python - numpy向量运算的核心使用上限

Question

我喜欢 numpy，因为向量运算会自动使用多核。

但是，我最近注意到，尽管我的机器有 8 个核心，但只使用了 4 个核心。

为什么 numpy 不使用机器中的所有核心？是否可以让numpy在numpy的向量运算上使用更多的核心？

我使用的是 Mac OSX 10.8。

更新

np.show_config() 显示如下:

lapack_opt_info:
    libraries = ['mkl_lapack95_lp64', 'mkl_intel_lp64', 'mkl_intel_thread', 'mkl_core', 'iomp5', 'pthread']
    library_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/lib']
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    include_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/include']
blas_opt_info:
    libraries = ['mkl_intel_lp64', 'mkl_intel_thread', 'mkl_core', 'iomp5', 'pthread']
    library_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/lib']
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    include_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/include']
openblas_lapack_info:
  NOT AVAILABLE
lapack_mkl_info:
    libraries = ['mkl_lapack95_lp64', 'mkl_intel_lp64', 'mkl_intel_thread', 'mkl_core', 'iomp5', 'pthread']
    library_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/lib']
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    include_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/include']
blas_mkl_info:
    libraries = ['mkl_intel_lp64', 'mkl_intel_thread', 'mkl_core', 'iomp5', 'pthread']
    library_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/lib']
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    include_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/include']
mkl_info:
    libraries = ['mkl_intel_lp64', 'mkl_intel_thread', 'mkl_core', 'iomp5', 'pthread']
    library_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/lib']
    define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]
    include_dirs = ['/Users/john/.pyenv/versions/anaconda-2.0.1/include']

Answer 1

MKL，理想情况下已经为您的系统使用优化的线程数。您可以这样检查:

python -c "import ctypes; mkl_rt = ctypes.CDLL('mkl_rt.dll'); print mkl_rt.mkl_get_max_threads()"

如果您需要更改它，您可能需要更改 MKL_NUM_THREADS 环境变量:

python -c "import ctypes; mkl_rt = ctypes.CDLL('mkl_rt.dll'); print mkl_rt.mkl_get_max_threads()"

有关 MKL 多线程检查的更多信息 this ，以及 API .

我引用Anton Malakhov ，来自英特尔。您可能想检查该问题(配置 MKL NumPy 可用的处理器数量)，因为它与您的问题非常相似。