NN_conv2d()

void NN_conv2d	(	Tensor *	out,
		const Tensor *	in,
		const Tensor *	weight,
		const Tensor *	bias,
		const size_t *	stride,
		const size_t *	padding,
		const size_t *	dilation,
		size_t	groups
	)

Applies a 2D convolution over an input signal composed of several input planes.

Parameters

out	the output tensor of shape (batch_size, channels_in, height, width)
in	the input tensor of shape (batch_size, channels_out, height, width)
weight	the learnable weights of the module of shape (kernel_height, kernel_width, channels_in, channels_out)
bias	the learnable bias of the module of shape (channels_out), or NULL if no bias is applied
stride	stride for the cross-correlation
padding	the amount of padding applied to the input
dilation	the spacing between kernel elements
groups	number of blocked connections from input channels to output channels