# Backpropagate error and store in neurons
def backward_propagate_error(network, expected):
 for i in reversed(range(len(network))):
  layer = network[i]
  errors = list()
  # calculate the loss forr each layer
  if i != len(network)-1:
   ...
        # ∂L/∂a of the hidden layer  
    for neuron in network[i + 1]: 
     error += (neuron['weights'][j] * neuron['delta']) 
    errors.append(error)
  else: 
     ...