Welcome to the Elema team's codebase. This codebase is about CIFAR100 image classification
Here's the revised introduction:
CIFAR-100 Classification
The CIFAR-100 dataset consists of 60,000 32x32 color images in 100 classes, with 600 images per class. It is split into 50,000 training and 10,000 test images. CIFAR-100 includes two levels of labels:
- Superclasses: 20 high-level categories grouping the classes.
- Fine classes: 100 specific classes.
For CIFAR-100 classification, we used ResNet and Wide-ResNet models. Performance metrics often focus on Top-1 and Top-5 Accuracy, and sometimes Super-Class Accuracy for broader category classification. This dataset is widely used to benchmark image recognition algorithms due to its small image size and challenging classification task with many similar classes.
Here are the latest results from our team
| Random Seed | Test Acc | Top-5 Acc | Super-Class Acc |
|---|---|---|---|
| 4 | 79.05% | 95.33% | 87.21% |
| 8 | 79.44% | 95.25% | 87.28% |
| 12 | 79.14% | 95.27% | 87.74% |