from gensim.models import Word2Vec as wtv
import jiebas1 = "刘新宇是一个自然语言处理算法工程师"
s2 = "他从事算法方面的工作很多年"
s3 = "刘新宇是一名非常资深的算法工程师"sentences = [jieba.lcut(s1),jieba.lcut(s2),jieba.lcut(s3)]
print(sentences)model = wtv(sentences,min_count=1)print("this is the summary of the model:")
print(model)words = list(model.wv.vocab)
print('this is the vocabulary for our corpus:')
print(words)print("the vector for the word singer:")
print(model['资深'])w1 = ['资深']
print(model.wv.most_similar(positive=w1,topn=5))
结果：
[['刘新宇', '是', '一个', '自然语言', '处理', '算法', '工程师'], ['他', '从事', '算法', '方面', '的', '工作', '很多年'], ['刘新宇', '是', '一名', '非常', '资深', '的', '算法', '工程师']]
this is the summary of the model:
Word2Vec(vocab=16, size=100, alpha=0.025)
this is the vocabulary for our corpus:
['刘新宇', '是', '一个', '自然语言', '处理', '算法', '工程师', '他', '从事', '方面', '的', '工作', '很多年', '一名', '非常', '资深']
the vector for the word singer:
[-3.7734744e-03 2.6095781e-04 -3.1862229e-03 4.2619775e-03-4.5439765e-07 -1.8814839e-03 -1.9608627e-03 -4.3091276e-032.0104370e-03 2.4313631e-03 3.6958823e-04 -4.6373094e-03-3.8697310e-03 -1.3030261e-03 3.4550454e-03 3.6497047e-04-3.0894275e-03 2.7997990e-04 2.4094102e-03 3.7222090e-03-3.3406885e-03 3.8989806e-03 -1.9044009e-03 -1.2546520e-03-2.7420574e-03 1.9589183e-03 4.6422374e-03 2.2412005e-03-4.5687910e-03 -3.0045302e-03 -7.5545040e-04 -1.9913551e-03-5.4244912e-04 3.4368648e-03 3.1319596e-03 3.5465839e-034.0631965e-04 -8.0981071e-04 -4.5625023e-03 1.8198669e-03-2.3173515e-03 1.3505023e-03 4.2371401e-03 4.9272538e-03-2.1169472e-03 4.2408700e-03 3.7939013e-03 -1.9469961e-03-1.0268842e-03 -3.4248799e-03 -4.0382403e-03 -8.0300641e-04-4.3166843e-03 4.0071514e-03 -5.1711878e-04 -1.0944011e-03-9.2390249e-04 -2.0183630e-03 -1.1692114e-04 3.7988871e-033.0223508e-03 -2.7847637e-03 -2.0709957e-03 -3.2283876e-03-7.3188142e-04 6.4730411e-06 2.4504904e-03 -1.8213416e-03-7.2910590e-04 -3.1336993e-03 -2.1612353e-03 3.4241637e-039.4859622e-04 -1.1737887e-03 -4.3117562e-03 -2.7182067e-03-6.3206785e-04 3.5696046e-03 -2.9301932e-03 9.1675809e-041.7115782e-03 1.6887123e-03 4.1562999e-03 3.1984923e-032.4283223e-04 4.2053428e-03 2.4675422e-03 2.9653152e-034.8725074e-03 3.6773803e-03 3.8778691e-03 7.9092768e-043.6476396e-03 -1.1947335e-03 -2.0735445e-03 2.1955518e-03-1.3067436e-03 -2.5959394e-03 4.0679227e-04 7.9005008e-04]
[('工作', 0.1197129413485527), ('非常', 0.08151963353157043), ('一个', 0.061945877969264984), ('的', 0.05767860636115074), ('他', 0.03252919763326645)]
C:/Users/user/Desktop/pycharm_test/test_gensim.py:21: DeprecationWarning: Call to deprecated `__getitem__` (Method will be removed in 4.0.0, use self.wv.__getitem__() instead).print(model['资深'])
Word2vec可调整的参数如下：
def __init__(self, sentences=None, corpus_file=None, size=100, alpha=0.025, window=5, min_count=5,max_vocab_size=None, sample=1e-3, seed=1, workers=3, min_alpha=0.0001,sg=0, hs=0, negative=5, ns_exponent=0.75, cbow_mean=1, hashfxn=hash, iter=5, null_word=0,trim_rule=None, sorted_vocab=1, batch_words=MAX_WORDS_IN_BATCH, compute_loss=False, callbacks=(),max_final_vocab=None):
