skip to main content
Idioma:
Tipo de recurso Mostra resultados com: Mostra resultados com: Índice

Kernel Operations on the GPU, with Autodiff, without Memory Overflows

Charlier, Benjamin ; Feydy, Jean ; Glaunès, Joan Alexis ; François-David, Collin ; Durif, Ghislain

Journal of machine learning research, 2021-01, Vol.22 (74), p.1-6 [Periódico revisado por pares]

Ithaca: Cornell University Library, arXiv.org

Texto completo disponível

Citações Citado por
  • Título:
    Kernel Operations on the GPU, with Autodiff, without Memory Overflows
  • Autor: Charlier, Benjamin ; Feydy, Jean ; Glaunès, Joan Alexis ; François-David, Collin ; Durif, Ghislain
  • Assuntos: Computation ; Computer Science - Learning ; Graphics processing units ; High level languages ; Kernels ; Libraries ; Machine Learning ; Mathematical analysis ; Matrix methods ; Statistics ; Tensors
  • É parte de: Journal of machine learning research, 2021-01, Vol.22 (74), p.1-6
  • Descrição: The KeOps library provides a fast and memory-efficient GPU support for tensors whose entries are given by a mathematical formula, such as kernel and distance matrices. KeOps alleviates the major bottleneck of tensor-centric libraries for kernel and geometric applications: memory consumption. It also supports automatic differentiation and outperforms standard GPU baselines, including PyTorch CUDA tensors or the Halide and TVM libraries. KeOps combines optimized C++/CUDA schemes with binders for high-level languages: Python (Numpy and PyTorch), Matlab and GNU R. As a result, high-level "quadratic" codes can now scale up to large data sets with millions of samples processed in seconds. KeOps brings graphics-like performances for kernel methods and is freely available on standard repositories (PyPi, CRAN). To showcase its versatility, we provide tutorials in a wide range of settings online at \url{www.kernel-operations.io}.
  • Editor: Ithaca: Cornell University Library, arXiv.org
  • Idioma: Inglês
Links
Voltar para lista de resultados

  • Realização: ABCD-USP USP   Logos de Redes Sociais: Freepik

Buscando em bases de dados remotas. Favor aguardar.