def performanceForCDAITAt(noOfTweets, fileName, **stream_settings):
ts = time.time()
sstObject = SimilarStreamAggregation(dict(iterateTweetUsersAfterCombiningTweets(fileName, **stream_settings)), stream_settings['ssa_threshold'])
sstObject.estimate()
documentClusters = list(sstObject.iterateClusters())
te = time.time()
return Evaluation.getEvaluationMetrics(noOfTweets, documentClusters, te-ts)
def getStatsForSSA(self):
print "SSA"
ts = time.time()
sstObject = SimilarStreamAggregation(dict(self._iterateUserDocuments()), self.stream_settings["ssa_threshold"])
sstObject.estimate()
documentClusters = list(sstObject.iterateClusters())
te = time.time()
return self.getEvaluationMetrics(documentClusters, te - ts)
def getStatsForSSA(self):
print 'SSA'
ts = time.time()
sstObject = SimilarStreamAggregation(
dict(self._iterateUserDocuments()),
self.stream_settings['ssa_threshold'])
sstObject.estimate()
documentClusters = list(sstObject.iterateClusters())
te = time.time()
return self.getEvaluationMetrics(documentClusters, te - ts)
def performanceForCDAITAt(noOfTweets, fileName, **stream_settings):
ts = time.time()
sstObject = SimilarStreamAggregation(
dict(
iterateTweetUsersAfterCombiningTweets(fileName,
**stream_settings)),
stream_settings['ssa_threshold'])
sstObject.estimate()
documentClusters = list(sstObject.iterateClusters())
te = time.time()
return Evaluation.getEvaluationMetrics(noOfTweets, documentClusters,
te - ts)
def test_estimate(self):
nn = SimilarStreamAggregation(vectors, 0.99)
nn.estimate()
self.assertEqual([['1', '3', '2'], ['5', '7']], list(nn.iterateClusters()))
