def analyzeCSV(groupingType, model, type=2, analysisType="3",limit = "100"):
"""
Parameters.
model : model to analyize
type: type: 1 -> single level sim analysis
2 -> range analysis
Get all sim files in model
iterate over categories
Distinguish between sngle and range sim files
Get top n row from sim file, n = numer of items searched for
Calculate IR measures
"""
keys = ['category', 'depth','idLevel','ocID','sim','nrOcc']
path = "testData/%s/%s/sim/SummaryCSV/%s/" %(groupingType,model,limit)
categories = getMainCat()
for cat in categories:
if type==1:
sysOutFile = path+cat+".txt"
else:
sysOutFile = path+cat+"_Range.txt"
sys.stdout = open(sysOutFile, 'wb')
#get cat debth
sqlmaxDepth = "select max(categoryDepth) from dmoz_categories where Topic like 'Top/"+str(cat)+"/%' and filterOut = 0"
maxDebthRS = dbQuery(sqlmaxDepth)
maxDebth = maxDebthRS[0]
maxDebth = maxDebth[0]
ranger = [x for x in range(2,maxDebth+1)]
print "Category: ", cat
for level in ranger:
if type == 1:
fileName = "%s%s_%s_%s.csv" % (path,model,cat,level)
originalIDFile = "%s%s_%s_%s_original.csv" % (path,model,cat,level)
elif type == 2:
fileName = "%s%s_%s_1_%s.csv" % (path,model,cat,level)
originalIDFile = "%s%s_%s_1_%s_original.csv" % (path,model,cat,level)
#original categories from sim csv or oid csv
if os.path.isfile(originalIDFile):
nrUID = uniq(originalIDFile, 0)
else:
nrUID = uniq(fileName, 2)
nrUID = [int(i) for i in nrUID]
#print fileName
returnedID = uniq(fileName, 3)
returnedID = [int(i) for i in returnedID[:len(nrUID)]]
print "level: ", level,"\t",fileName
#print nrUID
#print returnedID
print len(nrUID)
print len(returnedID)
if len(nrUID) ==len(returnedID):
print sklearn.metrics.classification_report(nrUID,returnedID)
def visit(self, recurse, directory, names):
settings = self.get_settings('', directory)
if settings.prune and (os.path.abspath(directory) in settings.prune):
print >> sys.stderr, '/// ...Skipping directory (pruned):', directory
sys.stderr.flush()
names[:] = []
return
if not self.initial_settings.silent:
print >> sys.stderr, '/// Processing directory:', directory
sys.stderr.flush()
# settings.ignore grows many duplicate entries as we recurse
# if we add patterns in config files or on the command line.
for pattern in utils.uniq(settings.ignore):
for i in range(len(names) - 1, -1, -1):
if fnmatch(names[i], pattern):
# Modify in place!
del names[i]
prune = 0
for name in names:
if name.endswith('.txt'):
prune = self.process_txt(directory, name)
if prune:
break
if not recurse:
del names[:]
def visit(self, directory, names):
# BUG prune and ignore do not work
settings = self.get_settings('', directory)
errout = ErrorOutput(encoding=settings.error_encoding)
if settings.prune and (os.path.abspath(directory) in settings.prune):
print >>errout, ('/// ...Skipping directory (pruned): %s' %
directory)
sys.stderr.flush()
names[:] = []
return
if not self.initial_settings.silent:
print >>errout, '/// Processing directory: %s' % directory
sys.stderr.flush()
# settings.ignore grows many duplicate entries as we recurse
# if we add patterns in config files or on the command line.
for pattern in utils.uniq(settings.ignore):
for i in range(len(names) - 1, -1, -1):
if fnmatch(names[i], pattern):
# Modify in place!
del names[i]
prune = 0
for name in names:
if name.endswith('.txt'):
prune = self.process_txt(directory, name)
if prune:
break
def indirect_target_error(self, target, explanation):
naming = ''
reflist = []
if target['names']:
naming = '"%s" ' % target['names'][0]
for name in target['names']:
reflist.extend(self.document.refnames.get(name, []))
for id in target['ids']:
reflist.extend(self.document.refids.get(id, []))
naming += '(id="%s")' % target['ids'][0]
msg = self.document.reporter.error(
'Indirect hyperlink target %s refers to target "%s", %s.' %
(naming, target['refname'], explanation),
base_node=target)
msgid = self.document.set_id(msg)
for ref in utils.uniq(reflist):
prb = nodes.problematic(ref.rawsource, ref.rawsource, refid=msgid)
prbid = self.document.set_id(prb)
msg.add_backref(prbid)
ref.replace_self(prb)
target.resolved = 1
def indirect_target_error(self, target, explanation):
naming = ''
reflist = []
if target['names']:
naming = '"%s" ' % target['names'][0]
for name in target['names']:
reflist.extend(self.document.refnames.get(name, []))
for id in target['ids']:
reflist.extend(self.document.refids.get(id, []))
naming += '(id="%s")' % target['ids'][0]
msg = self.document.reporter.error(
'Indirect hyperlink target %s refers to target "%s", %s.'
