2D convolution in pytorch

Can we define a 2d convolution inpytorch in the following way:

pred_bb = nn.conv2d(x_BB,4, (net_shape[1], net_shape[2]))

x_BB is the output of previous convolutional layers
4 is the number of output channels
and, (net_shape[1], net_shape[2]) is the shape of the kernel

Thanks and regards

Our doc has a pretty detailed description and examples on how to use nn.Conv2d
http://pytorch.org/docs/0.2.0/nn.html#torch.nn.Conv2d