Python 基于特定查询从SparSematrix相似度gensim中获取前N个最相似向量

我看到了

gensim

文档，但遇到了以下问题：

-一旦我得到一个

mXj

相似性矩阵，其中

m

是文档的数量，

j

是唯一单词的总数，我不知道如何提取最相似的

N

文档

-长期目标是以

xlsx

或

csv

格式存储和保存，但这是另一个问题

这里使用

相似性

from gensim.test.utils import common_corpus, common_dictionary, get_tmpfile
index_tmpfile = get_tmpfile("index")

query = [(1, 2), (6, 1), (7, 2)]

index = Similarity(index_tmpfile, common_corpus, num_features=len(common_dictionary))  # build the index

similarities = index[query]  # get similarities between the query and all index documents

在那之后，我得到了10个最相似的文档

预期产出：

arr=[('This is the most similar',0.99),('This is one of the most similar',0.98),('This is another very similar docs,0.98)]

或

在我个人的情况下，我有以下使用相似性的代码

#type here the query
query_vector='house is clean'
tokenized_query_vector = regexp_tokenize(query_vector,r"\w+")  

#Create an object of corpora.Dictionary() 
dictionary = corpora.Dictionary()
#Passing to dictionary.doc2bow() object
BoW_corpus = [dictionary.doc2bow(doc, allow_update=True) for doc in tokenized_lines]
feature_cnt = len(dictionary.token2id)

#n=raw
#t=zero corrected idf
#c=cosine
tfidf = TfidfModel(BoW_corpus, smartirs='ntc')

query_vector = dictionary.doc2bow(tokenized_query_vector)

index = similarities.SparseMatrixSimilarity(tfidf[BoW_corpus], num_features = feature_cnt)
#this way I get the similarity between the query and the similarity matrix
sim = index[tfidf[query_vector]]

#type here the query
query_vector='house is clean'
tokenized_query_vector = regexp_tokenize(query_vector,r"\w+")  

#Create an object of corpora.Dictionary() 
dictionary = corpora.Dictionary()
#Passing to dictionary.doc2bow() object
BoW_corpus = [dictionary.doc2bow(doc, allow_update=True) for doc in tokenized_lines]
feature_cnt = len(dictionary.token2id)

#n=raw
#t=zero corrected idf
#c=cosine
tfidf = TfidfModel(BoW_corpus, smartirs='ntc')

query_vector = dictionary.doc2bow(tokenized_query_vector)

index = similarities.SparseMatrixSimilarity(tfidf[BoW_corpus], num_features = feature_cnt)
#this way I get the similarity between the query and the similarity matrix
sim = index[tfidf[query_vector]]