In the convolutional neural network below, the numbers inside the rectangles represent the layer number and the numbers at the bottom of the rectangles represent the number of units in the layers (e.g., the input layer has 146 imes 146 = 21,316 input units.) Conv indicates a convolutional layer; Pool indicates a pooling layer and FC indicates a fully-connected layer. Assume that each convolutional layer in the network has only one filter of size 7 imes 7.