I wish to get the qualcomm SNPE (snapdragon neural processing engine) working on my linux (not Android) board (flightPro w/ qualcomm 820.) . It works fine on the cpu.
I've successfully followed the examples provided to load alexnet onto my 820 board and run snpe (snpe-net-run) in cpu mode. It does not run in gpu mode.
Searching the web and forums (e.g., https://developer.qualcomm.com/forum/qdn-forums/software/qualcomm-neural-processing-sdk/59207/quickpayportal) it seems that all (?) linux boards may be missing the opencl driver that would be required to make this work.
Following the example...
> snpe-net-run --container bvlc_alexnet.dlc --input_list target_raw_list.txt --use_gpuThe selected runtime is not available on this platform. Continue anyway to observe the failure at network creation time.
Aborted
I expected the gpu to work (and hopefully, fingers crossed to be substantially faster than the cpu!)
Hi, Sorry to bother you for non-relative problem, but can you teach me how to use snpe-net-run command?
I have followed this tutorial https://developer.qualcomm.com/docs/snpe/tutorial_setup.html and got inception v3 dlc file.
However, when I test dlc file by command "snpe-net-run --container dlc/inception_v3.dlc --input_list data/cropped/raw_list.txt"
It shows "snpe-net-run: command not found". It seems that I miss some packages or dependecis, but I can't find any solution in the toturial.
Do you have any idea for this problem?
Any advice will be appreciated, thanks.
Hi frank.huang_1,
$ export SNPE_ROOT=<obsolute path to NPE sdk>
Ex: export SNPE_ROOT=/home/user/git/QDN/snpe-1.37
$ export ANDROID_NDK_ROOT=<obsolute path to android ndk>
Ex: export ANDROID_NDK_ROOT=/home/sysadmin/git/QDN/snpe-1.37/android/Ndk/android-ndk-r21
#Execute the below enviroment setup script in NPE sdk folder
$ source ./bin/envsetup.sh -t <path to tensorflow>
Ex: source ./bin/envsetup.sh -t /usr/local/lib/python3.5/dist-packages/tensorflow