Hi Everyone,
I'm trying to run a face detection code in tensorflow on Snapdragon. Now I'm bring up stage and pretty new to this.
I have few questions:
1. To run my machine learning on snapdragon GPU, is SNPE SDK enough? Do I have to install Android SDK/NDK as well or not needed?
2. First wanted to start with running inception on SNPE. To complete the inception section in tutorial, I see "Permisison Denied" error as follows:
LAB: ~/snpe-1.18.0$ models/inception_v3/scripts/setup_inceptionv3.py -h-bash: models/inception_v3/scripts/setup_inceptionv3.py: Permission denied
Once I resolved this error, I'm planning to run the following command in examples which currently i see an error:
LAB:~/snpe-1.18.0/models/inception_v3$ snpe-net-run --container dlc/inception_v3_quantized.dlc --input_list data/cropped/raw_list.txt
error_code=310; error_message=Dlc read failure. Failed to read archive file /home/amir/snpe-1.18.0/models/inception_v3/dlc; error_component=Dl Container; line_no=477; thread_id=140563612323712
Hi,
1. As mentioned in the SNPE docs, we have to install Android SDK/NDK along with SNPE SDK. You can follow the below link for detailed requirements and steps for installation of SNPE SDK
https://developer.qualcomm.com/software/qualcomm-neural-processing-sdk/getting-started.
2. The permission issue is mostly because the SDK might be installed under super user permissions. Hence the access might be restricted to run the Python script. Try doing the same as super user (su).
Thank you.