ddangelov / Top2Vec

Top2Vec learns jointly embedded topic, document and word vectors.
BSD 3-Clause "New" or "Revised" License
2.95k stars 374 forks source link

Clarification on Reusing a pretrained saved top2vec model #356

Open amruthakannan opened 9 months ago

amruthakannan commented 9 months ago

Can I reuse the same pretrained model on a different dataset but of the same domain?Say I trained a top2vec model on a dataset with 26k records on artificial intelligence. Now can I reuse this same model to get the topics on a smaller dataset on artificial intelligence by calling get_document_topics()?

I am asking this because, when I looked through your source code, it seems like the pretrained model has already saved each document id, and then it matches the document id saved in the model with the new dataset. I would appreciate some clarification! thanks