Intel® Distribution of OpenVINO™ Toolkit
Community support and discussions about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all things computer vision-related on Intel® platforms.

tensorflow reshape not working on Myriad X

Community Manager



I got this error when I tried to run mvNCCheck on a graph with the tf.reshape() op.


Network Input tensors ['inp:0#6']


Network Output tensors ['out:0#11']


Blob generated


USB: Transferring Data…


USB: Myriad Execution Finished


Output is in Channel Minor format


USB: Myriad Connection Closing.


E: [ 0] dispatcherEventReceive:247 dispatcherEventReceive() Read failed -1


E: [ 0] eventReader:254 Failed to receive event, the device may have reset


E: [ 0] checkGraphMonitorResponse:922 XLink error


Traceback (most recent call last):


File "/usr/local/bin/mvNCCheck", line 228, in


quit_code = check_net(, args.image, args.inputnode, args.outputnode, args.nshaves, args.inputsize, args.weights, args)


File "/usr/local/bin/mvNCCheck", line 202, in check_net


timings, myriad_output = run_myriad(graph_file, args)


File "/usr/local/bin/ncsdk/Controllers/", line 335, in run_myriad




File "/usr/local/lib/python3.6/dist-packages/mvnc/", line 778, in destroy


raise Exception(Status(status))


Exception: Status.ERROR




Below is simple code to reproduce the error:


import tensorflow as tf


import os


os.environ['CUDA_VISIBLE_DEVICES'] = str(-1)


x = tf.placeholder(tf.float32, [1, 3, 6, 9], name='inp')


x = tf.contrib.layers.conv2d(x, 9, 1)


y = tf.reshape(x, [1, 162], name='out')


saver = tf.train.Saver()


with tf.Session() as sess:, ''path/to/save/my_model')




This is how I ran mvNCCheck


mvNCCheck path/to/save/my_model.meta -w path/to/save/my_model -in inp -on out


If I remove tf.reshape(), everything works fine. Any thoughts on why this is happening?

0 Kudos
1 Reply
Community Manager

Hi @yatao


Are you using the Intel Neural Compute Stick 2? If so, you will need to use OpenVINO as the Neural Compute SDK only supports the original Intel Neural Compute Stick.


Take a look at the get started guide: