def inputAnalysis(self, corpus):
self.resetWidget()
self.corpus = corpus
OWWidget.progressBarInit(self)
duplicateRefStartEndsArray = list(self.EMPTYLIST)
if self.corpus is None:
self.label.setText("No corpus available")
else:
text = self.EMPTYSTRING
self.fieldIdDate = self.getFieldId(self.corpus, self.FIELDNAMEDATE)
self.fieldIdText = self.getFieldId(self.corpus, self.FIELDNAMETEXT)
self.fieldIdExtra = self.getFieldId(self.corpus,
self.FIELDNAMEEXTRA)
for msgId in range(0, len(self.corpus.metas)):
date = datetime.datetime.fromtimestamp(
self.corpus.metas[msgId][self.fieldIdDate],
tz=datetime.timezone.utc)
text = self.prepareText(
str(self.corpus.metas[msgId][self.fieldIdText]))
duplicateRefStartEnds = self.countPhrases(date, text, msgId)
duplicateRefStartEndsArray.append(
[list(duplicateRefStartEnds)])
self.label.setText(str(duplicateRefStartEnds))
self.corpus.metas[msgId][self.fieldIdExtra] = list(
duplicateRefStartEnds)
self.corpus.metas[msgId][
self.fieldIdText] = self.markDuplicates(
text, duplicateRefStartEnds)
OWWidget.progressBarSet(
self, 100 * (msgId + 1) / len(self.corpus.metas))
# np.append(self.corpus.metas,np.array(duplicateRefStartEndsArray),axis=1)
self.Outputs.corpus.send(self.corpus)
def inputAnalysis(self, corpus):
self.resetWidget()
self.corpus = corpus
OWWidget.progressBarInit(self)
if self.corpus is None:
self.label.setText("No corpus available")
else:
text = self.EMPTYSTRING
self.fieldIdExtra = self.getFieldId(self.corpus,
self.FIELDNAMEEXTRA)
self.fieldIdText = self.getFieldId(self.corpus, self.FIELDNAMETEXT)
for msgId in range(0, len(self.corpus.metas)):
text = self.prepareText(
str(self.corpus.metas[msgId][self.fieldIdText]))
averageWeights = self.daap(text)
averageWeights = np.mean(averageWeights)
OWWidget.progressBarSet(self, len(self.corpus))
owfeatureconstructor.OWFeatureConstructor()
self.corpus.metas[msgId][self.fieldIdExtra] = averageWeights
self.Outputs.corpus.send(self.corpus)
