There is a dataset including images of dogs, cats, birds, fish, and humans. In addition, some images in the dataset do not belong to any of the categories mentioned above.
Build a deep neural network based on all the images in the dataset (all images must be used) to recognize the category of the image, whether it is a dog, cat, bird, fish, or human.
• Plot the architecture of the deep neural network. The network must include at least 3 convolutional layers and 2 fully connected layers (excluding the output layer).
• The dimensions of the features (or activation shape) in each layer must be indicated.
• Provide clear descriptions of the filters used in each layer. Include information such as the number of filters, filter size, stride, padding (if any), and padding size.
• All images have dimensions of 32x64x3 pixels (row x column x depth).