Oraoto

It's in `/opt/movidius/NCSDK/ncsdk-x86_64/tk/Controllers/TensorFlowParser.py`.

Sorry, i don't have NCSDK 1 installation right now, maybe you should try NCSDK 2.

I have come across similar problem, solved by modifying ncsdk source. In `/usr/local/bin/ncsdk/Controllers/TensorFlowParser.py` line 1059, add a feed_dict to eval: ``` # desired_shape = node.inputs[1].eval() desired_shape = node.inputs[1].eval(feed_dict={inputnode + ':0'...

@Derekabc here's my [modified TensorFlowParser.py](https://gist.github.com/oraoto/7656c7593b5bc7e481118e8c167d4d13), and [some working examples](https://github.com/oraoto/learn_ml/tree/master/ncs).

@Derekabc th.jpg is random zebra image from [Bing image search](https://bing.com/images/search?q=zebra):smile:, you can try different images from the [ImageNet](http://image-net.org/) project.

@Derekabc I use `tensorboard --logdir logs`to view the graph and get the names.