Algorithm 1 Training the Classification Model Based on DCGAN-Based Classifier

Input: 1: Load the dataset ISIC2017_Training_Data, S17; 2. Split the dataset: Training 70% and Testing 30% 3: Preprocessing of S17: interp2(), histeq(), imsharpen(), imfilter(), rgb2lab(), gaussian_median_filter() 4: Initialize the networks: Generator G(latent_noise), Discriminator D (); 5. Create optimizers to update the weights using backpropagation sgdmupdate () and learning rate(); 6. Train the networks G with noise and D with real and G-generated images for a number of epochs.     $mi{n}_{G}ma{x}_{D}V\left(G,D\right)=E_{x~Pdata\left(x\right)}\left[\log \left(x\right)\right]+E_{x~Pz\left(z\right)}\left[\log \left(1-D\left(G\left(z\right)\right)\right)\right]$ 7. ReLu and Tanh activation function for G; Leaky ReLu and SoftMax for D; 8. Calculate the loss function and repeat 5–7.           $D\_loss=D\_real\_loss+D\_fake\_loss$ 9. D acts as Classifier N + 1 output Output: 1: N + 1 Output 2: Confusion Matrix of Classification; Plots of AUC_ROC. 3: return Accuracy, Recall, Precision, Specificity and F1_Score.