
doheatmap(variables, yvar, xvar, paneldownvar, panelacrossvar, finalweight, autoheatmap)
spss.Submit("WEIGHT BY " + finalweight)
def catvalues(items=None, prefix=None, names=None, specialvars=None,
specialsorder="after", order="d", missing="exclude",
categorylabels="varlabels", specialvalues=None, other=False,
variables=None, countvalues=None, macroname=None,
mincount=0, minpercent=0, maxcount = None, maxpercent=None, separator=" ",
customattr=False, attrname="ORDER"):
"""Construct macros, custom attributes and MR set definitions"""
##debugging
# makes debug apply only to the current thread
#try:
#import wingdbstub
#if wingdbstub.debugger != None:
#import time
#wingdbstub.debugger.StopDebug()
#time.sleep(1)
#wingdbstub.debugger.StartDebug()
#import thread
#wingdbstub.debugger.SetDebugThreads({thread.get_ident(): 1}, default_policy=0)
# for V19 use
## ###SpssClient._heartBeat(False)
#except:
#pass
weightvar = spss.GetWeightVar()
# tODO: allow both names and macroname to be None if customattr
if variables is None:
if sum([prefix is None, names is None]) != 1 and not customattr:
raise ValueError(_("""Either a prefix or a set of macro or MR set names or a custom attribute must be specified"""))
if names is not None:
names = spssaux._buildvarlist(names)
if len(names) != len(items):
raise ValueError(_("""The number of macro or MR set names specified is different from the number of items"""))
if prefix is not None:
names = [prefix + "_" + name for name in items] # MR sets will have a $ in the macro name
resolver = Resolver()
allvars, vartypes = resolver.resolve(items) # check existence and get all variables and variable types
alltypes = vartypes.values()
# variables, direct or in MR sets, must be either all strings or all numeric
numerics = any([vartypes[item] == 0 for item in allvars])
strings = any([vartypes[item] != 0 for item in allvars])
if numerics and strings:
raise ValueError(_("""Items to process must be either all numeric or all string"""))
try:
if numerics and specialvalues is not None:
specialvalues = [float(item) for item in specialvalues]
except:
raise ValueError(_("""A nonnumeric special value was given for numeric variables"""))
regularvars = [item for item in items if not item.startswith("$")]
mrsets = [item for item in items if item.startswith("$")]
resolver.close()
if regularvars and other:
valuelabelsdict, missingvaluesdict = getmetadata(regularvars, missing)
else:
valuelabelsdict = None
missingvaluesdict = None
macrosgenerated = []
if regularvars:
macrosgenerated, customattrsgenerated = genVarsCategoryList(regularvars,
specialvalues=specialvalues, macroname=names,
missing=missing, order=order, weightvar=weightvar, specialsorder=specialsorder,
valuelabelsdict=valuelabelsdict, missingvaluesdict=missingvaluesdict,
customattr=customattr, attrname=attrname)
spss.StartProcedure("STATS CATEGORY ORDER", "STATSCATEGORYMACRO")
if customattrsgenerated:
caption = _("Custom attribute name: %s") % attrname
else:
caption = ""
table = spss.BasePivotTable("Generated Macros or Custom Attributes",
"STATSCATMACROS", caption=caption)
if macrosgenerated:
gen = macrosgenerated
else:
gen = customattrsgenerated
table.SimplePivotTable(rowdim=_("Name"), rowlabels=[names for names, values in gen],
collabels = [_("Definition")], cells=[values for names, values in gen])
spss.EndProcedure() # can't issue MRSETS command in a procedure state
# specialvars do not have to have been in the original set.
if mrsets:
mrsetsgenerated = genSetsCategoryList(mrsets, allvars, vartypes,
resolver, specialvalues=specialvars, macroname=names,
missing=missing, order=order, weightvar=weightvar,
categorylabels=categorylabels, specialsorder=specialsorder, other=other)
# mrset generation displayed via MRSET creation command so not repeated here
for m in macrosgenerated:
###spss.SetMacroValue(m[0], m[1]) # This api mishandles Unicode characters so use syntax instead
spss.Submit("""DEFINE %s() %s""" %(m[0], m[1]))
if variables is not None:
if items is not None or names is not None:
raise ValueError(_("""ITEMS and NAMES cannot be used with COUNTORDERING specifications."""))
if countvalues is None or macroname is None:
raise ValueError(_("""The values to count and a macroname must be specified when using COUNTORDERING specifications."""))
