Package index
-
backend_config() - Build embedding backend configuration
-
backend_embed_texts() - Embed texts via configured backend
-
backend_info() - Get embedding backend model/service information
-
backend_read() - Read backend configuration from YAML
-
backend_save() - Save backend configuration to YAML
-
batch_collect_openai() - Collect completed OpenAI batch embedding jobs
-
batch_status_openai() - Inspect OpenAI batch state for a label
-
batch_submit_openai() - Submit OpenAI Batch jobs for corpus embeddings (asynchronous)
-
calibrate_threshold() - Calibrate threshold from Parquet scores by streaming batches
-
clean_abstract_for_embedding() - Clean title/abstract rows into embedding-ready text
-
demo_finalize_openai_batch() - Finalize OpenAI demo batch jobs and compare direct vs batch embeddings
-
distance_cosine() - Cosine distance between two numeric vectors
-
distance_reference_cosine() - Pairwise cosine distances with centroid axis between label partitions
-
distance_ridge() - Compute corpus distance to a reference embedding area
-
distances() - Join prototype and ridge distances lazily via Arrow
-
embed_corpus() - Stream a corpus dataset, embed in batches, and write Parquets
-
embed_texts() - Embed texts through a configured backend
-
fit_ridge() - Fit a reference-area model from embeddings parquet
-
plot_embeddings_pca() - Plot embeddings via PCA, colored by arbitrary labels
-
plot_embeddings_umap() - Plot embeddings via UMAP, colored by arbitrary labels
-
run_demo_openai() - Create and optionally run an OpenAI-based demo project via Quarto
-
run_demo_openalex() - Create and optionally run a self-contained demo project via Quarto
-
score_reference_cosine() - Convert reference-cosine distances to scores
-
score_ridge() - Convert ridge distances to ridge scores
-
similarity_cosine() - Cosine similarity between two numeric vectors