Algorithm 2: Fitness Evaluation of ETLBOCBL-CNN
Inputs: $X_n.Pos$, ${ℛ}^{train}$, ${ℛ}^{valid}$, $S^{batch}$, ${\epsilon }^{train}$, $R^L$, $C^{num}$
01:	Construct a candidate CNN architecture based on the network and learning hyperparameters decoded from $X_n.Pos$ and insert a fully connected layer with $C^{num}$ output neurons;
02:	Compute ${\tau }^{train}$ and ${\tau }^{valid}$ using Equations (4) and (6), respectively;
03:	Generate the initial weights of the CNN model as $\mathit{\varpi }=\left\{{\varpi }_1,{\varpi }_2,\dots \right\}$ using the selected weight initializer;
04:	for $\epsilon =1$ to ${\epsilon }^{train}$ do
05:	for $i=1$ to ${\tau }^{train}$ do
06:	Calculate $f\left(\mathit{\varpi },{ℛ}_i^{train} \right)$ of CNN model;
07:	Update the weights ${\mathit{\varpi }}^{\mathit{n}\mathit{e}\mathit{w}}=\left\{{\varpi }_1^{new},{\varpi }_2^{new},\dots \right\}$ based on Equation (5);
08:	end for
09:	end for
10:	for $j=1$ to ${\tau }^{valid}$ do
11:	Classify the ${ℛ}_j^{valid}$ dataset using the trained CNN model;
12:	Record the classification errors for solving the ${ℛ}_j^{valid}$ dataset as $Err\_Batc{h}_j$;
13:	end for
14:	Calculate $X_n,Err$ of the candidate CNN architecture built from $X_n.Pos$ with Equation (7);
Output:  $X_n.Err$