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.
6408 Discussions

[Error 15] Toolkit Error: Parser Not Supported: None

idata
Employee
1,152 Views

When I try to compile some of my own Caffe models I get the error "[Error 15] Toolkit Error: Parser Not Supported: None". What does this mean?

 

My prototxt file looks like this:

 

input: "data"

 

input_shape {

 

dim: 1

 

dim: 1

 

dim: 48

 

dim: 48

 

}

 

layer {

 

name: "convh0"

 

type: "Convolution"

 

bottom: "data"

 

top: "convh0"

 

param {

 

lr_mult: 1

 

decay_mult: 1

 

}

 

param {

 

lr_mult: 2

 

decay_mult: 0

 

}

 

convolution_param {

 

num_output: 64

 

kernel_size: 5

 

stride: 1

 

weight_filler {

 

value: 0

 

}

 

bias_filler {

 

value: 0

 

}

 

}

 

}

 

layer {

 

name: "reluh0"

 

type: "ReLU"

 

bottom: "convh0"

 

top: "convh0"

 

}

 

layer {

 

name: "poolh0"

 

type: "Pooling"

 

bottom: "convh0"

 

top: "poolh0"

 

pooling_param {

 

pool: MAX

 

kernel_size: 3

 

stride: 2

 

}

 

}

 

layer {

 

name: "convh1"

 

type: "Convolution"

 

bottom: "poolh0"

 

top: "convh1"

 

param {

 

lr_mult: 1

 

decay_mult: 1

 

}

 

param {

 

lr_mult: 2

 

decay_mult: 0

 

}

 

convolution_param {

 

num_output: 64

 

kernel_size: 5

 

stride: 1

 

weight_filler {

 

value: 0

 

}

 

bias_filler {

 

value: 0

 

}

 

}

 

}

 

layer {

 

name: "reluh1"

 

type: "ReLU"

 

bottom: "convh1"

 

top: "convh1"

 

}

 

layer {

 

name: "convh2"

 

type: "Convolution"

 

bottom: "convh1"

 

top: "convh2"

 

param {

 

lr_mult: 1

 

decay_mult: 1

 

}

 

param {

 

lr_mult: 2

 

decay_mult: 0

 

}

 

convolution_param {

 

num_output: 128

 

kernel_size: 4

 

stride: 1

 

weight_filler {

 

value: 0

 

}

 

bias_filler {

 

value: 0

 

}

 

}

 

}

 

layer {

 

name: "reluh2"

 

type: "ReLU"

 

bottom: "convh2"

 

top: "convh2"

 

}

 

layer {

 

name: "fch3"

 

type: "InnerProduct"

 

bottom: "convh2"

 

top: "fch3"

 

inner_product_param {

 

num_output: 1024

 

weight_filler {

 

value: 0

 

}

 

bias_filler {

 

value: 0

 

}

 

}

 

}

 

layer {

 

name: "reluh3"

 

type: "ReLU"

 

bottom: "fch3"

 

top: "fch3"

 

}

 

layer {

 

name: "y"

 

type: "InnerProduct"

 

bottom: "fch3"

 

top: "y"

 

param {

 

lr_mult: 1

 

decay_mult: 1

 

}

 

param {

 

lr_mult: 2

 

decay_mult: 0

 

}

 

inner_product_param {

 

num_output: 38

 

weight_filler {

 

value: 0

 

}

 

bias_filler {

 

value: 0

 

}

 

}

 

}
0 Kudos
4 Replies
idata
Employee
822 Views

@graspergazon Hi, which version of the SDK are you using to compile your network? The latest SDK can be found here: https://ncsforum.movidius.com/discussion/98/latest-version-movidius-neural-compute-sdk#latest

 

I did not have any trouble compiling your prototxt file with the newest sdk using the mvNCCompile command.

 

Example: $mvNCCompile sample.prototxt -s 12

0 Kudos
idata
Employee
822 Views

I am using the latest version of the SDK, cloned from the git repository.

 

When try mvNCCompile on the original prototxt files I still get Error 15. However, when I copy the contents of a prototxt file to a new file it does compile.
0 Kudos
idata
Employee
822 Views

I got the same error report today using freshly installed ncsdk:

 

parallels@parallels-vm:~/MTCNN/model$ mvNCCompile --version

 

mvNCCompile v02.00, Copyright @ Movidius Ltd 2016
0 Kudos
idata
Employee
822 Views

I was providing .caffemodel instead of .prototxt, so one problem solved.

0 Kudos
Reply