def removeLabels(self, start=None, end=None, labelFilter=None):
"""
Remove labels from each record with record ROWID in range from
start to end, noninclusive of end. Removes all records if labelFilter is
None, otherwise only removes the labels eqaul to labelFilter.
This will recalculate all points from end to the last record stored in the
internal cache of this classifier.
"""
if len(self.saved_states) == 0:
raise CLAModelInvalidRangeError("Invalid supplied range for "
"'removeLabels'. Model has no saved records.")
startID = self.saved_states[0].ROWID
clippedStart = 0 if start is None else max(0, start - startID)
clippedEnd = len(self.saved_states) if end is None else \
max(0, min( len( self.saved_states) , end - startID))
if clippedEnd <= clippedStart:
raise CLAModelInvalidRangeError("Invalid supplied range for "
"'removeLabels'.", debugInfo={
'requestRange': {
'startRecordID': start,
'endRecordID': end
},
'clippedRequestRange': {
'startRecordID': clippedStart,
'endRecordID': clippedEnd
},
'validRange': {
'startRecordID': startID,
'endRecordID': self.saved_states[len(self.saved_states)-1].ROWID
},
'numRecordsStored': len(self.saved_states)
})
# Remove records within the cache
recordsToDelete = []
for state in self.saved_states[clippedStart:clippedEnd]:
if labelFilter is not None:
if labelFilter in state.anomalyLabel:
state.anomalyLabel.remove(labelFilter)
else:
state.anomalyLabel = []
state.setByUser = False
recordsToDelete.append(state)
self._deleteRecordsFromKNN(recordsToDelete)
# Remove records not in cache
self._deleteRangeFromKNN(start, end)
# Recompute [clippedEnd, ...)
for state in self.saved_states[clippedEnd:]:
self._updateState(state)
return {'status': 'success'}
def addLabel(self, start, end, labelName):
"""
Add the label labelName to each record with record ROWID in range from
start to end, noninclusive of end.
This will recalculate all points from end to the last record stored in the
internal cache of this classifier.
"""
if len(self._recordsCache) == 0:
raise CLAModelInvalidRangeError("Invalid supplied range for 'addLabel'. "
"Model has no saved records.")
try:
start = int(start)
except Exception:
start = 0
try:
end = int(end)
except Exception:
end = int(self._recordsCache[-1].ROWID)
startID = self._recordsCache[0].ROWID
clippedStart = max(0, start - startID)
clippedEnd = max(0, min( len( self._recordsCache) , end - startID))
if clippedEnd <= clippedStart:
raise CLAModelInvalidRangeError("Invalid supplied range for 'addLabel'.",
debugInfo={
'requestRange': {
'startRecordID': start,
'endRecordID': end
},
'clippedRequestRange': {
'startRecordID': clippedStart,
'endRecordID': clippedEnd
},
'validRange': {
'startRecordID': startID,
'endRecordID': self._recordsCache[len(self._recordsCache)-1].ROWID
},
'numRecordsStored': len(self._recordsCache)
})
# Add label to range [clippedStart, clippedEnd)
for state in self._recordsCache[clippedStart:clippedEnd]:
if labelName not in state.anomalyLabel:
state.anomalyLabel.append(labelName)
state.setByUser = True
self._addRecordToKNN(state)
assert len(self.saved_categories) > 0
# Recompute [end, ...)
for state in self._recordsCache[clippedEnd:]:
self.classifyState(state)
def getLabels(self, start=None, end=None):
"""
Get the labels on classified points within range start to end. Not inclusive
of end.
reval - dict of format:
{
'isProcessing': boolean,
'recordLabels': list of results
}
isProcessing - currently always false as recalculation blocks; used if
reprocessing of records is still being performed;
Each item in recordLabels is of format:
{
'ROWID': id of the row,
'labels': list of strings
}
"""
if len(self._recordsCache) == 0:
return {'isProcessing': False, 'recordLabels': []}
try:
start = int(start)
except Exception:
start = 0
try:
end = int(end)
except Exception:
end = self._recordsCache[-1].ROWID
if end <= start:
raise CLAModelInvalidRangeError(
"Invalid supplied range for 'getLabels'.",
debugInfo={
'requestRange': {
'startRecordID': start,
'endRecordID': end
},
'numRecordsStored': len(self._recordsCache)
})
results = {'isProcessing': False, 'recordLabels': []}
ROWIDX = numpy.array(
self._knnclassifier.getParameter('categoryRecencyList'))
validIdx = numpy.where((ROWIDX >= start) & (ROWIDX < end))[0].tolist()
categories = self._knnclassifier.getCategoryList()
for idx in validIdx:
row = dict(ROWID=int(ROWIDX[idx]),
labels=self._categoryToLabelList(categories[idx]))
results['recordLabels'].append(row)
return results
def getLabels(self, start=None, end=None):
if len(self.saved_states) == 0:
return {
'isProcessing': False,
'recordLabels': []
}
if start is None:
start = 0
if end is None:
end = self.saved_states[-1].ROWID
if end <= start:
raise CLAModelInvalidRangeError("Invalid supplied range for 'getLabels'.",
debugInfo={
'requestRange': {
'startRecordID': start,
'endRecordID': end
},
'numRecordsStored': len(self.saved_states)
})
results = {
'isProcessing': False,
'recordLabels': []
}
classifier = self.clamodel._getAnomalyClassifier()
knn = classifier.getSelf()._knn
ROWIDX = numpy.array(
classifier.getSelf().getParameter('categoryRecencyList'))
validIdx = numpy.where((ROWIDX >= start) & (ROWIDX < end))[0].tolist()
categories = knn._categoryList
for idx in validIdx:
row = dict(
ROWID=int(ROWIDX[idx]),
labels=self._categoryToLabelList(categories[idx]))
results['recordLabels'].append(row)
return results
