python如何實(shí)現(xiàn)word2vec

發(fā)布時(shí)間：2020-07-17 14:36:02 來(lái)源：億速云閱讀：146 作者：小豬欄目：開發(fā)技術(shù)

這篇文章主要為大家展示了python如何實(shí)現(xiàn)word2vec，內(nèi)容簡(jiǎn)而易懂，希望大家可以學(xué)習(xí)一下，學(xué)習(xí)完之后肯定會(huì)有收獲的，下面讓小編帶大家一起來(lái)看看吧。

一、前言

一開始看到word2vec環(huán)境的安裝還挺復(fù)雜的，安了半天Cygwin也沒太搞懂。后來(lái)突然發(fā)現(xiàn)，我為什么要去安c語(yǔ)言版本的呢，我應(yīng)該去用python版本的，然后就發(fā)現(xiàn)了gensim，安裝個(gè)gensim的包就可以用word2vec了，不過gensim只實(shí)現(xiàn)了word2vec里面的skip-gram模型。若要用到其他模型，就需要去研究其他語(yǔ)言的word2vec了。

二、語(yǔ)料準(zhǔn)備

有了gensim包之后，看了網(wǎng)上很多教程都是直接傳入一個(gè)txt文件，但是這個(gè)txt文件長(zhǎng)啥樣，是什么樣的數(shù)據(jù)格式呢，很多博客都沒有說明，也沒有提供可以下載的txt文件作為例子。進(jìn)一步理解之后發(fā)現(xiàn)這個(gè)txt是一個(gè)包含巨多文本的分好詞的文件。如下圖所示，是我自己訓(xùn)練的一個(gè)語(yǔ)料，我選取了自己之前用爬蟲抓取的7000條新聞當(dāng)做語(yǔ)料并進(jìn)行分詞。注意，詞與詞之間一定要用空格：

python如何實(shí)現(xiàn)word2vec

這里分詞使用的是結(jié)巴分詞。

這部分代碼如下：

import jieba
f1 =open("fenci.txt")
f2 =open("fenci_result.txt", 'a')
lines =f1.readlines() # 讀取全部?jī)?nèi)容
for line in lines:
  line.replace('\t', '').replace('\n', '').replace(' ','')
  seg_list = jieba.cut(line, cut_all=False)
  f2.write(" ".join(seg_list))
 
f1.close()
f2.close()

還要注意的一點(diǎn)就是語(yǔ)料中的文本一定要多，看網(wǎng)上隨便一個(gè)語(yǔ)料都是好幾個(gè)G，而且一開始我就使用了一條新聞當(dāng)成語(yǔ)料庫(kù)，結(jié)果很不好，輸出都是0。然后我就用了7000條新聞作為語(yǔ)料庫(kù)，分詞完之后得到的fenci_result.txt是20M，雖然也不大，但是已經(jīng)可以得到初步結(jié)果了。

三、使用gensim的word2vec訓(xùn)練模型

相關(guān)代碼如下：

from gensim.modelsimport word2vec
import logging
 
# 主程序
logging.basicConfig(format='%(asctime)s:%(levelname)s: %(message)s', level=logging.INFO)
sentences =word2vec.Text8Corpus(u"fenci_result.txt") # 加載語(yǔ)料
model =word2vec.Word2Vec(sentences, size=200) #訓(xùn)練skip-gram模型，默認(rèn)window=5
 
print model
# 計(jì)算兩個(gè)詞的相似度/相關(guān)程度
try:
  y1 = model.similarity(u"國(guó)家", u"國(guó)務(wù)院")
except KeyError:
  y1 = 0
print u"【國(guó)家】和【國(guó)務(wù)院】的相似度為：", y1
print"-----\n"
#
# 計(jì)算某個(gè)詞的相關(guān)詞列表
y2 = model.most_similar(u"控?zé)?quot;, topn=20) # 20個(gè)最相關(guān)的
print u"和【控?zé)煛孔钕嚓P(guān)的詞有：\n"
for item in y2:
  print item[0], item[1]
print"-----\n"
 
# 尋找對(duì)應(yīng)關(guān)系
print u"書-不錯(cuò)，質(zhì)量-"
y3 =model.most_similar([u'質(zhì)量', u'不錯(cuò)'], [u'書'], topn=3)
for item in y3:
  print item[0], item[1]
print"----\n"
 
# 尋找不合群的詞
y4 =model.doesnt_match(u"書 書籍 教材 很".split())
print u"不合群的詞：", y4
print"-----\n"
 
# 保存模型，以便重用
model.save(u"書評(píng).model")
# 對(duì)應(yīng)的加載方式
# model_2 =word2vec.Word2Vec.load("text8.model")
 
# 以一種c語(yǔ)言可以解析的形式存儲(chǔ)詞向量
#model.save_word2vec_format(u"書評(píng).model.bin", binary=True)
# 對(duì)應(yīng)的加載方式
# model_3 =word2vec.Word2Vec.load_word2vec_format("text8.model.bin",binary=True)

