Community
cancel
Showing results for 
Search instead for 
Did you mean: 
idata
Community Manager
410 Views

Error compiling TensorFlow network

Hi all,

 

I've got the following error compiling my TensorFlow network:

 

mvNCCompile TF_Model/tf_model.meta -in=input_1 -on=filtered_detections/map/TensorArrayStack_2/TensorArrayGatherV3

 

/usr/local/bin/ncsdk/Controllers/Parsers/TensorFlowParser/Convolution.py:44: SyntaxWarning: assertion is always true, perhaps remove parentheses?

 

assert(False, "Layer type not supported by Convolution: " + obj.type)

 

mvNCCompile v02.00, Copyright @ Intel Corporation 2017

 

/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/tf_inspect.py:45: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

shape: [1, None, None, 3]

 

Traceback (most recent call last):

 

File "/usr/local/bin/mvNCCompile", line 169, in

 

create_graph(args.network, args.image, args.inputnode, args.outputnode, args.outfile, args.nshaves, args.inputsize, args.weights, args.explicit_concat, args.ma2480, args.scheduler, args.new_parser, args)

 

File "/usr/local/bin/mvNCCompile", line 148, in create_graph

 

load_ret = load_network(args, parser, myriad_config)

 

File "/usr/local/bin/ncsdk/Controllers/Scheduler.py", line 100, in load_network

 

parse_ret = parse_tensor(arguments, myriad_conf)

 

File "/usr/local/bin/ncsdk/Controllers/TensorFlowParser.py", line 259, in parse_tensor

 

input_data = np.random.uniform(0, 1, shape)

 

File "mtrand.pyx", line 1302, in mtrand.RandomState.uniform

 

File "mtrand.pyx", line 242, in mtrand.cont2_array_sc**

 

TypeError: 'NoneType' object cannot be interpreted as an integer**

 

I'm running on Ubuntu 16.04 with NCAP2, trying to compile a Retinanet network to the NCS.

 

Any idea about what went wrong?

 

Thanks in advance!
Tags (1)
0 Kudos
1 Reply
idata
Community Manager
81 Views

Hi @fpartika

 

The error you're getting is due to your model using variable input sizes. Visit this thread for more information.

 

Best Regards,

 

Sahira
Reply