model,weights,mAP50,mAP50-95,precision,recall,size_mb
PyTorch FP32,runs/detect/runs/sar/yolov8n_sar2/weights/best.pt,0.9168,0.6646,0.9183,0.84,6.26
TorchScript FP16,runs/detect/runs/sar/yolov8n_sar2/weights/best.torchscript,0.9124,0.6271,0.9278,0.8274,12.43