Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.
6392 Discussions

mvNCProfile fails with Error 25 - 5x5 Convolution outputChannels Dimension too small

idata
Employee
820 Views

Hi,

 

I am trying to run a caffe model for anti-aliasing (AA) for inclusion in rendering pipeline - replicating work from Deep Shading paper. The idea is to compare performance of trained model with traditional methods of AA, or other graphics techniques.

 

The sample prototxt is below - 2 conv and 2 RELU layers. Using _mvNCCompile_ on this works fine, but _mvNCProfile_ fails with _[Error 25] Myriad Error: "5x5 Convolution outputChannels Dimension too small."_.

 

Any help in solving this is appreciated.

 

aaforNC.prototxt

 

name: "ShadingNet"

 

input: "data"

 

input_shape {

 

dim: 1

 

dim: 6

 

dim: 256

 

dim: 256

 

}

 

layer {

 

name: "down_level_0_conv_top"

 

type: "Convolution"

 

bottom: "data"

 

top: "down_level_0_conv_top"

 

#blobs_lr: 1

 

#blobs_lr: 2

 

convolution_param {

 

num_output: 8

 

kernel_size: 5

 

pad: 2

 

stride: 1

 

engine: CAFFE

 

weight_filler {

 

type: "gaussian"

 

std: 0.011

 

}

 

bias_filler {

 

type: "constant"

 

value: 0

 

}

 

}

 

}

 

layer {

 

name: "down_level_0_relu"

 

type: "ReLU"

 

bottom: "down_level_0_conv_top"

 

top: "down_level_0_relu"

 

relu_param{

 

negative_slope: 0.01

 

engine: CAFFE

 

}

 

}

 

layer {

 

name: "up_level_0_conv_top"

 

type: "Convolution"

 

bottom: "down_level_0_relu"

 

top: "up_level_0_conv_top"

 

#blobs_lr: 1

 

#blobs_lr: 2

 

convolution_param {

 

num_output: 1

 

kernel_size: 5

 

pad: 2

 

stride: 1

 

engine: CAFFE

 

weight_filler {

 

type: "gaussian"

 

std: 0.01

 

}

 

bias_filler {

 

type: "constant"

 

value: 0

 

}

 

}

 

}

 

layer {

 

name: "up_level_0_relu"

 

type: "ReLU"

 

bottom: "up_level_0_conv_top"

 

top: "result"

 

relu_param{

 

engine: CAFFE

 

negative_slope: 0.01

 

}

 

}

 

layer {

 

name: "loss"

 

type: "Softmax"

 

bottom: "result"

 

top: "loss"

 

}

 

Message and Error Log

 

spl@spl-Blade:~/Desktop$ mvNCCompile netMV.prototxt

 

mvNCCompile v02.00, Copyright @ Movidius Ltd 2016

 

/usr/local/lib/python3.5/dist-packages/matplotlib/backends/backend_gtk3.py:58: DeprecationWarning: Gdk.Cursor.new is deprecated

 

cursors.MOVE : Gdk.Cursor.new(Gdk.CursorType.FLEUR),

 

/usr/local/lib/python3.5/dist-packages/matplotlib/backends/backend_gtk3agg.py:18: UserWarning: The Gtk3Agg backend is known to not work on Python 3.x with pycairo. Try installing cairocffi.

 

"The Gtk3Agg backend is known to not work on Python 3.x with pycairo. "

 

/usr/local/lib/python3.5/dist-packages/scipy/lib/decorator.py:219: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

first = inspect.getargspec(caller)[0][0] # first arg

 

/usr/local/lib/python3.5/dist-packages/scipy/optimize/nonlin.py:1498: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

args, varargs, varkw, defaults = inspect.getargspec(jac.init)

 

/usr/local/lib/python3.5/dist-packages/scipy/stats/_distn_infrastructure.py:611: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

sign = inspect.getargspec(self._stats)

 

/usr/local/lib/python3.5/dist-packages/scipy/stats/_distn_infrastructure.py:648: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

shapes_args = inspect.getargspec(meth)

 

__ WARNING: using empty weights __

 

spl@spl-Blade:~/Desktop$ mvNCProfile netMV.prototxt

 

mvNCProfile v02.00, Copyright @ Movidius Ltd 2016

 

/usr/local/lib/python3.5/dist-packages/matplotlib/backends/backend_gtk3.py:58: DeprecationWarning: Gdk.Cursor.new is deprecated

 

cursors.MOVE : Gdk.Cursor.new(Gdk.CursorType.FLEUR),

 

/usr/local/lib/python3.5/dist-packages/matplotlib/backends/backend_gtk3agg.py:18: UserWarning: The Gtk3Agg backend is known to not work on Python 3.x with pycairo. Try installing cairocffi.

 

"The Gtk3Agg backend is known to not work on Python 3.x with pycairo. "

 

/usr/local/lib/python3.5/dist-packages/scipy/lib/decorator.py:219: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

first = inspect.getargspec(caller)[0][0] # first arg

 

/usr/local/lib/python3.5/dist-packages/scipy/optimize/nonlin.py:1498: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

args, varargs, varkw, defaults = inspect.getargspec(jac.init)

 

/usr/local/lib/python3.5/dist-packages/scipy/stats/_distn_infrastructure.py:611: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

sign = inspect.getargspec(self._stats)

 

/usr/local/lib/python3.5/dist-packages/scipy/stats/_distn_infrastructure.py:648: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead

 

shapes_args = inspect.getargspec(meth)

 

__ WARNING: using empty weights __

 

USB: Transferring Data…

 

[Error 25] Myriad Error: "5x5 Convolution outputChannels Dimension too small.".
0 Kudos
0 Replies
Reply