Package: fastEmbedR
Type: Package
Title: Native UMAP and Interpolation-Based t-SNE Embeddings
Version: 0.99.17
Authors@R: person(
    given = "Stefano",
    family = "Cacciatore",
    role = c("aut", "cre"),
    email = "tkcaccia@gmail.com",
    comment = c(ORCID = "0000-0001-7052-7156")
    )
Copyright: See inst/COPYRIGHTS and inst/NOTICE for source-level
        ownership, adapted components, and optional linked
        dependencies.
Description: A native float32 UMAP implementation and a configurable
        interpolation-based t-SNE implementation for embeddings from
        data matrices or precomputed nearest-neighbor indices and
        distances. One-call workflows use package-native CPU HNSW and
        Apple Metal exact or IVF-Flat search. Optional CUDA support
        links directly to FAISS GPU and the RAPIDS cuVS C API while
        retaining nearest-neighbor results on the device.
        Preprocessing, graph or affinity construction, fixed-reference
        transform, and layout optimization use native code with
        explicit CPU, Metal, and CUDA backend reporting. UMAP exposes
        neighborhood, metric, graph mode, backend, seed, and
        preprocessing controls, while optimizer controls such as
        epochs, minimum distance, spread, learning rate, repulsion, and
        negative sampling follow a recorded release policy rather than
        a general-purpose sweep API. Optional graph utilities support
        downstream Louvain, Leiden, and Walktrap community detection.
License: MIT + file LICENSE
URL: https://github.com/tkcaccia/fastEmbedR,
        https://tkcaccia.github.io/fastEmbedR/,
        https://github.com/tkcaccia/fastEmbedR-benchmark
BugReports: https://github.com/tkcaccia/fastEmbedR/issues
biocViews: Software, Infrastructure, Preprocessing, DimensionReduction,
        Visualization, Clustering, SingleCell, FlowCytometry,
        GeneExpression, GPU
Encoding: UTF-8
Imports: Biobase, methods, Rcpp, withr
Suggests: float, igraph, jsonlite, knitr, RhpcBLASctl, rmarkdown,
        testthat
LinkingTo: Rcpp
SystemRequirements: C++17; optional Metal backend: Apple Silicon with
        macOS 14 or later and Xcode 15 or later; the required
        Foundation, Accelerate, Metal, Metal Performance Shaders, and
        Metal Performance Shaders Graph frameworks are supplied by the
        Apple SDK; optional CUDA backend: CUDA Toolkit with cuFFT,
        cuBLAS, cuSOLVER, and CUB/Thrust (CCCL); FAISS GPU and RAPIDS
        cuVS for CUDA nearest-neighbor search; RAPIDS RAFT and RMM only
        when CUDA TSVD initialization is enabled.
VignetteBuilder: knitr
Roxygen: list(markdown = TRUE)
Config/roxygen2/version: 8.1.0
Config/testthat/edition: 3
Config/pak/sysreqs: nvidia-cuda-dev libfaiss-dev
Repository: https://biocstaging.r-universe.dev
Date/Publication: 2026-08-30 12:59:37 UTC
RemoteUrl: https://github.com/BiocStaging/fastEmbedR
RemoteRef: HEAD
RemoteSha: 918492c82ebe4804299b8713b5315b0fcd6f47d2
NeedsCompilation: yes
Packaged: 2026-09-30 11:01:36 UTC; root
Author: Stefano Cacciatore [aut, cre] (ORCID:
    <https://orcid.org/0000-0001-7052-7156>)
Maintainer: Stefano Cacciatore <tkcaccia@gmail.com>