Building LDA models with a different number of topics

The number of topics is an exogenous parameter. We will build several models with a range of topics. We will use the BoW data and the TF-IDF data. To economize our code, we will develop a small function, build_lda_models(). This function takes three arguments, as follows:

• input_data: Either BoW or TF-IDF
• name: The filename to save the model
• k: The number of topics

The code is shown as follows:

def build_lda_models(input_data, name, k):    np.random.seed(42)
    # Train the model
    lda = LdaModel(input_data,
              num_topics=k,
              id2word = gensim_dictionary)
    # Save the model
    from gensim.test.utils import datapath
    tempfile...