nihui
nihui
yolov5n v6.1
https://zhuanlan.zhihu.com/p/471357671
the data is collected from https://github.com/Tencent/ncnn/issues/3485 https://qengineering.eu/install-ncnn-on-jetson-nano.html Note, during the test the Jetson Nano CPU was overclocked to 2014.5 MHz and the GPU to 998.4 MHz.
> > 你的pb我看了一下，没有不支持的op，你这个h5 netron都打不开，一般来说肯定是转换出问题了。 > > 我明天帮你转一下试试看吧，tf这个一堆模型转换的问题是真的头疼= = > > 感谢大佬！tf 想在 ncnn下跑真的是太难了（就是钟情于ncnn, 莫得办法） 你这个错误我也遇到了，于是只能导出 mlir，然后 mlir2ncnn ...
binaryop pow only handles positive base, this is a known limitation. try unaryop square or binaryop mul for x^2 purpose
``` nihui@nanopik2-s905:~/ncnn/build/benchmark$ ./benchncnn 4 4 0 -1 1 loop_count = 4 num_threads = 4 powersave = 0 gpu_device = -1 cooling_down = 1 squeezenet min = 55.85 max = 56.29...
``` nihui@odroid:~/ncnn/build/benchmark$ ./benchncnn 4 4 2 -1 1 loop_count = 4 num_threads = 4 powersave = 2 gpu_device = -1 cooling_down = 1 squeezenet min = 41.74 max = 42.33...
``` rock@rock-5b:~/ncnn/build/benchmark$ ./benchncnn 4 4 2 -1 0 loop_count = 4 num_threads = 4 powersave = 2 gpu_device = -1 cooling_down = 0 squeezenet min = 4.21 max = 4.29...
``` nihui@nanopim3:~/ncnn/build/benchmark$ ./benchncnn 4 8 0 -1 1 loop_count = 4 num_threads = 8 powersave = 0 gpu_device = -1 cooling_down = 1 squeezenet min = 46.99 max = 48.23...
``` nihui@bananapim2zero:~/ncnn/build/benchmark$ ./benchncnn 4 4 0 -1 1 loop_count = 4 num_threads = 4 powersave = 0 gpu_device = -1 cooling_down = 1 squeezenet min = 218.73 max = 219.06...