% (naming, target['refname'], explanation), base_node=target)
msgid = self.document.set_id(msg)
for ref in utils.uniq(reflist):
prb = nodes.problematic(
ref.rawsource, ref.rawsource, refid=msgid)
prbid = self.document.set_id(prb)
msg.add_backref(prbid)
ref.replace_self(prb)
target.resolved = 1
def visit(self, directory, names, subdirectories):
settings = self.get_settings("", directory)
errout = ErrorOutput(encoding=settings.error_encoding)
if settings.prune and (os.path.abspath(directory) in settings.prune):
errout.write("/// ...Skipping directory (pruned): %s\n" % directory)
sys.stderr.flush()
del subdirectories[:]
return
if not self.initial_settings.silent:
errout.write("/// Processing directory: %s" % directory)
sys.stderr.flush()
# settings.ignore grows many duplicate entries as we recurse
# if we add patterns in config files or on the command line.
for pattern in utils.uniq(settings.ignore):
for i in range(len(names) - 1, -1, -1):
if fnmatch(names[i], pattern):
# Modify in place!
del names[i]
for name in names:
if name.endswith(".txt"):
self.process_txt(directory, name)
def visit(self, directory, names, subdirectories):
settings = self.get_settings('', directory)
errout = ErrorOutput(encoding=settings.error_encoding)
if settings.prune and (os.path.abspath(directory) in settings.prune):
errout.write('/// ...Skipping directory (pruned): %s\n' %
directory)
sys.stderr.flush()
del subdirectories[:]
return
if not self.initial_settings.silent:
errout.write('/// Processing directory: %s\n' % directory)
sys.stderr.flush()
# settings.ignore grows many duplicate entries as we recurse
# if we add patterns in config files or on the command line.
for pattern in utils.uniq(settings.ignore):
for i in range(len(names) - 1, -1, -1):
if fnmatch(names[i], pattern):
# Modify in place!
del names[i]
for name in names:
if name.endswith('.txt'):
self.process_txt(directory, name)
def visit(self, recurse, directory, names):
settings = self.get_settings("", directory)
if settings.prune and (os.path.abspath(directory) in settings.prune):
print >>sys.stderr, "/// ...Skipping directory (pruned):", directory
sys.stderr.flush()
names[:] = []
return
if not self.initial_settings.silent:
print >>sys.stderr, "/// Processing directory:", directory
sys.stderr.flush()
# settings.ignore grows many duplicate entries as we recurse
# if we add patterns in config files or on the command line.
for pattern in utils.uniq(settings.ignore):
for i in range(len(names) - 1, -1, -1):
if fnmatch(names[i], pattern):
# Modify in place!
del names[i]
for name in names:
if name.endswith(".txt"):
if self.process_txt(directory, name):
break # prune
if not recurse:
del names[:]
def analyzeSummary(groupingType, model="0.1"):
"""
Make graphs from summary files: input
CSV summary structure:
"Category","Level","Model","docsInModel","ReturnedDocsForModel","NrInputDocs"
"""
path = "testData/%s/%s/summary_100.csv" %(groupingType,model)
labels = []
categories = uniq(path, 0)
plt.rc('legend',**{'fontsize':6})
contentSingleDocsInModel = []
contentSingleReturnedDocs= []
contentRangeDocsInModel= []
contentRangeReturnedDocs= []
rangeLabels = []
#graphs for summary
for cat in categories:
content = []
#range data
f = open(path, "rb")
reader = csv.reader(f)
contentRange = [row for row in reader if row[0] == cat and row[2].count('_') == 3]
tempRangeDocsInModel = [int(x[3]) for x in contentRange]
tempRangerangeReturnedDocsForModel = [int(x[4]) for x in contentRange]
f.close()
#single data
f = open(path, "rb") # don't forget the 'b'!
reader = csv.reader(f)
contentSingle = [row for row in reader if row[0] == cat and row[2].count('_') == 2]
#values
singleDocsInModel = [int(x[3]) for x in contentSingle]
singleReturnedDocsForModel = [int(x[4]) for x in contentSingle]
position = len(singleDocsInModel) + 1
while position < 14:
singleDocsInModel.insert(position,0)
singleReturnedDocsForModel.insert(position,0)
tempRangeDocsInModel.insert(position,0)
tempRangerangeReturnedDocsForModel.insert(position,0)
position += 1
#single model
contentSingleDocsInModel.append(singleDocsInModel)
contentSingleReturnedDocs.append(singleReturnedDocsForModel)
#range model
contentRangeDocsInModel.append(tempRangeDocsInModel)
contentRangeReturnedDocs.append(tempRangerangeReturnedDocsForModel)
f.close()
#labels
labels.append(cat)
#print contentSingleDocsInModel
#print contentSingleReturnedDocs
summaryGraph(contentSingleDocsInModel, contentSingleReturnedDocs,labels, "single",model)
summaryGraph(contentRangeDocsInModel, contentRangeReturnedDocs,labels, "range", model)
