Embedding analysis: DBSCAN clustering and UMAP 2D projection. Operates on raw float[] vectors; not coupled to any text-specific code.
Embedding analysis: DBSCAN clustering and UMAP 2D projection. Operates on raw float[] vectors; not coupled to any text-specific code.
(cluster embeddings & {:keys [radius min-pts] :or {radius 0.75 min-pts 1}})Run DBSCAN on a seq of float[] embeddings. Returns a vector of cluster indices (0-based); -1 means unclustered/noise. Options: :radius - neighborhood radius in cosine distance (default 0.75) :min-pts - minimum neighbors for a core point (default 1)
Run DBSCAN on a seq of float[] embeddings. Returns a vector of cluster indices (0-based); -1 means unclustered/noise. Options: :radius - neighborhood radius in cosine distance (default 0.75) :min-pts - minimum neighbors for a core point (default 1)
(project-2d embeddings)Reduce embeddings to 2D positions via UMAP. Takes a seq of float[] embeddings; returns a vector of [x y] pairs.
Reduce embeddings to 2D positions via UMAP. Takes a seq of float[] embeddings; returns a vector of [x y] pairs.
cljdoc builds & hosts documentation for Clojure/Script libraries
| Ctrl+k | Jump to recent docs |
| ← | Move to previous article |
| → | Move to next article |
| Ctrl+/ | Jump to the search field |