w is the convolution filter weight tensor, organized as tensor w[m, c, k, k], where m is the number of output feature maps, c is the number of input feature maps and k is the height and width of each filter. tensors are stored as multi-dimensional arrays in the memory. x is the input feature map, organized as a tensor x[c, h, w], where h is the height of each input feature map and w is the width of each input feature map. y is the output feature map, organized as a tensor y[m, h out, w out], where h out is the height of each output feature map and w out is the width of each output feature map. what are h out and w out in terms of m,c,k,h,w?