As I could see from below, XR2 chip is not in the supported list, is it possible that XR2 use the different Architecture for AI?
----------
- The Qualcomm Neural Processing SDK supports Snapdragon® 855, 845, 820, 835, 712, 675, 660, 653, 652, 650, 636, 632, 630, 626, 625, 450, 439, and 429 as well as Qualcomm® QCS605 and QCS403, Qualcomm® SM6125, the Snapdragon® 820Am automotive platform and Qualcomm Flight. For Qualcomm® Adreno™ GPU support, libOpenCL.so must be present on device.
- Due to the fast evolution of networks and layers, acceleration support is partial and will be expanded through time.
----------
And in SNPE Architecture, the computer Networks,
----------
Execute the network on the Snapdragon CPU, the Qualcomm® AdrenoTMGPU, or the HexagonDSP with HVX* and HMX* support,
----------
In XR2 chip, it includes GPU, HexagonDSP and NPU, is it same as SNPE Architecture?
The SXR2130 can use the SNPE architecture, too. Please use the most recent version, such as SNPE 1.56
It does work, thanks.
Hi All, has anyone succesfully run SNPE on XR2 (Oculus Quest 2)? I'm able to launch the sample image classification application but it seems to only have CPU option for runtime? Has anyone been able to use GPU or DSP for inference? I'm using the latest snpe-1.61.0.zip. Thanks.
Try to copy libsnpe_dsp_v66_domains_v2_skel.so, libsnpe_dsp_domains_v2.so and libsnpe_dsp_domains_v3.so to your device and set the environment variable before run: export ADSP_LIBRARY_PATH=/path/to/your/so
thanks for the suggestion. it is working for me now.