Hello,
I am developing apps on Galaxy S10+ (Snapdragon 855).
The inference performance of my deep learning model (Mobilenet V2) is as below:
GPU: 5 ~ 15 ms
CPU: 120 ~ 140 ms
Also, same deep learning model implemented by Tensorflow Lite (CPU) is considerably faster than SNPE.
TFLite CPU: 20 ~ 40 ms
My code is almost the same with the example app in SNPE SDK.
Why is the performance gap between SNPE and TFLite on CPU?
Thanks in advance.
BTW, quantization is NOT used here.
TFlite only use CPU. GPU or NNAPI is not used.
Yes. I also met the samilar case. I have to see SNPE CPU runtime is very slow. I think maybe Qualcomm didnot do much optimization on CPU runtime. BTW, SNPE CPU runtime can only running on one thread, not support multi-threads.