genVarMacro(variables, countvalues, order, macroname, mincount,
minpercent, maxcount, maxpercent, separator, weightvar, missing)
def dopropor(num=None,
denom=None,
id=None,
dsname="*",
alpha=.05,
adjust='bonferroni'):
if num is None or denom is None:
raise ValueError("Error: NUM and DENOM keywords are required")
if spss.PyInvokeSpss.IsUTF8mode():
unistr = str
else:
unistr = str
currentds = spss.ActiveDataset()
if currentds == "*":
currentds = "S" + str(random.uniform(0, 1))
spss.Submit("DATASET NAME %s" % currentds)
dsnamed = True
else:
dsnamed = False
numvec, denomvec, idvec = getvalues(num, denom, id, dsname)
# clean data, discard missing
droplist = []
for i in range(len(numvec)):
droplist.append(numvec[i] is not None
and denomvec[i] is not None) #missing data
if (droplist[i] and (numvec[i] > denomvec[i] or denomvec[i] <= 0)):
raise ValueError(
"Error: NUM value greater than DENOM value or zero denominator: %s, %s"
% (numvec[i], denomvec[i]))
for lis in numvec, denomvec, idvec:
lis = [x for f, x in zip(droplist, lis) if f] #prune missing values
if len(numvec) == 0:
raise ValueError("Error: No valid proportions were found to analyze")
alphalow = alpha / 2
alphahigh = 1 - alphalow
dotest = len(numvec) > 1
try:
spss.StartDataStep() #TODO: pending transformations
except:
spss.Submit("EXECUTE")
spss.StartDataStep()
# calculate ci's via SPSS IDFs
ds = spss.Dataset(name=None)
spss.SetActive(ds)
ds.varlist.append("p", 0)
ds.varlist.append("num", 0)
ds.varlist.append("denom", 0)
p0 = numvec[0] / denomvec[0]
sdvec = []
for i in range(len(numvec)):
p1 = numvec[i] / denomvec[i]
sdvec.append(
sqrt(p0 * (1 - p0) / denomvec[0] + p1 * (1 - p1) / denomvec[i]))
#p = (numvec[i] + numvec[0]) / (denomvec[i] + denomvec[0])
#z = (p1 - p0)/sqrt(p * (1 - p)*(1/denomvec[0] + 1/denomvec[i]))
ds.cases.append([p1, numvec[i], denomvec[i]])
spss.EndDataStep()
cmd =r"""COMPUTE PLOWBI = IDF.BETA(%(alphalow)s, num + .5, denom-num + .5).
COMPUTE PHIGHBI = IDF.BETA(%(alphahigh)s, num + .5, denom - num + .5).
DO IF num > 0.
COMPUTE PLOWPOIS = (IDF.CHISQ(%(alphalow)s, 2*num)/2)/denom.
ELSE.
COMPUTE PLOWPOIS = 0.
END IF.
COMPUTE PHIGHPOIS = (IDF.CHISQ(%(alphahigh)s, 2*(num+1))/2) / denom.
COMPUTE ZTAIL = IDF.NORMAL(%(alphahigh)s, 0,1).
EXECUTE."""\
% {"alphalow": alphalow, "alphahigh": alphahigh}
spss.Submit(cmd)
plowbi = []
phighbi = []
plowpois = []
phighpois = []
spss.StartDataStep()
ds = spss.Dataset(name="*")
for case in ds.cases:
i = 3
for v in plowbi, phighbi, plowpois, phighpois:
v.append(case[i])
i += 1
zalpha2 = case[-1]
try:
closeafter = False
spss.SetActive(spss.Dataset(name=currentds))
except:
closeafter = True
ds.close()
spss.EndDataStep()
from spss import CellText
spss.StartProcedure("Proportions")
table = spss.BasePivotTable("Proportion Confidence Intervals",
"Proportions")
titlefootnote = "Alpha = %.3f" % alpha
if 0. in numvec:
titlefootnote += " (One-sided %.3f when p = 0)" % (alpha / 2.)
table.TitleFootnotes(titlefootnote)
rowdim = table.Append(spss.Dimension.Place.row, "Proportions")
coldim = table.Append(spss.Dimension.Place.column, "Statistics")
cols = [
"p", "Binomial\nLower CI", "Binomial\nUpper CI", "Poisson\nLower CI",
"Poisson\nUpper CI", "Difference\nfrom p0",
"Difference from p0\nLower CI", "Difference from p0\nUpper CI"
]
table.SetCategories(coldim, [CellText.String(v) for v in cols])
idvec = [
not v is None and unistr(v) or unistr(i + 1)
for i, v in enumerate(idvec)
]
table.SetCategories(rowdim, [CellText.String(v) for v in idvec])
for i in range(len(numvec)):
p1 = numvec[i] / denomvec[i]
if i > 0:
zdifflow = p1 - p0 - sdvec[i] * zalpha2
zdiffhigh = p1 - p0 + sdvec[i] * zalpha2
else:
zdifflow = zdiffhigh = 0.
table.SetCellsByRow(CellText.String(idvec[i]), [
CellText.Number(v)
for v in (numvec[i] / denomvec[i], plowbi[i], phighbi[i],
plowpois[i], phighpois[i], p1 - p0, zdifflow, zdiffhigh)
])
if i == 0:
table[(CellText.String(idvec[0]),
CellText.String(cols[-3]))] = CellText.String("-")
table[(CellText.String(idvec[0]),
CellText.String(cols[-2]))] = CellText.String("-")
table[(CellText.String(idvec[0]),
CellText.String(cols[-1]))] = CellText.String("-")
spss.EndProcedure()
if closeafter:
spss.Submit(r"""NEW FILE.
