Function reference • BertopicR

Library Setup

Functions for setting up BertopicR

check_python_dependencies(): Check that dependencies are loaded

install_python_dependencies(): Install Python Dependencies

Embedding Documents

Functions for creating and manipulating document embeddings.

bt_make_embedder_st(): Create an embedding model using sentence-transformers

bt_make_embedder_spacy(): Create an embedding model using a model available from the Spacy Library

bt_make_embedder_flair(): Create an embedding model using a model available from the Flair Library

bt_make_embedder_openai(): Create an embedding model using a model available from the OpenAI Library

bt_do_embedding(): Embed your documents

bt_make_reducer_umap(): Create umap dimensionality reduction model

bt_make_reducer_pca(): Create pca dimensionality reduction model

bt_make_reducer_truncated_svd(): Created Truncated SVD dimensionality reduction model

bt_do_reducing(): Perform dimensionality reduction on your embeddings

Clustering

Functions for clustering embedded documents

bt_make_clusterer_hdbscan(): Create an HDBSCAN clustering model

bt_make_clusterer_kmeans(): Create a kmeans clustering model

bt_make_clusterer_agglomerative(): Create an Agglomerative Clustering clustering model

bt_do_clustering(): Cluster your data

Tokenise

Functions for tokenising document words/phrases in order to represent topics

bt_make_vectoriser(): Create a text vectoriser

bt_make_ctfidf(): Create an instance of the ClassTfidfTransformer from the bertopic.vectorizers module

Build Model

Functions for compiling and buiding model

bt_compile_model(): Build a BERTopic model

bt_fit_model(): Fit a topic model on your documents & embeddings

bt_update_topics(): Update Topic Representations

Representation

Functions for clustering embedded documents

bt_representation_mmr(): Create representation model using Maximal Marginal Relevance

bt_representation_keybert(): Create representation model using keybert

bt_representation_openai(): Create representation model that uses OpenAI text generation models

bt_representation_hf(): Use Huggingface models to create topic representation

Topic Manipulation

Functions for merging topics and reducing outliers

bt_merge_topics(): Merges list(s) of topics together

bt_outliers_embeddings(): Redistributes outliers using embeddings

bt_outliers_tokenset_similarity(): Redistributes outliers using tokenset c-TF-IDF scores

bt_outliers_ctfidf(): Redistributes outliers using c-TF-IDF scores

Empty Models

Functions for creating empty instances of models

bt_empty_clusterer(): Create an empty clusterer for skipping clustering step of bertopic pipeline

bt_empty_embedder(): Create an empty embedder for skipping embedding step of bertopic pipeline

bt_empty_reducer(): Create an empty reducer for skipping dimensionality reduction step of bertopic pipeline

Detaching Bertopic

Functions for detaching python bertopic library

bertopic_detach(): Detach bertopic from the python session