I would like to know your approaches for selecting the best hyperparameters for DNN models such as:

• Number of epochs
• Choice of optimizer
• Varying the regularization method
• Choice of activation function
• Number of layers and number of neurons in each layer
• Learning rate decay methods
• Mini-batch size
• Weight initialization methods

I need to specially know about tuning "Number of layers and number of neurons in each layer"

You can suggest papers, books or any other reliable reference.