Classification of Citrus Diseases Using Optimization Deep Learning Approach

Algorithm 1

SGDM optimizer.

(1)	Select W0 as the initial parameter vector and as the objective function
(2)	Select the decay rate moving average γ and step size η
(3)	Create the first moment M0 to zero vector
(4)	while true do
(5)	W0 is not converge do
(6)	j = j + 1
(7)	at timestep, j, calculate the gradients
(8)	Using the updated biased first-moment estimate Mj = γ Mj−1 + (1 − γ)
(9)	Use to change the parameters Wj = Wj−1 − η Mj
(10)	end