762078742de4aad07627f1add50b5c638a40f18b,skcuda/cublas.py,,,#,5404

Before Change

                                    int(C), ldc)
    cublasCheckStatus(status)

cublasDgeam.__doc__ = _GEAM_doc.substitute(precision="double precision",
                                           real="real",
                                           num_type="numpy.float64",
                                           alpha_data="np.float64(np.random.rand())",
                                           beta_data="np.float64(np.random.rand())",
                                           a_data_1="np.random.rand(2, 3).astype(np.float64)",
                                           b_data_1="np.random.rand(2, 3).astype(np.float64)",
                                           a_data_2="np.random.rand(2, 3).astype(np.float64)",
                                           b_data_2="np.random.rand(3, 2).astype(np.float64)",
                                           c_data_1="alpha*a+beta*b",
                                           c_data_2="alpha*a.T+beta*b",
                                           func="cublasDgeam")

if _cublas_version >= 5000:
    _libcublas.cublasCgeam.restype = int
    _libcublas.cublasCgeam.argtypes = [_types.handle,

After Change

////// BLAS-like extension routines //////

// SGEAM, DGEAM, CGEAM, ZGEAM
_GEAM_doc = Template(

    Matrix-matrix addition/transposition (${precision} ${real}).

    Computes the sum of two ${precision} ${real} scaled and possibly (conjugate)
    transposed matrices.

    Parameters
    ----------
    handle : int
        CUBLAS context
    transa, transb : char
        "t" if they are transposed, "c" if they are conjugate transposed,
        "n" if otherwise.
    m : int
        Number of rows in `A` and `C`.
    n : int
        Number of columns in `B` and `C`.
    alpha : ${num_type}
        Constant by which to scale `A`.
    A : ctypes.c_void_p
        Pointer to first matrix operand (`A`).
    lda : int
        Leading dimension of `A`.
    beta : ${num_type}
        Constant by which to scale `B`.
    B : ctypes.c_void_p
        Pointer to second matrix operand (`B`).
    ldb : int
        Leading dimension of `A`.
    C : ctypes.c_void_p
        Pointer to result matrix (`C`).
    ldc : int
        Leading dimension of `C`.

    Examples
    --------
    >>> import pycuda.autoinit
    >>> import pycuda.gpuarray as gpuarray
    >>> import numpy as np
    >>> alpha = ${alpha_data}
    >>> beta = ${beta_data}
    >>> a = ${a_data_1}
    >>> b = ${b_data_1}
    >>> c = ${c_data_1}
    >>> a_gpu = gpuarray.to_gpu(a)
    >>> b_gpu = gpuarray.to_gpu(b)
    >>> c_gpu = gpuarray.empty(c.shape, c.dtype)
    >>> h = cublasCreate()
    >>> ${func}(h, "n", "n", c.shape[0], c.shape[1], alpha, a_gpu.gpudata, a.shape[0], beta, b_gpu.gpudata, b.shape[0], c_gpu.gpudata, c.shape[0])
    >>> np.allclose(c_gpu.get(), c)
    True
    >>> a = ${a_data_2}
    >>> b = ${b_data_2}
    >>> c = ${c_data_2}
    >>> a_gpu = gpuarray.to_gpu(a.T.copy())
    >>> b_gpu = gpuarray.to_gpu(b.T.copy())
    >>> c_gpu = gpuarray.empty(c.T.shape, c.dtype)
    >>> transa = "c" if np.iscomplexobj(a) else "t"
    >>> ${func}(h, transa, "n", c.shape[0], c.shape[1], alpha, a_gpu.gpudata, a.shape[0], beta, b_gpu.gpudata, b.shape[0], c_gpu.gpudata, c.shape[0])
    >>> np.allclose(c_gpu.get().T, c)
    True
    >>> cublasDestroy(h)

    References
    ----------
    `cublas<t>geam <http://docs.nvidia.com/cuda/cublas///cublas-lt-t-gt-geam>`_
)

if _cublas_version >= 5000:
    _libcublas.cublasSgeam.restype = int
    _libcublas.cublasSgeam.argtypes = [_types.handle,

Instances