It also supports 8-bit quantized models and provides GPU performance on par with their float versions. For details on the GPU delegate, see TensorFlow Lite ...
確定! 回上一頁