A versatile Python package engineered for seamless topic modeling, topic evaluation, and topic visualization. Ideal for text analysis, natural language processing (NLP), and research in the social sciences, STREAM simplifies the extraction, interpretation, and visualization of topics from large, complex datasets.
Currently we encode all words for the embedding based metrics
This takes very long and is fairly inefficient