ResNet-50 ImageNet-1k Classifier

Upload an image to classify it into one of 1000 ImageNet categories.

This model is a ResNet-50 trained on the ImageNet-1k dataset with modern optimization techniques:

• Architecture: ResNet-50 with Bottleneck blocks [3, 4, 6, 3]
• Parameters: ~25.6M trainable parameters
• Training Optimizations:
  • Progressive resizing (128→160→192→224px)
  • CutMix and MixUp augmentation
  • Label smoothing (0.1)
  • Exponential Moving Average (EMA)
  • Automatic Mixed Precision (AMP)
  • PyTorch 2.0 compilation
  • FFCV high-performance data loading
• Target Accuracy: 78%+ (Top-1), 94%+ (Top-5)
• Training Time: ~90 minutes on 8x A100 GPUs

Class labels are from the official HuggingFace ImageNet-1k dataset.

The model works best with natural images containing objects, animals, or scenes from the ImageNet categories.

Training code: github.com/arghyaiitb/assignment_9