輸出如下：

"D:\program files\python2.7.0\python.exe" "D:/pycharm workspace/畢設(shè)/cluster_test/word2vec.py"
D:\program files\python2.7.0\lib\site-packages\gensim\utils.py:840: UserWarning: detected Windows; aliasing chunkize to chunkize_serial
 warnings.warn("detected Windows; aliasing chunkize to chunkize_serial")
D:\program files\python2.7.0\lib\site-packages\gensim\utils.py:1015: UserWarning: Pattern library is not installed, lemmatization won't be available.
 warnings.warn("Pattern library is not installed, lemmatization won't be available.")
2016-12-12 15:37:43,331: INFO: collecting all words and their counts
2016-12-12 15:37:43,332: INFO: PROGRESS: at sentence #0, processed 0 words, keeping 0 word types
2016-12-12 15:37:45,236: INFO: collected 99865 word types from a corpus of 3561156 raw words and 357 sentences
2016-12-12 15:37:45,236: INFO: Loading a fresh vocabulary
2016-12-12 15:37:45,413: INFO: min_count=5 retains 29982 unique words (30% of original 99865, drops 69883)
2016-12-12 15:37:45,413: INFO: min_count=5 leaves 3444018 word corpus (96% of original 3561156, drops 117138)
2016-12-12 15:37:45,602: INFO: deleting the raw counts dictionary of 99865 items
2016-12-12 15:37:45,615: INFO: sample=0.001 downsamples 29 most-common words
2016-12-12 15:37:45,615: INFO: downsampling leaves estimated 2804247 word corpus (81.4% of prior 3444018)
2016-12-12 15:37:45,615: INFO: estimated required memory for 29982 words and 200 dimensions: 62962200 bytes
2016-12-12 15:37:45,746: INFO: resetting layer weights
2016-12-12 15:37:46,782: INFO: training model with 3 workers on 29982 vocabulary and 200 features, using sg=0 hs=0 sample=0.001 negative=5 window=5
2016-12-12 15:37:46,782: INFO: expecting 357 sentences, matching count from corpus used for vocabulary survey
2016-12-12 15:37:47,818: INFO: PROGRESS: at 1.96% examples, 267531 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:37:48,844: INFO: PROGRESS: at 3.70% examples, 254229 words/s, in_qsize 3, out_qsize 1
2016-12-12 15:37:49,871: INFO: PROGRESS: at 5.99% examples, 273509 words/s, in_qsize 3, out_qsize 1
2016-12-12 15:37:50,867: INFO: PROGRESS: at 8.18% examples, 281557 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:37:51,872: INFO: PROGRESS: at 10.20% examples, 280918 words/s, in_qsize 5, out_qsize 0
2016-12-12 15:37:52,898: INFO: PROGRESS: at 12.44% examples, 284750 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:37:53,911: INFO: PROGRESS: at 14.17% examples, 278948 words/s, in_qsize 0, out_qsize 0
2016-12-12 15:37:54,956: INFO: PROGRESS: at 16.47% examples, 284101 words/s, in_qsize 2, out_qsize 1
2016-12-12 15:37:55,934: INFO: PROGRESS: at 18.60% examples, 285781 words/s, in_qsize 6, out_qsize 1
2016-12-12 15:37:56,933: INFO: PROGRESS: at 20.84% examples, 288045 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:37:57,973: INFO: PROGRESS: at 23.03% examples, 289083 words/s, in_qsize 6, out_qsize 2
2016-12-12 15:37:58,993: INFO: PROGRESS: at 24.87% examples, 285990 words/s, in_qsize 6, out_qsize 1
2016-12-12 15:38:00,006: INFO: PROGRESS: at 27.17% examples, 288266 words/s, in_qsize 4, out_qsize 1
2016-12-12 15:38:01,081: INFO: PROGRESS: at 29.52% examples, 290197 words/s, in_qsize 1, out_qsize 2
2016-12-12 15:38:02,065: INFO: PROGRESS: at 31.88% examples, 292344 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:38:03,188: INFO: PROGRESS: at 34.01% examples, 291356 words/s, in_qsize 2, out_qsize 2
2016-12-12 15:38:04,161: INFO: PROGRESS: at 36.02% examples, 290805 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:38:05,174: INFO: PROGRESS: at 38.26% examples, 292174 words/s, in_qsize 3, out_qsize 0
2016-12-12 15:38:06,214: INFO: PROGRESS: at 40.56% examples, 293297 words/s, in_qsize 4, out_qsize 1
2016-12-12 15:38:07,201: INFO: PROGRESS: at 42.69% examples, 293428 words/s, in_qsize 4, out_qsize 1
2016-12-12 15:38:08,266: INFO: PROGRESS: at 44.65% examples, 292108 words/s, in_qsize 1, out_qsize 1
2016-12-12 15:38:09,295: INFO: PROGRESS: at 46.83% examples, 292097 words/s, in_qsize 4, out_qsize 1
