4ace4ef2fb1956ec4df46f78c9edd02154780913,bugbug/similarity.py,NeighborsSimilarity,__init__,#NeighborsSimilarity#Any#Any#,202

Before Change

class NeighborsSimilarity(BaseSimilarity):
    def __init__(self, k=10, vectorizer=TfidfVectorizer()):
        self.vectorizer = vectorizer
        self.similarity_calculator = NearestNeighbors(n_neighbors=k)
        text = []
        self.bug_ids = []

        for bug in bugzilla.get_bugs():
            text.append(text_preprocess(get_text(bug), join=True))
            self.bug_ids.append(bug["id"])

        self.vectorizer.fit(text)
        self.similarity_calculator.fit(self.vectorizer.transform(text))

    def get_similar_bugs(self, query):

        processed_query = self.vectorizer.transform([get_text(query)])

After Change

        indexes = list(range(len(self.corpus)))
        random.shuffle(indexes)
        self.corpus = [self.corpus[idx] for idx in indexes]
        self.bug_ids = [self.bug_ids[idx] for idx in indexes]

        self.w2vmodel = Word2Vec(self.corpus, size=100, min_count=5)
        self.w2vmodel.init_sims(replace=True)

Instances