Anaconda Accelerate

Accelerate provides access to numerical libraries optimized for performance on Intel CPUs and NVidia GPUs.

Features

  • Bindings to CUDA libraries: cuBLAS, cuFFT, cuSPARSE, cuRAND, and sorting algorithms from the CUB and Modern GPU libraries
  • Speed-boosted linear algebra operations in NumPy, SciPy, scikit-learn and NumExpr libraries using Intel’s Math Kernel Library (MKL).
  • Accelerated variants of Numpy’s built-in UFuncs.
  • Increased-speed Fast Fourier Transformations (FFT) in NumPy.

Requirements

  • 64-bit operating system: Linux, OS X or Windows

  • Supported Python and Numpy combinations:
    • Python 2.7 with Numpy 1.9 or 1.10
    • Python 3.4 with Numpy 1.9 or 1.10
    • Python 3.5 with Numpy 1.9 or 1.10
  • Numba 0.25

For the CUDA features:

  • NVidia driver version 349.00 or later
  • CUDA toolkit 7.0
  • At least one CUDA GPU with compute capability 2.0 or above

Installation

Accelerate is included with Anaconda Workgroup and Anaconda Enterprise subscriptions.

To start a 30-day free trial just download and install the Anaconda Accelerate package.

If you already have Anaconda (free Python distribution) installed:

conda update conda
conda install accelerate

If you do not have Anaconda installed, you can download it here.

Anaconda Accelerate can also be installed into your own (non-Anaconda) Python environment. For more information about Accelerate please contact sales@continuum.io.

Update Instructions

If you have Anaconda (free Python distribution) installed:

conda update conda
conda update accelerate

If you already have NumbaPro installed, you must manually upgrade NumbaPro to install the NumbaPro compatibility layer:

conda update conda
conda update numbapro