2016-12-12 15:38:10,315: INFO: PROGRESS: at 49.13% examples, 292968 words/s, in_qsize 2, out_qsize 2
2016-12-12 15:38:11,326: INFO: PROGRESS: at 51.37% examples, 293621 words/s, in_qsize 5, out_qsize 0
2016-12-12 15:38:12,367: INFO: PROGRESS: at 53.39% examples, 292777 words/s, in_qsize 2, out_qsize 2
2016-12-12 15:38:13,348: INFO: PROGRESS: at 55.35% examples, 292187 words/s, in_qsize 5, out_qsize 0
2016-12-12 15:38:14,349: INFO: PROGRESS: at 57.31% examples, 291656 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:38:15,374: INFO: PROGRESS: at 59.50% examples, 292019 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:38:16,403: INFO: PROGRESS: at 61.68% examples, 292318 words/s, in_qsize 4, out_qsize 2
2016-12-12 15:38:17,401: INFO: PROGRESS: at 63.81% examples, 292275 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:38:18,410: INFO: PROGRESS: at 65.71% examples, 291495 words/s, in_qsize 4, out_qsize 1
2016-12-12 15:38:19,433: INFO: PROGRESS: at 67.62% examples, 290443 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:38:20,473: INFO: PROGRESS: at 69.58% examples, 289655 words/s, in_qsize 6, out_qsize 2
2016-12-12 15:38:21,589: INFO: PROGRESS: at 71.71% examples, 289388 words/s, in_qsize 2, out_qsize 2
2016-12-12 15:38:22,533: INFO: PROGRESS: at 73.78% examples, 289366 words/s, in_qsize 0, out_qsize 1
2016-12-12 15:38:23,611: INFO: PROGRESS: at 75.46% examples, 287542 words/s, in_qsize 5, out_qsize 1
2016-12-12 15:38:24,614: INFO: PROGRESS: at 77.25% examples, 286609 words/s, in_qsize 3, out_qsize 0
2016-12-12 15:38:25,609: INFO: PROGRESS: at 79.33% examples, 286732 words/s, in_qsize 5, out_qsize 1
2016-12-12 15:38:26,621: INFO: PROGRESS: at 81.40% examples, 286595 words/s, in_qsize 2, out_qsize 0
2016-12-12 15:38:27,625: INFO: PROGRESS: at 83.53% examples, 286807 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:38:28,683: INFO: PROGRESS: at 85.32% examples, 285651 words/s, in_qsize 5, out_qsize 3
2016-12-12 15:38:29,729: INFO: PROGRESS: at 87.56% examples, 286175 words/s, in_qsize 6, out_qsize 1
2016-12-12 15:38:30,706: INFO: PROGRESS: at 89.86% examples, 286920 words/s, in_qsize 5, out_qsize 0
2016-12-12 15:38:31,714: INFO: PROGRESS: at 92.10% examples, 287368 words/s, in_qsize 6, out_qsize 0
2016-12-12 15:38:32,756: INFO: PROGRESS: at 94.40% examples, 288070 words/s, in_qsize 4, out_qsize 2
2016-12-12 15:38:33,755: INFO: PROGRESS: at 96.30% examples, 287543 words/s, in_qsize 1, out_qsize 0
2016-12-12 15:38:34,802: INFO: PROGRESS: at 98.71% examples, 288375 words/s, in_qsize 4, out_qsize 0
2016-12-12 15:38:35,286: INFO: worker thread finished; awaiting finish of 2 more threads
2016-12-12 15:38:35,286: INFO: worker thread finished; awaiting finish of 1 more threads
Word2Vec(vocab=29982, size=200, alpha=0.025)
【國(guó)家】和【國(guó)務(wù)院】的相似度為： 0.387535493256
-----
2016-12-12 15:38:35,293: INFO: worker thread finished; awaiting finish of 0 more threads
2016-12-12 15:38:35,293: INFO: training on 17805780 raw words (14021191 effective words) took 48.5s, 289037 effective words/s
2016-12-12 15:38:35,293: INFO: precomputing L2-norms of word weight vectors
和【控?zé)煛孔钕嚓P(guān)的詞有：
禁煙 0.6038454175
防煙 0.585186183453
執(zhí)行 0.530897378922
煙控 0.516572892666
廣而告之 0.508533298969
履約 0.507428050041
執(zhí)法 0.494115233421
禁煙令 0.471616715193
修法 0.465247869492
該項(xiàng) 0.457907706499
落實(shí) 0.457776963711
控制 0.455987215042
這方面 0.450040221214
立法 0.44820779562
控?zé)熮k 0.436062157154
執(zhí)行力 0.432559013367
控?zé)煏?huì) 0.430508673191
進(jìn)展 0.430286765099
監(jiān)管 0.429748386145
懲罰 0.429243773222
-----
書-不錯(cuò)，質(zhì)量-
生存 0.613928854465
穩(wěn)定 0.595371186733
整體 0.592055797577
----
不合群的詞： 很
-----
2016-12-12 15:38:35,515: INFO: saving Word2Vec object under 書評(píng).model, separately None
2016-12-12 15:38:35,515: INFO: not storing attribute syn0norm
2016-12-12 15:38:35,515: INFO: not storing attribute cum_table
2016-12-12 15:38:36,490: INFO: saved 書評(píng).model
Process finished with exit code 0

python如何實(shí)現(xiàn)word2vec