DATASET NAME %s.""" % "S" + str(random.uniform(0, 1)))
value.remove (value[4])
stat_labels = ['B [95% CI]', 'SE B', 'Beta', 'p']
#renaming the statistics labels in order
# footnote for our table - uses values from Model Summary and ANOVA tables
footnote = "Note: R Square = {}, adjusted R Square = {}, overall F (df {}, {}) = {}, p = {}.".format(\
modelsuminfo['R Square'], modelsuminfo['Adjusted R Square'], anovavals['Regression'][0][1], anovavals ['Residual'][0][1], \
anovavals ['Regression'][1][1], anovavals ['Regression'][2][1])
##finally, create new pivot table with the values we've pulled above
import spss
mytablecells = []
for var in varnames: # because the variable names are sorted as we want them in this list
for i in range (len(coeffdict[var])):
statdata = coeffdict[var][i][1]
mytablecells.append (statdata)
spss.StartProcedure("myTABLE")
mytitle = 'Linear Model of Predictors of the Mental Health Score. 95% CI and standard errors based on parametric model.'
table = spss.BasePivotTable (mytitle, "OMS table subtype")
table.SimplePivotTable(rowdim = "Variables in Model",
rowlabels = varnames,
coldim = "Statistics",
collabels = stat_labels,
cells = mytablecells)
table.TitleFootnotes (footnote) # adds footnote to the title of the table
spss.EndProcedure() #then the table will print to the output
def plots(yvars,
xvars,
color=None,
size=None,
shape=None,
label=None,
linear=False,
quadratic=False,
cubic=False,
loess=False,
ignore=False,
title="",
categorical="bars",
group=1,
boxplots=False,
hexbin=False,
applyfitto="total",
indent=15,
yscale=75,
pagex=None,
pagey=None):
"""Create plots per specifcation described in help above"""
# debugging
# makes debug apply only to the current thread
#try:
#import wingdbstub
#if wingdbstub.debugger != None:
#import time
#wingdbstub.debugger.StopDebug()
#time.sleep(2)
#wingdbstub.debugger.StartDebug()
#import thread
#wingdbstub.debugger.SetDebugThreads({thread.get_ident(): 1}, default_policy=0)
## for V19 use
## ###SpssClient._heartBeat(False)
#except:
#pass
npage = [pagex, pagey].count(None) # 0 means both specified
if npage == 1:
raise ValueError(
_("Page specification must include both x and y sizes"))
if group > 1:
boxplots = False
spssweight = spss.GetWeightVar()
if not spssweight:
spssweight = None
vardict = spssaux.VariableDict()
# display pivot table of legend information
fits = []
for i, fittype in enumerate([linear, quadratic, cubic, loess]):
if fittype:
fits.append(fittypetable[i])
spss.StartProcedure("STATS REGRESS", _("Relationship Plots"))
ttitle = _("Chart Legend Information")
if title:
ttitle = ttitle + "\n" + title
tbl = spss.BasePivotTable(
ttitle,
"CHARTLEGENDINFO",
caption=
_("Legend Settings for the charts that follow. Some settings do not apply to categorical charts."
))
tbl.SimplePivotTable(_("Settings"),
rowlabels=[
_("Color by"),
_("Size by"),
_("Shape by"),
_("Label by"),
_("Fit Lines")
],
collabels=[_("Value")],
cells=[
labelit(color, vardict) or "---",
labelit(size, vardict) or "---",
labelit(shape, vardict) or "---",
labelit(label, vardict) or "---", "\n".join(fits)
or "---"
])
spss.EndProcedure()
# group fitlines only available for categorically defined groups
if not color or (color and vardict[color].VariableLevel == "scale"):
applyfitto = "total"
aesthetics = set([
item for item in [color, size, shape, label, spssweight]
if not item is None
])
for y in yvars:
yobj = vardict[y]
if yobj.VariableLevel != "scale":
raise ValueError(
_("Y variables must have a scale measurement level: %s") % y)
yvarlabel = yobj.VariableLabel or y
# construct one possibly multi-part chart for each numcharts variables
for xpart in xgen(xvars, group):
first = True
cmd = []
numcharts = len(xpart)
mostvariables = " ".join(
set(xpart +
list(aesthetics))) # eliminate duplicates (except with y)
if spssweight:
options = ", weight(%s)" % spssweight
else:
options = ""
cmd.append(ggraphtemplate % {
"allvars": y + " " + mostvariables,
"options": options
})
indentx = indent
if npage == 0: # page specs were given
if numcharts < group: # short row
shortpagex = pagex * indent / 100. + pagex * (
100. - indent) / 100. * (float(numcharts) / group)
