Research Article
Multiobjective Genetic Algorithm and Convolutional Neural Network Based COVID-19 Identification in Chest X-Ray Images
Algorithm 3
Hyperparameter tuning of CNN using multiobjective genetic algorithm.
| | output: optimized population | | | input: {Fitness function, demand, crossover_ratio} | | | begin | | (1) | Generate the random ; / represents the initial population /; | | (2) | Calculate the fitness of ; | | (3) | Sort ; | | (4) | / Selection / | | (5) | set = ; / denotes final population / | | (6) | while ordo | | (7) | / and represent children elimination and last generation / | | (8) | Generate random ; / denotes children / | | (9) set ; | | (10) | for eachdo | | (11) | Compute the fitness of ; | | (12) | ifthen | | (13) | remove ; | | (14) | set ; | | (15) | else | | (16) | set = ; | | (17) | end | | (18) | end | | (19) | / Mutation / | | (20) | for crossover do | | (21) | select and randomly; / , , and are children / | | (22) | ; | | (23) | Evaluate the fitness of ; | | (24) | ifthen | | (25) | remove ; | | (26) | else | | (27) | remove ; | | (28) | end | | (29) | end | | (30) | next generation | | (31) | end | | (32) | / Ranking / sort the ; | | (33) | return / returns the most dominant solution w.r.t. fitness function / | | | end |
|