def test_calcDiffs():
cvs = finddup.loadLocalCaseversions('tests/data/small_274_0.json')
comb_it = finddup.genAllCombinations(cvs)
selected_pairs = [next(comb_it) for i in range(2)]
diffs = filters.calcDiffs(cvs, selected_pairs)
assert (len(selected_pairs) == len(diffs))
assert (type(diffs[0]) == type([]))
assert (type(diffs[0][0]) == type(""))
def extractFeatures(caseversions, selected_pairs):
#caseversions_sorted_by_id = sorted(caseversions['objects'], key=lambda x: x['id'])
#print(caseversions_sorted_by_id)
#idx_from_caseversion_id = dict((str(d['id']), i) for (i, d) in enumerate(caseversions_sorted_by_id))
#TODO: can we reduce the number of cases here?
#TODO: find the intersection between the groundtruth and the caseversions
#caseversion_texts = map(lambda x: json.dumps(x), caseversions_sorted_by_id)
logging.info("Prepare to extract features from " +
str(len(selected_pairs)) + " pairs")
counter = 0
# Extracting similarity related features
p = ProgressBar(3)
p.update(1)
#similarities = filters.calcSimilarity(caseversions, selected_pairs)
#TODO enable similarity
#vect = TfidfVectorizer(min_df=1)
#tfidf = vect.fit_transform(caseversion_texts)
#pairwise_similarity = tfidf * tfidf.T
#p = ProgressBar(len(selected_pairs))
# Extracting diff related features
diffs = filters.calcDiffs(caseversions, selected_pairs)
p.update(2)
isonoffs = map(filters.isOnOffPairs, diffs)
p.update(3)
isdiffmodules = map(filters.isDifferentModule, diffs)
#isdiffmodule = filters.isDifferentModule(diff)
# Feature re-formatting
def toDict(fields):
return {
"isonoff": fields[0],
"isdiffmodule": fields[1],
#"similarity": fields[2]
}
#features = map(toDict, zip(isonoffs, isdiffmodules, similarities))
features = map(toDict, zip(isonoffs, isdiffmodules))
vec = DictVectorizer()
vectorized_features = vec.fit_transform(features)
p.done()
#p.done()
return vectorized_features
def extractFeatures(caseversions, selected_pairs):
#caseversions_sorted_by_id = sorted(caseversions['objects'], key=lambda x: x['id'])
#print(caseversions_sorted_by_id)
#idx_from_caseversion_id = dict((str(d['id']), i) for (i, d) in enumerate(caseversions_sorted_by_id))
#TODO: can we reduce the number of cases here?
#TODO: find the intersection between the groundtruth and the caseversions
#caseversion_texts = map(lambda x: json.dumps(x), caseversions_sorted_by_id)
logging.info("Prepare to extract features from " + str(len(selected_pairs)) + " pairs")
counter = 0
# Extracting similarity related features
p = ProgressBar(3)
p.update(1)
#similarities = filters.calcSimilarity(caseversions, selected_pairs)
#TODO enable similarity
#vect = TfidfVectorizer(min_df=1)
#tfidf = vect.fit_transform(caseversion_texts)
#pairwise_similarity = tfidf * tfidf.T
#p = ProgressBar(len(selected_pairs))
# Extracting diff related features
diffs = filters.calcDiffs(caseversions, selected_pairs)
p.update(2)
isonoffs = map(filters.isOnOffPairs, diffs)
p.update(3)
isdiffmodules = map(filters.isDifferentModule, diffs)
#isdiffmodule = filters.isDifferentModule(diff)
# Feature re-formatting
def toDict(fields):
return {
"isonoff": fields[0],
"isdiffmodule": fields[1],
#"similarity": fields[2]
}
#features = map(toDict, zip(isonoffs, isdiffmodules, similarities))
features = map(toDict, zip(isonoffs, isdiffmodules))
vec = DictVectorizer()
vectorized_features = vec.fit_transform(features)
p.done()
#p.done()
return vectorized_features
