5. Phi and Theta Extraction. Transform Method

Detailed description of all parameters and methods of BigARTM Python API classes can be found in Python Interface.

  • Phi/Theta exraction

Let’s assume, that you have a data and a model, fitted on this data. You had tuned all necessary regularizers and used scores. But the set of quality measures of the library wasn’t enough for you, and you need to compute your own scores using \Phi and \Theta matrices. In this case you are able to extract these matrices using next code:

phi = model.get_phi()
theta = model.get_theta()

Note, that you need a cache_theta flag to be set True if you are planning to extract \Theta in the future without using transform(). You also can extract not whole matrices, but part of them, that corresponds different topics (using the same topic_names parameter of the methods, as in previous sections). Also you can extract only necessary modalities of the \Phi matrix, if you want.

Both methods return pandas.DataFrame.

  • Transform new documents

Now we will go to the usage of fitted ready model for the classification of test data (to do it you need to have a fitted multimodal model with @labels_class modality, see 4. Multimodal Topic Models).

In the classification task you have the train data (the collection you used to train your model, where for each document the model knew it’s true class labels), and test one. For the test data true labels are known to you, but are unknown to the model. Model need to forecast these labels, using test documents, and your task is to compute the quality of the predictions by counting some metrics, AUC, for instance.

Computation of the AUC or any other quality measure is your task, we won’t do it. Instead, we will learn how to get p(c|d) vectors for each document, where each value is the probability of class c in the given document d.

Well, we have a model. We assume you put test documents into separate file in Vowpal Wabbit format, and created batches using it, which are covered by the variable batch_vectorizer_test. Also we assume you have saved your test batches into the separate directory (not into the one containing train batches).

Your test documents shouldn’t contain information about true labels (e.g. the Vowpal Wabbit file shouldn’t contain string ‘|@labels_class’), also text document shouldn’t contain tokens, that doesn’t appear in the train set. Such tokens will be ignored.

If all these conditions are met, we can use the ARTM.transform() method, that allows you to get p(t|d) (e.g. \Theta) or p(c|d) matrix for all documents from your BatchVectorizer object.

Run this code to get \Theta:

theta_test = model.transform(batch_vectorizer=batch_vectorizer_test)

And this one to achieve p(c|d):

p_cd_test = model.transform(batch_vectorizer=batch_vectorizer_test,

In this way you have got the predictions of the model in pandas.DataFrame. Now you can score the quality of the predictions of your model in all ways, you need.

Method allows you to extract dense od sparse matrix. Also you can use for p_{tdw} matrix (see 8. Deal with Ptdw Matrix).

  • Held-out perplexity calculation

To compute perplexity for held-out sample you should use the following code sample

# create score if you haven't do so yet

# apply `transform` to process the held-out sample;
# don't create theta matrix to save memory
model.transform(batch_vectorizer=your_heldout_sample, theta_matrix_type=None)

# retrieve score with get_score(score_name)
perplexity = model.get_score('perplexity')

If the code above looks like magic, remember important facts about scores implementation, described in the 3. Regularizers and Scores Usage