Mastering Computer Vision with TensorFlow 2.x

Front module

The following code describes the front module:

def create_front_module(input, num_channels, bottleneck):    _x = Conv2D(64, kernel_size=(7, 7), strides=(2, 2), padding='same',         activation='relu', name='front_conv_1x1_x1')(input)    _x = BatchNormalization()(_x)    _x = bottleneck(_x, num_channels // 2, 'front_residual_x1')    _x = MaxPool2D(pool_size=(2, 2), strides=(2, 2))(_x)    _x = bottleneck(_x, num_channels // 2, 'front_residual_x2')    _x = bottleneck(_x, num_channels, 'front_residual_x3')    return _xfront_features = create_front_module(input, num_channels, bottleneck)

As discussed previously, this consists of a Conv2D block with a total of 64 filters with a filter size of 7 x 7 and a stride of 2. The output of the block is (None, 32, 32,6) ...