Popular repositories Loading

1. yolov12-vs-yolov13-attention-study yolov12-vs-yolov13-attention-study Public

   Python 2

2. yolov13-validation-study yolov13-validation-study Public

   Independent validation study comparing YOLOv13 vs YOLOv12/YOLO11 on COCO dataset

   1

3. first-repository first-repository Public

   Jupyter Notebook

4. yolov12-sdpa-flashattention-pytorch yolov12-sdpa-flashattention-pytorch Public

   PyTorch implementation of YOLOv12 with Scaled Dot-Product Attention (SDPA) optimized by FlashAttention for fast and efficient object detection.

   HTML

5. pytorch-sdpa-vs-flash-attention pytorch-sdpa-vs-flash-attention Public

   First scientific comparison proving PyTorch SDPA as viable Flash Attention alternative. 96.7% mAP, 6% speed diff, 100% portable. No CUDA compilation needed.

   Python

6. Scientific-README-Why-SDPA-Achieves-Flash-Attention-Performance Scientific-README-Why-SDPA-Achieves-Flash-Attention-Performance Public