dog-qiuqiu / Yolo-Fastest

:zap: Based on yolo's ultra-lightweight universal target detection algorithm, the calculation amount is only 250mflops, the ncnn model size is only 666kb, the Raspberry Pi 3b can run up to 15fps+, and the mobile terminal can run up to 178fps+
Other
1.98k stars 427 forks source link

8 bit problem #10

Open chuck0518 opened 3 years ago

chuck0518 commented 3 years ago

hey,man,thanks for the work,it's cool. When i convert the model to 8bit by ncnn tool,the accuracy has dropped a lot,can you give me some suggetions?

dog-qiuqiu commented 3 years ago

int8 ncnn is not well optimized,int8 does not bring inference speed advantages

chuck0518 commented 3 years ago

got it,thank you