CUDArray: CUDA-based NumPy
Department of Applied Mathematics and Computer Science, Technical University of Denmark
Technical University of Denmark, DTU Compute Technical Report-2014-21, 2014
@article{larsen2014cudarray,
title={CUDArray: CUDA-based NumPy},
author={Larsen, Anders Boesen Lindbo},
year={2014}
}
This technical report introduces CUDArray – a CUDA-accelerated subset of the NumPy library. The goal of CUDArray is to combine the ease of development from NumPy with the computational power of Nvidia GPUs in a lightweight and extensible framework. Since the motivation behind CUDArray is to facilitate neural network programming, CUDArray extends NumPy with a neural network submodule. This module has both a CPU and a GPU back-end to allow for experiments without requiring a GPU.
November 16, 2014 by hgpu