indentx = indent * (pagex / shortpagex)
cmd.append(pagestarttemplate % {
"pagex": shortpagex,
"pagey": pagey
})
else:
cmd.append(pagestarttemplate % {
"pagex": pagex,
"pagey": pagey
})
cmd.append(datatemplate % {"varname": y, "unitcategory": ""})
alldatastatements = set([y.lower()])
if spssweight:
cmd.append(gendata(spssweight, vardict, alldatastatements))
# loop over one or more x variables for this chart
for currentn, x in enumerate(xpart):
xobj = vardict[x]
ml = xobj.VariableLevel
if numcharts > 1:
cmd.append(
graphstarttemplate % {
"originandscale":
scaling(numcharts, currentn, indentx, yscale)
})
if boxplots and ml == "scale":
cmd.append(
graphstarttemplate %
{"originandscale": "origin(15%, 10%), scale(75%,75%)"})
if ml == "scale": # build scatterplot specs
uc = ""
options = ""
if size:
options = options + ", size(%s)" % size
cmd.append(gendata(size, vardict, alldatastatements))
if numcharts > 1:
cmd.append(aesth % {"atype": "size"})
if color:
options = options + ", color.exterior(%s)" % color
cmd.append(gendata(color, vardict, alldatastatements))
if numcharts > 1:
cmd.append(aesth % {"atype": "color.exterior"})
if shape:
if vardict[shape].VariableLevel == "scale":
raise ValueError(
_("The shape variable must be categorical: %s")
% shape)
options = options + ", shape(%s)" % shape
cmd.append(gendata(shape, vardict, alldatastatements))
if numcharts > 1:
cmd.append(aesth % {"atype": "shape"})
else:
uc = iscat
if categorical == "bars":
cmd.append(include0)
if not first:
other = ", null()"
else:
other = ""
if title and numcharts == 1 and not boxplots:
cmd.append(titletemplate % {"title": title})
cmd.append(gendata(x, vardict, alldatastatements))
if label:
cmd.append(gendata(label, vardict, alldatastatements))
#cmd.append(datatemplate % {"varname": x, "unitcategory": uc})
cmd.append(guidetemplate % {
"dim": 1,
"varlabel": xobj.VariableLabel or x,
"other": ""
})
if first:
cmd.append(guidetemplate % {
"dim": 2,
"varlabel": yvarlabel,
"other": other
})
else:
cmd.append(noyaxis)
if ml == "scale":
if label:
options = options + ", label(%s))" % label
if hexbin:
cmd.append(hexbinscatterelement % {
"y": y,
"x": x,
"options": options
})
else:
cmd.append(scatterelement % {
"y": y,
"x": x,
"options": options
})
for i, fittype in enumerate(
[linear, quadratic, cubic, loess]):
if fittype:
if applyfitto == "group":
colorspec = ", color(%s)" % color
else:
colorspec = ""
if numcharts > 1:
cmd.append(aesth % {"atype": "color"})
cmd.append(fitlineelement % \
{"fittype": fittypekwd[i], "y": y, "x": x, "lineshape" : lineshapes[i], "color" : colorspec})
if boxplots: # bordered boxplot if single variable chart
cmd.append(graphendtemplate)
cmd.append(graphstarttemplate % {
"originandscale":
"origin(15%, 0%), scale(75%,8%)"
})
cmd.append("""GUIDE: axis(dim(1), ticks(null()))""")
cmd.append("""COORD: rect(dim(1))""")
cmd.append(oneboxplotelement % {"variable": x})
cmd.append(graphendtemplate)
cmd.append(graphstarttemplate % {
"originandscale":
"origin(92%, 10%), scale(8%, 75%)"
})
cmd.append("COORD: transpose(rect(dim(1)))")
cmd.append("""GUIDE: axis(dim(1), ticks(null()))""")
cmd.append(oneboxplotelement % {"variable": y})
cmd.append(graphendtemplate)
else:
if categorical != "boxplot":
cmd.append(categoricalelement % {
"etype": elementmap[categorical],
"y": y,
"x": x
})
else:
if label:
options = ", label(%s)" % label
else:
options = ""
cmd.append(boxplotelement % {
"y": y,
"x": x,
"options": options
})
first = False
if numcharts > 1:
cmd.append(graphendtemplate)
if npage == 0:
cmd.append(pageendtemplate)
cmd.append(endgpl)
spss.Submit(cmd)
def convert(filespec,
outputloc,
recursive=False,
copyall=False,
overwrite=False):
"""Convert SPSS syntax files or Python files from Python 2 to Python 3
filespec is a file name or a wildcard specification.
path separator must be written with /, not \ ??
outputloc is the location where converted files are written.
copyall specifies whether to copy unchanged files or not
overwrite specifies whether to overwrite existing files or not
The directory and any intermediate directories will be created if necessary.
The recursive parameter is not yet supported.
