Model Version 1.2

Metrics

• train_loss: 0.5705
• train_accuracy: 0.7812
• val_loss: 0.5975
• val_accuracy: 0.7752

Configuration

• architecture: Tensorflow CNN
• input_size: (224, 224, 3)
• num_classes: 6
• augmentation: {'rotation_range': 20, 'width_shift_range': 0.2, 'height_shift_range': 0.2, 'horizontal_flip': True, 'vertical_flip': True, 'shear_range': 0.2, 'brightness_range': [0.8, 1.2]}
• optimizer: adam
• learning_rate: 0.001
• batch_size: 32
• epochs: 50