CuPy: A NumPy-compatible array library accelerated by CUDA
Project description
CuPy : NumPy & SciPy for GPU
CuPy is a NumPy/SciPy-compatible array library for GPU-accelerated computing with Python.
This package (cupy) is a source distribution. For most users, use of pre-build wheel distributions are recommended:
cupy-cuda113 (for CUDA 11.3)
cupy-cuda112 (for CUDA 11.2)
cupy-cuda111 (for CUDA 11.1)
cupy-cuda110 (for CUDA 11.0)
cupy-cuda102 (for CUDA 10.2)
cupy-cuda101 (for CUDA 10.1)
cupy-cuda100 (for CUDA 10.0)
cupy-cuda92 (for CUDA 9.2)
cupy-cuda90 (for CUDA 9.0)
cupy-rocm-4-2 (for ROCm 4.2)
cupy-rocm-4-0 (for ROCm 4.0)
Please see Installation Guide for the detailed instructions.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
cupy-9.2.0.tar.gz
(1.6 MB
view hashes)