When merging the Topic_model (including all data, with outliers) and the Out_Topic_model (consisting only of outliers), the 'Count' of the Topic_model for -1 increases by the number of outliers, instead of effectively concat them.

I have a hard time understanding what you exactly mean here. Could you give an example? Perhaps showcase what is happening and what you would expect to happen?

The Representative_docs are displayed as NaN.
=> is the only way?

The representative documents are indeed displayed as NaN since merge_models is also meant for federated learning. If you want representative documents re-calculated, I would advise checking the issues page. I believe there are a number of issues that describe in detail how you can do this.

Hi @MaartenGr and @OnAnd0n , let me share my code as an example here, as I also try to extract outliers from a topic model, calculate a new topic model using the outliers only and then merge both models. What I tried so far:

# lets assume there is a topic model called topic_model with roughly 50% outliers (topic -1)
# first, lets filter the outliers from the initial topic model
filtered_docs = [doc for doc, topic in zip(docs, topics) if topic != -1] 

# new topic model with outlier docs only
outlier_topic_model = BERTopic()
outlier_topics, outlier_probs = outlier_topic_model.fit_transform(filtered_docs)

# merge these two models together
merged_model = BERTopic.merge_models([topic_model, outlier_topic_model])

Is this a possible approach? And why do I not get e.g. the topic frequencies for the merged model? This: merged_model.get_topic_freq() does not work?

Is this a possible approach? And why do I not get e.g. the topic frequencies for the merged model? This: merged_model.get_topic_freq() does not work?

Could you share a bit more information? Which version of BERTopic are you using? Can you share the output? You mention that it does not work, but what exactly does that mean? Do you get an error or perhaps unexpected results?