I have tested with Inception_v3 model. It works fine with all(cpu/gpu/dsp/aip) runtimes.
when I tryed to run my model net converted from tensorflow's pb file, it words fine with cpu/gpu runtimes but failed with dsp/aip runtimes.
I execute the command below:
snpe-net-run --container model_q.dlc --input_list data/target_raw_list.txt --use_dsp
and I recieve the log below:
-------------------------------------------------------------------------------Model String: N/ASNPE v1.49.0.2587-------------------------------------------------------------------------------Processing DNN input(s):data/fullbody.rawApplying paddingApplying paddingApplying paddingerror_code=910; error_message=DSP runtime system error. Failed to execute network. Error code: -6, layer id: 0, op type: 0, extra: 0, backendErrorCode: 0.; error_component=DSP Runtime; line_no=610; thread_id=-192352832
Enviroment
Snapdragon 855
SNPE version 1.49.0
The message in this log is not enough, I want to know the reason why dsp failed. Is it beause any op or layer is not supported? and how can I get more log about the failure?