Does torch implement 3D DeConvolution (Upsampling)? Found that the documentation of SpatialFullConvolution says, Other frameworks call this operation "In-network Upsampling", "Fractionally-strided convolution", "Backwards Convolution," "Deconvolution", or "Upconvolution.".
So I guess that VolumetricFullConvolution could do 3D deconvolution, but not sure, can anyone tell me?
Does torch implement 3D DeConvolution (Upsampling)? Found that the documentation of SpatialFullConvolution says, Other frameworks call this operation "In-network Upsampling", "Fractionally-strided convolution", "Backwards Convolution," "Deconvolution", or "Upconvolution.". So I guess that VolumetricFullConvolution could do 3D deconvolution, but not sure, can anyone tell me?
Best Regards, Kui