... impact on inference performance (e.g. Jetson TX2 FP32 6 cl/s using TensorFlow and 13 cl/s using TensorRT). 12.6. Outlook.
確定! 回上一頁