"""
tempdir = tempfile.TemporaryDirectory()
temp1 = tempdir.name
# support file handles
fh = spssaux.FileHandles()
outputloc = fh.resolve(outputloc)
filespec = fh.resolve(filespec)
if not os.path.isdir(outputloc):
os.mkdir(outputloc)
###os.makedirs(outputloc, exist_ok=True) # raises exception if not possible
# the lib2to3 conversion code and the SPSS print output functions interfere
# with each other, so we divert the logging information to a file
convlog = outputloc + os.sep + "conversionMessages.txt"
logging.basicConfig(filename=convlog, filemode="w", level=logging.INFO)
filespec = os.path.abspath(filespec)
#if recursive:
#makeoutputlocs(filespec, outputloc)
# py success failure skipped
# sps success failure skipped
counts = [0, 0, 0, 0, 0, 0]
for f in glob.glob(filespec, recursive=recursive):
ext = os.path.splitext(f)[1].lower()
if not ext in [".py", ".sps"]:
continue
thefile = os.path.basename(f)
foutputloc = getoutputloc(f, outputloc)
# ensure that file will not be overwritten or allow
if not overwrite:
try:
if os.path.exists(foutputloc + os.path.sep + thefile):
print(
"*** %s already exists in output location. Skipping conversion"
% fq(thefile))
if ext == ".py":
counts[2] += 1
else:
counts[5] += 1
continue
except:
pass
# convert file according to type, accumulating statistics
if ext == ".sps":
if cvtsyntax(f, temp1, foutputloc, copyall):
counts[3] += 1
else:
counts[4] += 1
else:
if cvtpy(f, foutputloc, copyall):
counts[0] += 1
else:
counts[1] += 1
print(
"\nAdditional conversion information (usually of limited usefulness) written to file: {0}"
.format(convlog))
pt = spss.StartProcedure("Convert Python 2")
spss.AddProcedureFootnotes(
"Successful conversions should be checked for correctness")
spss.AddProcedureFootnotes(
"Existing files overwrite option: {0}".format(overwrite))
spss.AddProcedureFootnotes(
"Copy unchanged files option: {0}".format(copyall))
pt = spss.BasePivotTable("File Conversions: {0}".format(filespec),
"PythonConversionStats")
pt.SimplePivotTable(rowlabels=['py', 'sps'],
collabels=['Success', 'Failure', "Skipped"],
cells=counts)
spss.EndProcedure()
logging.shutdown()
def createText(strings=[""],
otitle="Comment",
heading="Comment",
pagebreak=False,
wrap=None):
"""Create a textblock in the Viewer with contents strings.
strings is a sequence of lines of text to insert in the block. If omitted,
the block will be empty.
otitle is an optional title to appear in the outline.
heading is the procedure name that will appear first in the outline and the associated item on the right.
If pagebreak is True and this is version 17 or later, a pagebreak is inserted.
If the text appears to be html or rtf, it is inserted with formatting (using a scripting api) if 17 or later
"""
# debugging
# makes debug apply only to the current thread
#try:
#import wingdbstub
#if wingdbstub.debugger != None:
#import time
#wingdbstub.debugger.StopDebug()
#time.sleep(1)
#wingdbstub.debugger.StartDebug()
#import thread
#wingdbstub.debugger.SetDebugThreads({thread.get_ident(): 1}, default_policy=0)
## for V19 use
### ###SpssClient._heartBeat(False)
#except:
#pass
try:
spss.StartProcedure(heading)
except:
raise ValueError(
_("Error: There must be a nonempty active dataset before using this command."
))
user = getpass.getuser()
curdate = time.asctime()
for i in range(len(strings)):
strings[i] = strings[i].replace(")USER", user)
strings[i] = strings[i].replace(")CURDATE", curdate)
start = strings[0][:7].lower()
# The rtf code below screams for an r prefix, but the start text is coming through non-raw
if not (start.startswith("<html>") or start.startswith("{\rtf")
) or spss.GetDefaultPlugInVersion()[4:] < "170":
strings, nitems = reducer(strings)
if not wrap is None:
strings = "\n".join(
textwrap.wrap(strings, width=wrap, break_long_words=False))
t = spss.TextBlock(otitle, strings)
# lines are appended at once for better performance
spss.EndProcedure()
else:
spss.TextBlock(otitle, "")
spss.EndProcedure()
# do the rest with scripting apis
import SpssClient
SpssClient.StartClient()
time.sleep(
.1) # text block should have arrived in Viewer, but pause briefly
odoc = SpssClient.GetDesignatedOutputDoc()
items = odoc.GetOutputItems()
# Allow for some delay in the text block getting into the Viewer
for i in range(5):
size = items.Size()
item = items.GetItemAt(size - 1)
if item.GetType() == SpssClient.OutputItemType.TEXT:
break
time.sleep(.5)
specificitem = item.GetSpecificType()
specificitem.SetTextContents("".join(strings))
item.SetDescription(otitle)
items.GetItemAt(size - 2).SetVisible(False)
if pagebreak:
item.SetPageBreak(True)
SpssClient.StopClient()
return
if pagebreak and spss.GetDefaultPlugInVersion()[4:] >= '170':
import SpssClient
try:
SpssClient.StartClient()
items = SpssClient.GetDesignatedOutputDoc().GetOutputItems()
item = items.GetItemAt(items.Size() - 1)
item.SetPageBreak(True)
SpssClient.StopClient()
except: # do not propagate
SpssClient.StopClient()
def weightedkappaextension(variables, wttype=1, cilevel=95):
varnames = expandvarnames(variables)
caption = varnames[0] + _(" vs. ") + varnames[1]
vardict = spssaux.VariableDict(varnames)
if len(vardict) != len(varnames):
spss.StartProcedure(_("Weighted Kappa"), "Weighted Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row, "rowdim", hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""An invalid variable has been specified. This command is not executed."""
))
spss.EndProcedure()
elif len(varnames) != 2:
spss.StartProcedure(_("Weighted Kappa"), "Weighted Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row, "rowdim", hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""Exactly two variables must be specified. This command is not executed."""
))
spss.EndProcedure()
else:
try:
warntext = []
if cilevel < 50:
warntext.append(
_("CILEVEL cannot be less than 50%. It has been set to 50%."
