frame

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Alert: Beginning Tuesday, June 25th, we will be freezing this site and migrating the content and forums to our new home at https://forums.intel.com/s/topic/0TO0P000000PqZDWA0/intel-neural-compute-sticks. Check it out now!

Export Keras VGG16 fine tuned model to Movidius

Hello,
I have fine tuned the Keras implementation of the VGG16 net loaded with the following code:
vgg16 = VGG16(weights="imagenet", include_top=False, input_shape=(IMAGE_SIZE, IMAGE_SIZE, 3))

After replacing the classification layers and performed the training, I have exported the json model and the weights to convert the net for tensorflow.

Then, I have exported the checkpoint, .data, .index and .meta files using the tensorflow functions.

I have all the file necessary to create the model compatible with Movidius with the command:
mvNCCompile tf_export/garbage_vgg16.meta -in=vgg16_input -on=dense_2/Softmax

But the command fails with the following error:
[Error 5] Toolkit Error: Stage Details Not Supported: Switch

How can I fix this problem?

Thanks in advance for your help.

Comments

  • 1 Comment sorted by Votes Date Added
  • Hi @macc.n

    Can you please post the whole output?
    "Stage details not supported" errors usually mean that your network requires operations or layers that aren't yet supported by the NCSDK. If you enable debugging (change to debug = true in /ncsdk/Controllers/TensorFlowParser.py) you can see exactly which layers or operations are unsupported.

    Best Regards,
    Sahira

Sign In or Register to comment.