CPU-Only + Torch V100 + Torch. V100 + TensorRT. Laten cy (ms). Im ages/sec. Inference throughput (sentences/sec) on OpenNMT 692M. V100 + TensorRT: NVIDIA ...
確定! 回上一頁