gensim載入word2vec訓練結果(bin檔案)並進行相似度實驗

# -*- coding: utf-8 -*-  
import gensim  

# 匯入模型
model = gensim.models.KeyedVectors.load_word2vec_format('vectors.bin', binary=True)

# 得到兩組詞的相似度
list1 = [u'核能']
list2 = [u'電能']
list3 = [u'電力']
list_sim1 =  model.n_similarity(list1, list2)
print list_sim1
list_sim2 = model.n_similarity(list2, list3)
print list_sim2, '\n'

# 得到一組詞中最無關的詞
list4 = [u'汽車', u'火車', u'飛機', u'北京']
print model.doesnt_match(list4)
print '\n'

# 得到與一個詞最相關的若干詞及相似程度
result = model.most_similar(u'脫水工藝')
for each in result:
    print each[0] , each[1]