Repository navigation
Tags
Tags
v0.12 (MaartenGr#668) * Online/incremental topic modeling with .partial_fit * Expose c-TF-IDF model for customization with bertopic.vectorizers.ClassTfidfTransformer * Expose attributes for easier access to internal data * Major changes to the Algorithm page of the documentation, which now contains three overviews of the algorithm * Added an example of combining BERTopic with KeyBERT * Added many tests with the intention of making development a bit more stable * Fix MaartenGr#632, MaartenGr#648, MaartenGr#673, MaartenGr#682, MaartenGr#667, MaartenGr#664
v0.11.0 (MaartenGr#578) * Perform hierarchical topic modeling with `.hierarchical_topics` * Visualize hierarchical topic representations with `.visualize_hierarchy` * Extract a text-based hierarchical topic representation with `.get_topic_tree` * Visualize 2D documents with `.visualize_documents()` * Visualize 2D hierarchical documents with `.visualize_hierarchical_documents()` * Create custom labels to the topics throughout most visualizations with `.generate_topic_labels` and `.set_topic_labels` * Manually merge topics with `.merge_topics()` * Added example for finding similar topics between two models in the tips & tricks page * Add multi-modal example in the tips & tricks page * Added native Hugging Face transformers support
v0.10.0 (MaartenGr#492) * Use any dimensionality reduction technique instead of UMAP * Use any clustering technique instead of HDBSCAN * Add a CountVectorizer page with tips and tricks on how to create topic representations that fit your use case * Added pages on how to use other dimensionality reduction and clustering algorithms * Additional instructions on how to reduce outliers in the FAQ * Fixed `None` being returned for probabilities when transforming unseen documents * Replaced all instances of `arg:` with `Arguments:` for consistency * Before saving a fitted BERTopic instance, we remove the stopwords in the fitted CountVectorizer model as it can get quite large due to the number of words that end in stopwords if `min_df` is set to a value larger than 1 * Set `"hdbscan>=0.8.28"` to prevent numpy issues * Update gensim dependency to `>=4.0.0` (MaartenGr#371) * Fix topic 0 not appearing in visualizations (MaartenGr#472) * Fix MaartenGr#506 * Fix MaartenGr#429
v0.9.4 (MaartenGr#335) * Expose diversity parameter * Improve stability of topic reduction * Added property to c-TF-IDF that all IDF values are positive (MaartenGr#351) * Improve stability of `.visualize_barchart()` and `.visualize_hierarchy()` * Major documentation overhaul (including MaartenGr#330) * Drop python 3.6 (MaartenGr#333) * Relax plotly dependency (MaartenGr#88) * Additional logging for `.transform` (MaartenGr#356)
v0.9.3 - Quickfix (MaartenGr#284) * Fix MaartenGr#282, MaartenGr#285, MaartenGr#288
v0.9.2 (MaartenGr#239) * Update default embedding model from 'paraphrase' to 'all' * Fix probability mapping * Optimize cTFIDF topic extraction * Fix algorithm image, update documentation, fix spelling, etc. * Fix MaartenGr#258 * Update README with visualization example
v0.9.1 (MaartenGr#211) Fix MaartenGr#210, MaartenGr#208, MaartenGr#205, and MaartenGr#213
v0.9 (MaartenGr#176) * Get most representative documents per topic: `topic_model.get_representative_docs(topic=1)` * Added `normalize_frequency` parameter to `visualize_topics_per_class` and `visualize_topics_over_time` * Return flat probabilities as default, only calculate the probabilities of all topics per document if `calculate_probabilities` is True * Implemented a guided BERTopic by defining seed topics: `BERTopic(seed_topic_list=seed_topic_list)` * Fix loading embedding model * Fix probability mapping * Improve accuracy of probabilities * Additional FAQs
v0.8.1 (MaartenGr#138) * Fix tests * Add Kaggle example * Add interactive visualizations to API documentation * Set transformers in Flair
v0.8 (MaartenGr#120) Visualization Update: * Topic Hierarchy: topic_model.visualize_hierarchy() * Topic Similarity Heatmap: topic_model.visualize_heatmap() * Topic Representation Barchart: topic_model.visualize_barchart() * Term Score Decline: topic_model.visualize_term_rank() Improvements: * Created bertopic.plotting library to easily extend visualizations * Improved automatic topic reduction by using HDBSCAN to detect similar topics * Sort topic ids by their frequency. -1 is the outlier class and contains typically the most documents. After that 0 is the largest topic, 1 the second largest, etc. * Update MKDOCS with new visualizations Fixes: * Fix typo MaartenGr#113, MaartenGr#117 * Fix MaartenGr#121 * Fix mapping of topics after reduction (it now excludes 0) (MaartenGr#103)
PreviousNext