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
