`mkl_random` -- a NumPy-based Python interface to Intel (R) MKL Random Number Generation functionality

mkl_random has started as Intel (R) Distribution for Python optimizations for NumPy.

Per NumPy's community suggestions, voiced in https://github.com/numpy/numpy/pull/8209, it is being released as a stand-alone package.

Prebuilt mkl_random can be installed into conda environment from Intel's channel on Anaconda cloud:

  conda install -c intel mkl_random

To install mkl_random Pypi package please use following command:

   python -m pip install --i https://pypi.anaconda.org/intel/simple -extra-index-url https://pypi.org/simple mkl_random

If command above installs NumPy package from the Pypi, please use following command to install Intel optimized NumPy wheel package from Anaconda Cloud:

   python -m pip install --i https://pypi.anaconda.org/intel/simple -extra-index-url https://pypi.org/simple mkl_random numpy==<numpy_version>

Where <numpy_version> should be the latest version from https://anaconda.org/intel/numpy

mkl_random is not fixed-seed backward compatible drop-in replacement for numpy.random, meaning that it implements sampling from the same distributions as numpy.random.

For distributions directly supported in Intel (R) Math Kernel Library (MKL), method keyword is supported:

   mkl_random.standard_normal(size=(10**5, 10**3), method='BoxMuller')

Additionally, mkl_random exposes different basic random number generation algorithms available in MKL. For example to use SFMT19937 use

   mkl_random.RandomState(77777, brng='SFMT19937')

For generator families, such that MT2203 and Wichmann-Hill, a particular member of the family can be chosen by specifying brng=('WH', 3), etc.

See MKL reference guide for more details: https://software.intel.com/en-us/mkl-developer-reference-c-random-number-generators

The list of supported by mkl_random.RandomState constructor brng keywords is as follows:

'MT19937'
'SFMT19937'
'WH' or ('WH', id)
'MT2203' or ('MT2203', id)
'MCG31'
'R250'
'MRG32K3A'
'MCG59'
'PHILOX4X32X10'
'NONDETERM'
'ARS5'

Filters