登录论坛

查看完整版本 : 编写内核opn Matlab以进行GPU并行化的值函数迭代


poster
2019-11-28, 19:01
我想编写一个在GPU上执行Value Function Iteration的matlab函数。我的想法与用Julia写的代码非常相似:
https://github.com/giob1994/GPU-OpenCL-VFI-in-Julia/blob/master/GPU%20Value%20Function%20Iteration%20under%20Uncertainty.ipynb

我真的很难为GPU编写内核。在Julia-Code中,操作如下所示:

# Write kernel for GPU manually:
gpu_call(grid, (grid, V, policy, z, P, Float32(alpha), Float32(beta), Float32(delta), Float32(sigma), UInt32(SIZE_GRID), UInt32(SIZE_Z)))
do state, grid, V, policy, z, P, alpha, beta, delta, sigma, SIZE_GRID, SIZE_Z
# Each kernel executes for one value of the capital grid:
idx = @linearidx grid

Matlab中gpu_call( )和__ = @linearidx __的等效函数是什么?

我已经安装了并行计算工具包。

感谢您的任何帮助,提示或建议! :)



更多&回答... (https://stackoverflow.com/q/59087018)