Hi,
I followed the steps in http://developer.qualcomm.com/docs/snpe/tutorial_inceptionv3.html, but I failed running the Inception v3 Model with snpe-net-run program.
I am working on a docker which is tensorflow/tensorflow:1.6.0-py3. The version information about the environment is as follows.
Ubuntu==16.04.7
tensorflow==1.6.0
Python==3.5.2
SNPE==1.47.0.2501
command:
snpe-net-run --container dlc/inception_v3_quantized.dlc --input_list data/cropped/raw_list.txt
error log:
error_code=310; error_message=Dlc read failure. Failed to read archive file /workdir/snpe-1.47.0.2501/models/inception_v3/dlc/inception_v3_quantized.dlc; error_component=Dl Container; line_no=530; thread_id=139815578068928
Thanks for any replies.
Mats
I am facing the same issue, was there any solution found to this?
Hi r-matsumoto,
Can you please respond back the below points?
What is Permission of inception_v3_quantized.dlc file?
Can you try using the Tensorflow >1.12 , <1.14 for DLC Conversion.
Thanks & Regards,
Sahil Bandar
Dear customer,
Could you please help to check your running environment again from below notes?
tensorflow == 1.15
Python == 3.6
Looks like the wrong version framework you have used.
BR.
Wei
Hi,
I am also working on tensorflow inspection v3 model . First I also got same errors because of tensorflow version mismatch
Environmnet details :
Ubuntu:18.04
tensorflow:1.15.0
python:3.6.13
SNPE:1.54.2.2899
I followed the steps in https://developer.qualcomm.com/sites/default/files/docs/snpe/tutorial_inceptionv3.html
In the above link you can find how to set-up and run tensorflow inspection v3 model.
Step1: First set-up the environment.
Path: Tutorials and examples -> Tutorials Setup
Step2: Run the tensorflow inspection v3 model.
Path: Tutorials and examples -> Running Nets -> TensorFlow Inspection V3