JieShare

关于负样本代码labels = np.zeros((0,5), dtype=np.float32)

2

是不是写错了，应该是labels = np.zeros((1,5), dtype=np.float32)？代表负样本的标签为0，ground truth的坐标和高宽都为0？

Excessive inference time spent on attention_mask operations after ONNX→TensorRT conversion (DriveOS AGX Orin + TRT 8.6.15)

2

--- # Hi NVIDIA team, I encountered a severe performance issue when converting a GPT-like model (with `attention_mask` input) from ONNX to TensorRT. After successful conversion and deployment on **DriveOS...