Python 为什么dgemm和sgemm比numpy慢得多（200x）'；s点？

为什么dgemm和sgemm比numpy的dot慢得多（200倍）？这是正常的吗

以下是我用来测试的代码：

from scipy.linalg import blas
import numpy as np
import time


x2 = np.zeros((1000000, 512))
x1 = np.zeros((1, 512))

t1 = time.time()
for i in range(10):
    np.dot(x1, x2.T)
t2 = time.time()
print("np.dot: ", t2-t1)
t1 = time.time()
for i in range(10):
    blas.dgemm(alpha=1.0, a=x1, b=x2, trans_b=True)
t2 = time.time()
print("dgemm: ", t2-t1)
t1 = time.time()
for i in range(10):
    blas.sgemm(alpha=1.0, a=x1, b=x2, trans_b=True)
t2 = time.time()
print("sgemm: ", t2-t1)

我得到的结果是：

np.dot:  0.1820526123046875
dgemm:  34.11782765388489
sgemm:  25.33052659034729

以下是我的scipy配置，显示它是使用OpenBLAS编译的：

   >>> import scipy
    >>> scipy.__config__.show()
    openblas_lapack_info:
        libraries = ['openblas', 'openblas']
        library_dirs = ['/usr/local/lib']
        language = c
        define_macros = [('HAVE_CBLAS', None)]
    lapack_opt_info:
        libraries = ['openblas', 'openblas']
        library_dirs = ['/usr/local/lib']
        language = c
        define_macros = [('HAVE_CBLAS', None)]
    blas_mkl_info:
      NOT AVAILABLE
    openblas_info:
        libraries = ['openblas', 'openblas']
        library_dirs = ['/usr/local/lib']
        language = c
        define_macros = [('HAVE_CBLAS', None)]
    blas_opt_info:
        libraries = ['openblas', 'openblas']
        library_dirs = ['/usr/local/lib']
        language = c
        define_macros = [('HAVE_CBLAS', None)]

以下是我的numpy配置，它与scipy的基本相同：

>>> import numpy
>>> numpy.__config__.show()
blas_mkl_info:
  NOT AVAILABLE
blis_info:
  NOT AVAILABLE
openblas_info:
    libraries = ['openblas', 'openblas']
    library_dirs = ['/usr/local/lib']
    language = c
    define_macros = [('HAVE_CBLAS', None)]
blas_opt_info:
    libraries = ['openblas', 'openblas']
    library_dirs = ['/usr/local/lib']
    language = c
    define_macros = [('HAVE_CBLAS', None)]
lapack_mkl_info:
  NOT AVAILABLE
openblas_lapack_info:
    libraries = ['openblas', 'openblas']
    library_dirs = ['/usr/local/lib']
    language = c
    define_macros = [('HAVE_CBLAS', None)]
lapack_opt_info:
    libraries = ['openblas', 'openblas']
    library_dirs = ['/usr/local/lib']
    language = c
    define_macros = [('HAVE_CBLAS', None)]

---

我用错了吗？

我认为从f2py调用blas函数的开销是罪魁祸首。首先，使您的数组Fortran连续，否则将创建一个数组副本以传递给GEMM，同时使您的数据类型不为dgemm而浮动int@percusse为什么要把它设为int？我使用np.zero作为虚拟变量。我在处理浮点矩阵。我的意思不是int。如果我将

order='F'

添加到零函数中，dgemm就开始赢了。sgemm无论如何都会慢一些，因为数据应该在内部转换为单精度。@percusse我找不到numpy.dot的源代码。它是否在幕后使用相同的scipy.blas.gemm函数？速度非常接近，这让我怀疑区别在于函数调用的开销。