))
cilevel = 50
if cilevel > 99.999:
warntext.append(
_("CILEVEL cannot be greater than 99.999%. It has been set to 99.999%."
))
cilevel = 99.999
if cilevel == int(cilevel):
cilevel = int(cilevel)
if wttype != 1:
if wttype != 2:
warntext.append(
_("WTTYPE must be 1 or 2. It has been set to 1."))
wttype = 1
varlist = varnames[0] + ' ' + varnames[1]
spss.Submit("PRESERVE.")
tempdir = tempfile.gettempdir()
spss.Submit("""CD "%s".""" % tempdir)
wtvar = spss.GetWeightVar()
if wtvar != None:
spss.Submit(r"""
COMPUTE %s=RND(%s).""" % (wtvar, wtvar))
spss.Submit(r"""
EXECUTE.""")
maxloops = 2 * spss.GetCaseCount()
spss.Submit("""SET PRINTBACK=OFF MPRINT=OFF MXLOOPS=%s.""" %
maxloops)
activeds = spss.ActiveDataset()
if activeds == "*":
activeds = "D" + str(random.uniform(.1, 1))
spss.Submit("DATASET NAME %s" % activeds)
tmpvar1 = "V" + str(random.uniform(.1, 1))
tmpvar2 = "V" + str(random.uniform(.1, 1))
tmpvar3 = "V" + str(random.uniform(.1, 1))
tmpvar4 = "V" + str(random.uniform(.1, 1))
tmpvar5 = "V" + str(random.uniform(.1, 1))
tmpvar6 = "V" + str(random.uniform(.1, 1))
tmpdata1 = "D" + str(random.uniform(.1, 1))
tmpdata2 = "D" + str(random.uniform(.1, 1))
omstag1 = "T" + str(random.uniform(.1, 1))
omstag2 = "T" + str(random.uniform(.1, 1))
omstag3 = "T" + str(random.uniform(.1, 1))
omstag4 = "T" + str(random.uniform(.1, 1))
omstag5 = "T" + str(random.uniform(.1, 1))
omstag6 = "T" + str(random.uniform(.1, 1))
tmpfile1 = "F" + str(random.uniform(.1, 1))
tmpfile2 = "F" + str(random.uniform(.1, 1))
lowlabel = _("""Lower %s%% Asymptotic CI Bound""") % cilevel
upplabel = _("""Upper %s%% Asymptotic CI Bound""") % cilevel
spss.Submit(r"""
DATASET COPY %s WINDOW=HIDDEN.""" % tmpdata1)
spss.Submit(r"""
DATASET ACTIVATE %s WINDOW=ASIS.""" % tmpdata1)
filt = spssaux.GetSHOW("FILTER", olang="english")
if filt != "No case filter is in effect":
filtcond = filt.strip("(FILTER)")
select = "SELECT IF " + str(filtcond) + "."
spss.Submit("""%s""" % select)
spss.Submit("""EXECUTE.""")
spss.Submit("""USE ALL.""")
banana = spssaux.getDatasetInfo(Info="SplitFile")
if banana != "":
warntext.append(_("This procedure ignores split file status."))
spss.Submit(r"""SPLIT FILE OFF.""")
spss.Submit(r"""
COUNT %s=%s (MISSING).""" % (tmpvar1, varlist))
spss.Submit(r"""
SELECT IF %s=0.""" % tmpvar1)
spss.Submit(r"""
EXECUTE.""")
validn = spss.GetCaseCount()
if validn < 2:
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Weighted Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag1)
spss.StartProcedure(_("Weighted Kappa"), "Weighted Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""There are too few complete cases. This command is not executed."""
))
spss.EndProcedure()
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag1)
else:
spss.Submit(r"""
AGGREGATE
/OUTFILE=* MODE=ADDVARIABLES
/%s=SD(%s)
/%s=SD(%s).""" % (tmpvar2, varnames[0], tmpvar3, varnames[1]))
try:
cur = spss.Cursor(isBinary=False)
except:
cur = spss.Cursor()
datarow = cur.fetchone()
cur.close()
sd1 = datarow[-2]
sd2 = datarow[-1]
if min(sd1, sd2) == 0:
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Weighted Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag1)
spss.StartProcedure(_("Weighted Kappa"), "Weighted Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""All ratings are the same for at least one rater. This command is not executed."""
