使用 Python subprocess.CalledProcessError 进行主题建模答案

【问题标题】：Topic modelling with Python subprocess.CalledProcessError使用 Python subprocess.CalledProcessError 进行主题建模
【发布时间】：2021-03-29 11:16:45
【问题描述】：

我需要为我的 uni 项目创建一个主题建模，而我正在做的是试图重复这个人正在做的事情：https://www.youtube.com/watch?v=TgXLq1XIdA0

到目前为止，我一直在运行他的代码，直到第 45 行，我在这里复制它：

#creating our first topic model :)
#importing the necessary libraries
import gensim, nltk, os

#defining lists to work on with
#tokens from each file stored here
texts = []
#files' names
labels = []

#iterating through all files in the corpus folder
for root, dirs, files in os.walk('CEOs_speeches_all'):
    for file_name in files:
        with open(os.path.join(root, file_name),encoding='utf8') as rf:
            #opening each file as a text for reading
            text = rf.read()
            #tokenising words with nltk
            tokens = nltk.word_tokenize(text)
            #getting rid of all tokens that are not numerical or alphabetical i.e. spaces, punctuation
            cleaned = [word for word in tokens if word.isalnum()]
            texts.append(cleaned)
            #putting files' names without extension (.txt) into the list
            labels.append(file_name[:-4])

#transforming the corpus to run in gensim
corpus_dictionary = gensim.corpora.Dictionary(texts)
corpus_dictionary.filter_extremes(no_below=5)
#transforming the corpus text list into the list of bags of words
processed_corpus = [corpus_dictionary.doc2bow(text) for text in texts]

#printing the first file for checking
#print(processed_corpus[0])

#defining the number of topics
number_of_topics = 10

#specifying where mallet lives
mallet_path = os.path.join("C:\mallet", "bin", "mallet")

#creating the mallet modeling object
lda_model = gensim.models.wrappers.ldamallet.LdaMallet(
    mallet_path,
    corpus=processed_corpus,
    id2word=corpus_dictionary,
    num_topics=number_of_topics,
    optimize_interval=10,
    prefix='fed_'
)

我收到以下错误消息： subprocess.CalledProcessError: Command 'C:\mallet/bin/mallet import-file --preserve-case --keep-sequence --remove-stopwords --token-regex "\S+" --input fed_corpus.txt --output fed_corpus.mallet' 返回非零退出状态 127。

有人可以帮忙解决吗？有什么想法吗？

更新：我将 mallet_path = os.path.join("C:\mallet", "bin", "mallet") 行更改为 mallet_path = "~/mallet/bin/mallet" 但它仍然显示相同错误：/

【问题讨论】：

标签： python gensim topic-modeling

【解决方案1】：

您似乎在 Linux 环境中工作，但示例中的路径是为 Windows 编写的。您需要使用计算机上的正确路径更新变量mallet_path。

【讨论】：