nabu.processing.kernel_base module

Base class for CudaKernel and OpenCLKernel Should not be used directly

class nabu.processing.kernel_base.KernelBase(kernel_name, filename=None, src=None, automation_params=None)

Bases: object

A base class for OpenCL and Cuda kernels.

Parameters:

• kernel_name (str) – Name of the CUDA kernel.

• filename (str, optional) – Path to the file name containing kernels definitions

• src (str, optional) – Source code of kernels definitions

• automation_params (dict, optional) – Automation parameters, see below

• parameters (Automation) –

• ---------------------- –

• values. (automation_params is a dictionary with the following keys and default) –

  guess_block: bool (True)

  If block is not specified during calls, choose a block size based on the size/dimensions of the first array. Mind that it is unlikely to be the optimal choice.

  guess_grid: bool (True):

  If the grid size is not specified during calls, choose a grid size based on the size of the first array.

  follow_device_ptr: bool (True)

  specify gpuarray.gpudata for all cuda GPUArrays (and pyopencl.array.data for pyopencl arrays). Otherwise, raise an error.

check_filename_src(filename, src)

set_automation_params(automation_params)

static guess_grid_size(shape, block_size)

static guess_block_size(shape)

Guess a block size based on the shape of an array.

get_block_grid(*args, **kwargs)

follow_device_arr(args)