AlviGeo/cifar10-resnet18-finetuned

CIFAR-10 Image Classification - ResNet-18 Fine-tuned

Performa Model

• Test Accuracy: 95.01%
• Architecture: ResNet-18 (pretrained on ImageNet)
• Dataset: CIFAR-10 (60,000 images, 10 classes)

Kelas

airplane, automobile, bird, cat, deer, dog, frog, horse, ship, truck

Training Details

• Optimizer: Adam (lr=1e-3, weight_decay=1e-4)
• Scheduler: StepLR (step=5, gamma=0.5)
• Epochs: 5
• Augmentasi: RandomHorizontalFlip, RandomCrop, ColorJitter