))
spss.EndProcedure()
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag1)
else:
if len(warntext) > 0:
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Weighted Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag1)
if len(warntext) == 1:
spss.StartProcedure(_("Weighted Kappa"),
"Weighted Kappa")
table = spss.BasePivotTable(
"Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String("%s" %
warntext[0])
spss.EndProcedure()
if len(warntext) == 2:
spss.StartProcedure(_("Weighted Kappa"),
"Weighted Kappa")
table = spss.BasePivotTable(
"Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
"%s \n"
"%s" % (warntext[0], warntext[1]))
spss.EndProcedure()
if len(warntext) == 3:
spss.StartProcedure(_("Weighted Kappa"),
"Weighted Kappa")
table = spss.BasePivotTable(
"Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
"%s \n"
"%s \n"
"%s" % (warntext[0], warntext[1], warntext[2]))
spss.EndProcedure()
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag1)
spss.Submit(r"""
DELETE VARIABLES %s %s.""" % (tmpvar2, tmpvar3))
spss.Submit(r"""
AGGREGATE
/OUTFILE=%s
/BREAK=%s
/%s=N.""" % (tmpfile1, varlist, tmpvar4))
spss.Submit(r"""
OMS /SELECT ALL EXCEPT=WARNINGS
/IF COMMANDS=['Variables to Cases']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag2)
spss.Submit(r"""
VARSTOCASES
/MAKE %s FROM %s.""" % (tmpvar5, varlist))
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag2)
catdata = []
try:
cur = spss.Cursor(isBinary=False)
except:
cur = spss.Cursor()
while True:
datarow = cur.fetchone()
if datarow is None:
break
catdata.append(datarow[-1])
cur.close()
cats = list(set(catdata))
cattest = 0
if any(item != round(item) for item in cats):
cattest = 1
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Weighted Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag1)
spss.StartProcedure(_("Weighted Kappa"),
"Weighted Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""Some ratings are not integers. This command is not executed."""
))
spss.EndProcedure()
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag1)
elif min(cats) < 1.0:
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Weighted Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag1)
spss.StartProcedure(_("Weighted Kappa"),
"Weighted Kappa")
table = spss.BasePivotTable("Warnings ", "Warnings")
table.Append(spss.Dimension.Place.row,
"rowdim",
hideLabels=True)
rowLabel = CellText.String("1")
table[(rowLabel, )] = CellText.String(
_("""Some ratings are less than 1. This command is not executed."""
))
spss.EndProcedure()
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag1)
else:
spss.Submit(r"""
AGGREGATE
/OUTFILE=%s
/BREAK=%s
/%s=N.""" % (tmpfile2, tmpvar5, tmpvar6))
spss.Submit(r"""
DATASET DECLARE %s WINDOW=HIDDEN""" % tmpdata2)
spss.Submit(r"""
OMS /SELECT ALL EXCEPT=WARNINGS
/IF COMMANDS=['Matrix']
/DESTINATION VIEWER=NO
/TAG='"%s"'.""" % omstag3)
spss.Submit(r"""
MATRIX.
GET x
/FILE=%s
/VARIABLES=%s %s.
GET ratecats
/FILE=%s
/VARIABLES=%s.
COMPUTE size=MMAX(ratecats).
COMPUTE y=MAKE(size,size,0).
LOOP i=1 to NROW(y).
+ LOOP j=1 to NCOL(y).
+ LOOP k=1 to NROW(x).
+ DO IF (x(k,1)=i and x(k,2)=j).
+ COMPUTE y(i,j)=x(k,3).
+ END IF.
+ END LOOP.
+ END LOOP.
END LOOP.
COMPUTE wttype=%s.
COMPUTE wt=MAKE(NROW(y),NCOL(y),0).
LOOP i=1 to NROW(y).
+ LOOP j=1 to NCOL(y).
+ DO IF wttype=1.
+ COMPUTE wt(i,j)=1-(ABS(i-j)/(size-1)).
+ ELSE IF wttype=2.
+ COMPUTE wt(i,j)=1-((i-j)/(NROW(y)-1))**2.
+ END IF.
+ END LOOP.
END LOOP.
COMPUTE n=MSUM(y).
COMPUTE prop=y/n.
COMPUTE p_i=RSUM(prop).
COMPUTE p_j=CSUM(prop).
COMPUTE w_i=(wt*T(p_j))*MAKE(1,size,1).
COMPUTE w_j=MAKE(size,1,1)*(T(p_i)*wt).
COMPUTE po=MSUM(wt&*prop).
COMPUTE pe=MSUM(MDIAG(p_i)*wt*MDIAG(p_j)).
COMPUTE kstat=(po-pe)/(1-pe).
COMPUTE var0=(T(p_i)*((wt-(w_i+w_j))&**2)*T(p_j)-pe**2)/(n*(1-pe)**2).
DO IF var0>=0.
+ COMPUTE ase0=SQRT(var0).
ELSE.
+ COMPUTE ase0=-1.
END IF.
DO IF ase0>0.
+ COMPUTE z=kstat/ase0.
+ COMPUTE sig=1-CHICDF(z**2,1).
ELSE.
+ COMPUTE z=-1.
+ COMPUTE sig=-1.
END IF.
COMPUTE var1=(MSUM((prop&*((wt-(w_i+w_j)&*(1-kstat))&**2)))-(kstat-pe*(1-kstat))**2)/(n*(1-pe)**2).
DO IF var1>=0.
+ COMPUTE ase1=SQRT(var1).
ELSE.
+ COMPUTE ase1=-1.
END IF.
SAVE {wttype,kstat,ase1,z,sig,ase0}
/OUTFILE=%s
/VARIABLES=wttype,kstat,ase1,z,sig,ase0.
END MATRIX.""" % (tmpfile1, varlist, tmpvar4, tmpfile2, tmpvar5, wttype,
tmpdata2))
spss.Submit(r"""
OMSEND TAG=['"%s"'].""" % omstag3)
spss.Submit(r"""
DATASET ACTIVATE %s WINDOW=ASIS.""" % tmpdata2)
spss.Submit(r"""
DO IF ase0=-1.
