pytorch-extension

This is an example of a CUDA extension/function/layer for PyTorch which uses CuPy to compute the Hadamard product of two tensors.

For a more advanced extension that uses CuPy as well, please see: https://github.com/sniklaus/softmax-splatting
And yet another extension that uses CuPy can be found here: https://github.com/sniklaus/revisiting-sepconv

setup

Make sure to install CuPy, which can be done using pip install cupy or alternatively using one of the provided binary packages as outlined in the CuPy repository.

usage

There is no separate build process necessary, simply run python run.py to test it. A minimal example of how the sample extension can be used is also shown below.

import torch

import hadamard # the custom layer

class Network(torch.nn.Module):
	def __init__(self):
		super().__init__()
	# end

	def forward(self, tenOne, tenTwo):
		return hadamard.hadamard_func.apply(tenOne, tenTwo)
	# end
# end

netNetwork = Network().cuda()

tenOne = torch.rand(size=[64, 3, 128, 128], dtype=torch.float32, device=torch.device('cuda')).requires_grad_()
tenTwo = torch.rand(size=[64, 3, 128, 128], dtype=torch.float32, device=torch.device('cuda')).requires_grad_()

tenOut = netNetwork(tenOne, tenTwo)
tenExpected = torch.mul(tenOne, tenTwo)

print(torch.sum(tenOut.data - tenExpected.data), '<-- should be 0.0')

license

Please refer to the appropriate file within this repository.