CODEWITHSUNDEEP
pythongensimlda
Con
Con

Reputation: 89

Python Gensim LDA Model show_topics funciton

I am training a LDA Model using Gensim:

dictionary = corpora.Dictionary(section_2_sentence_df['Tokenized_Sentence'].tolist())
dictionary.filter_extremes(no_below=20, no_above=0.7)
corpus = [dictionary.doc2bow(text) for text in (section_2_sentence_df['Tokenized_Sentence'].tolist())]

num_topics = 15
passes = 200
chunksize = 100
lda_sentence_model = gensim.models.ldamulticore.LdaMulticore(corpus, num_topics=num_topics, 
                                                              id2word=dictionary, 
                                                              passes=passes, 
                                                              chunksize=chunksize,
                                                              random_state=100,
                                                              workers = 3)

After training i need the topics for further analysis. Unfortunately the show_topics function only returns 10 topics. I expected the defined number of 15 topics. Does anyone know if that is on purpose or an error that can be solved?

print(len(lda_sentence_model.show_topics(formatted=False)))

Upvotes: 0

Views: 2563

Answers (1)

gojomo
gojomo

Reputation: 54243

According to the gensim documentation for the .show_topics() method, its default num_topics parameter value ("Number of topics to be returned") is 10:

https://radimrehurek.com/gensim/models/ldamulticore.html#gensim.models.ldamulticore.LdaMulticore.show_topics

If you want it to return more than 10, supply your preferred non-default value for that method's num_topics parameter. For example:

len(lda_sentence_model.show_topics(formatted=False, num_topics=15))

Upvotes: 1

Related Questions