nihui
nihui
https://github.com/Tencent/ncnn/pull/4359
fixed in https://github.com/Tencent/ncnn/commit/6647396667426405cee4da8da79f723148f60dd5
``` nihui@jetson-nano:~/ncnn/build/benchmark$ ./benchncnn 10 1 0 -1 0 loop_count = 10 num_threads = 1 powersave = 0 gpu_device = -1 cooling_down = 0 squeezenet min = 77.27 max = 77.46...
``` nihui@starfive:~/ncnn/build/benchmark$ ./benchncnn 4 4 0 -1 0 loop_count = 4 num_threads = 4 powersave = 0 gpu_device = -1 cooling_down = 0 squeezenet min = 150.07 max = 150.45...
rpi b+ ``` pi@raspberrypi:~/ncnn/build/benchmark $ ./benchncnn 4 1 0 -1 1 syscall error -1 loop_count = 4 num_threads = 1 powersave = 0 gpu_device = -1 cooling_down = 1 squeezenet...
``` root@BeagleBone:/home/nihui/ncnn/build/benchmark# ./benchncnn 4 2 0 -1 1 loop_count = 4 num_threads = 2 powersave = 0 gpu_device = -1 cooling_down = 1 squeezenet min = 42.23 max = 42.54...
``` root@panther-x2:/home/nihui/ncnn/build/benchmark# ./benchncnn 8 4 0 -1 0 loop_count = 8 num_threads = 4 powersave = 0 gpu_device = -1 cooling_down = 0 squeezenet min = 25.91 max = 27.53...
The data in the benchmark readme is likely out of date The new version optimizes the inference performance of int8 Welcome to update and test again. If there are any...
move to https://github.com/Tencent/ncnn/pull/4463
``` net.opt.openmp_blocktime = 0; ``` before load_param 有助于减少发热