Exploiting Multiple Optimizers with Transfer Learning Techniques for the Identification of COVID-19 Patients

Figure 1

Block diagram of our study. Stage 1: data preprocessing: the number of channels of different images was made alike and data augmentation was performed. Stage 2: setting up training parameters: training parameters like number of epochs, mini-batch size, and number of folds per epoch opted in this stage. Data training was performed on each model after image resizing according to the distinct input size. Stage 3: data classification: here, our trained model displays the classified result as either normal or infected.