Aegis ChequeGuard: Swin Transformer Edition πŸ›‘οΈ

An advanced Vision Transformer (Swin-Tiny) fine-tuned for bank cheque authentication (Real vs. Fake/Photocopy).

πŸš€ Highlights

  • Architecture: microsoft/swin-tiny-patch4-window7-224
  • Validation Accuracy: 100.00%
  • Final Validation Loss: 0.0005
  • Formats: PyTorch (safetensors) and ONNX (.onnx)

πŸ“‚ Contents

  • model.safetensors and config.json: PyTorch weights and configs
  • preprocessor_config.json: Preprocessing specs (224x224)
  • swin_cheque_verification.onnx: Single-file production ONNX model
Downloads last month
-
Safetensors
Model size
27.5M params
Tensor type
I64
Β·
F32
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support