+ RECODE z sig (-1=SYSMIS).
END IF.
EXECUTE.
DELETE VARIABLES ase0.
RECODE ase1 (-1=SYSMIS).
COMPUTE lower=kstat-SQRT(IDF.CHISQUARE(%s/100,1))*ase1.""" % cilevel)
spss.Submit(r"""
COMPUTE upper=kstat+SQRT(IDF.CHISQUARE(%s/100,1))*ase1.""" % cilevel)
spss.Submit(r"""
FORMATS kstat ase1 z sig lower upper (F11.3).
VARIABLE LABELS kstat %s.""" % _smartquote(_("""Kappa""")))
spss.Submit(r"""
VARIABLE LABELS ase1 %s.""" % _smartquote(_("""Asymptotic Standard Error""")))
spss.Submit(r"""
VARIABLE LABELS z %s.""" % _smartquote(_("""Z""")))
spss.Submit(r"""
VARIABLE LABELS sig %s. """ % _smartquote(_("""P Value""")))
spss.Submit(r"""
VARIABLE LABELS lower %s. """ % _smartquote(_(lowlabel)))
spss.Submit(r"""
VARIABLE LABELS upper %s. """ % _smartquote(_(upplabel)))
if wttype == 1:
spss.Submit(r"""
VARIABLE LABELS wttype %s.""" % _smartquote(_("""Linear""")))
if wttype == 2:
spss.Submit(r"""
VARIABLE LABELS wttype %s.""" % _smartquote(_("""Quadratic""")))
spss.Submit(r"""
EXECUTE.
""")
spss.Submit(r"""
OMS
/SELECT TABLES
/IF COMMANDS=['Weighted Kappa'] SUBTYPES=['Notes']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag4)
spss.Submit(r"""
OMS
/SELECT TEXTS
/IF COMMANDS=['Weighted Kappa'] LABELS=['Active Dataset']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag5)
if len(warntext) > 0:
spss.Submit(r"""
OMS
/SELECT HEADINGS
/IF COMMANDS=['Weighted Kappa']
/DESTINATION VIEWER=NO
/TAG = '"%s"'.""" % omstag6)
try:
cur = spss.Cursor(isBinary=False)
except:
cur = spss.Cursor()
data = cur.fetchone()
cur.close()
spss.StartProcedure(_("Weighted Kappa"),
"Weighted Kappa")
table = spss.BasePivotTable(_("Weighted Kappa"),
"Kappa",
caption=caption)
table.SimplePivotTable(rowdim = _("Weighting"),
rowlabels = [CellText.String(spss.GetVariableLabel(0))],
coldim = "",
collabels = [spss.GetVariableLabel(1),spss.GetVariableLabel(2),spss.GetVariableLabel(3),spss.GetVariableLabel(4), \
spss.GetVariableLabel(5),spss.GetVariableLabel(6)],
cells = [data[1],data[2],data[3],data[4],data[5],data[6]])
spss.EndProcedure()
if len(warntext) > 0:
spss.Submit(r"""
OMSEND TAG = ['"%s"'].""" % omstag6)
finally:
try:
spss.Submit(r"""
DATASET CLOSE %s.""" % tmpdata1)
spss.Submit(r"""
DATASET ACTIVATE %s WINDOW=ASIS.""" % activeds)
if validn >= 2:
if min(sd1, sd2) > 0:
if cattest == 0:
if min(cats) >= 1:
spss.Submit(r"""
OMSEND TAG=['"%s"' '"%s"'].""" % (omstag4, omstag5))
spss.Submit(r"""
DATASET CLOSE %s.""" % tmpdata2)
spss.Submit(r"""
ERASE FILE=%s.""" % tmpfile2)
spss.Submit(r"""
ERASE FILE=%s.""" % tmpfile1)
except:
pass
spss.Submit(r"""
RESTORE.
""")
pt2 = spss.BasePivotTable(
title=_("""Excluded Variables for Macro Definition: %s""") % macroname,
templateName="COUNTMACROEX",
isSplit= False
)
if not weightvar:
pt2.SetDefaultFormatSpec(spss.FormatSpec.Count)
rowlabels = [item[0] for item in sorted(excluded.items(), key=lambda x: x[1],
reverse = order == "d")]
pt2.SimplePivotTable(
rowdim=_("""Excluded Variables"""),
rowlabels= [spss.CellText.VarName(vardict[v].VariableIndex) for v in rowlabels],
collabels=[_("""Counts""")],
cells = [excluded[v] for v in rowlabels]
)
spss.EndProcedure()
class ManageValues(object):
"""Manage mr set values"""
def __init__(self, mrsetinfo, allvars, allvalues, specials,
order, macroname, categorylabels, specialsorder, other):
"""mrsetinfo is the structure returned by the dataset multiResposneSet api
Set names are always in upper case
allvars is a list of all the expanded variable names across all the
specified variable sets.
allvalues is a list of dictionaries.
Each list item is a dictionary of numeric or string values and their
count for one variable
specials is the list of special values