- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
I have a problem using OpenVino R1.1 Model_Optimizer to convert inference_graph.pb file to IR.
Below the output of the error where is clearly mentioned to contact the MO developers, I can only open a request to the Support:
[ ERROR ] Please contact Model Optimizer developers
File Attached with all details.
Please, can you support here?
Thank you
Model Optimizer version: 2019.1.1-83-g28dfbfd
[ ERROR ] -------------------------------------------------
[ ERROR ] ----------------- INTERNAL ERROR ----------------
[ ERROR ] Unexpected exception happened.
[ ERROR ] Please contact Model Optimizer developers and forward the following information:
[ ERROR ] Exception occurred during running replacer "None (<class 'extensions.front.no_op_eraser.NoOpEraser'>)": The node training/group_deps must have just one input
[ ERROR ] Traceback (most recent call last):
File "/opt/intel/openvino_2019.1.144/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 167, in apply_replacements
replacer.find_and_replace_pattern(graph)
File "/opt/intel/openvino_2019.1.144/deployment_tools/model_optimizer/mo/front/common/replacement.py", line 141, in find_and_replace_pattern
apply_pattern(graph, action=self.replace_sub_graph, **self.pattern())
File "/opt/intel/openvino_2019.1.144/deployment_tools/model_optimizer/mo/middle/pattern_match.py", line 95, in apply_pattern
action(graph, match)
File "/opt/intel/openvino_2019.1.144/deployment_tools/model_optimizer/extensions/front/no_op_eraser.py", line 40, in replace_sub_graph
graph.erase_node(match['noop'])
File "/opt/intel/openvino_2019.1.144/deployment_tools/model_optimizer/mo/graph/graph.py", line 493, in erase_node
assert len(inputs) <= 1, "The node {} must have just one input".format(node.soft_get('name'))
AssertionError: The node training/group_deps must have just one input
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Cosma, is your *.pb a frozen one ? The error :
The node training/group_deps must have just one input
Leads me to think that maybe your model is not frozen.
Also please do a --log_level DEBUG and post the ERROR part to this forum. I think you and I have "talked before" on this IDZ thread
From where did you get this model ? Please tell me more about this model so that I may help you. Was it a Keras model that you converted to a Tensorflow Frozen pb ? If you are still trying to do Resnet 50 why not choose from one of our Model Optimizer Supported Tensorflow List ? We support Resnet50 v1 and v2.
Thanks,
Shubha
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dearest Cosma,
Thank you for the attachments. I will definitely take a look !
Sincerely,
Shubha
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Shubha,
during the last days, I did other tests using a new code and applying all suggestions to save TF.keras models described in the official link below:
https://www.tensorflow.org/tutorials/keras/save_and_restore_models
Before sending this new material, I would like to know if you prefer to finish analyzing the first attachements I posted few days ago.
Please, let me know
Thank you
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Shubha,
attached the code I used to generate the mentioned .pb file, all rows are commented.
Additionally, attached, you can find the error using Model Optimizer to convert also the other file "Model_1.pb" as frozen file generated also with the simple code attached.
In both cases, the Model Optimizer doean't convert .pb to IR files.
Consider the code used and the model are really simple.
I need to find the right procedure to freeze TF Models usable with Intel Open Vino and Model Optimizer before to consider OpenVino as useful framework in production (with all steps involved: define TF Model, Training Data, freeze model and the Inference).
Thank you!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Shubha,
I solved the problem!
In any case, many thanks for you support and time.
Regards
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Cosma,
Congrats ! Can you let me know how you solved the problem ? Please share your findings with the community if you have the time and inclination to do so.
Thanks,
Shubha
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Shubha,
After some months the solution was simple:
I reviewed all code and cleaned some lines in the Python code used to define the model and training data!
I invested a lot of time analyzing random errors about frozen .h5 file, useless activity!
Hoping it helps
Best Regards
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dearest Cosma,
Thank you for coming back to the forum and telling me ! Trust me, I have been where you are many many times ! Often times we learn the most from these silly errors !
I am super happy though that you were able to get up and running on OpenVino.
Please download the latest OpenVino 2019R2 which was just released today. Give it a try won't you ?
Thanks,
Shubha
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello, I have been facing this error for few days now and can't seem to resolve it. Am new to the Openvino platform. I get the error when I try to run the demos that came with Intel Openvino toolkit
Microsoft Windows [Version 6.3.9600]
(c) 2013 Microsoft Corporation. All rights reserved.
C:\WINDOWS\system32>cd ..\..\
C:\>cd program files (x86)
C:\Program Files (x86)>cd intelswtools
C:\Program Files (x86)\IntelSWTools>cd openvino_2019.3.379
C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379>cd bin
C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\bin>setupvars.bat
Python 3.6.0
ECHO is off.
PYTHONPATH=C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_to
ols\open_model_zoo\tools\accuracy_checker;C:\Program Files (x86)\IntelSWTools\op
envino_2019.3.379\python\python3.6;C:\Program Files (x86)\IntelSWTools\openvino_
2019.3.379\python\python3;
[setupvars.bat] OpenVINO environment initialized
C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\bin>cd ..\
C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379>cd deployment_tools
C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools>cd demo
C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools\demo>de
mo_squeezenet_download_convert_run.bat
target_precision = FP16
Python 3.6.0
ECHO is off.
PYTHONPATH=C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_to
ols\open_model_zoo\tools\accuracy_checker;C:\Program Files (x86)\IntelSWTools\op
envino_2019.3.379\python\python3.6;C:\Program Files (x86)\IntelSWTools\openvino_
2019.3.379\python\python3;C:\Program Files (x86)\IntelSWTools\openvino_2019.3.37
9\deployment_tools\open_model_zoo\tools\accuracy_checker;C:\Program Files (x86)\
IntelSWTools\openvino_2019.3.379\python\python3.6;C:\Program Files (x86)\IntelSW
Tools\openvino_2019.3.379\python\python3;
[setupvars.bat] OpenVINO environment initialized
INTEL_OPENVINO_DIR is set to C:\Program Files (x86)\IntelSWTools\openvino_2019.3
.379
Python 3.6.0
ECHO is off.
Requirement already satisfied: pyyaml in c:\program files\python36\lib\site-pack
ages (from -r C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment
_tools\demo\\..\open_model_zoo\tools\downloader\requirements.in (line 1)) (5.3)
Requirement already satisfied: requests in c:\users\manpasu\appdata\roaming\pyth
on\python36\site-packages (from -r C:\Program Files (x86)\IntelSWTools\openvino_
2019.3.379\deployment_tools\demo\\..\open_model_zoo\tools\downloader\requirement
s.in (line 2)) (2.18.4)
Requirement already satisfied: chardet<3.1.0,>=3.0.2 in c:\users\manpasu\appdata
\roaming\python\python36\site-packages (from requests->-r C:\Program Files (x86)
\IntelSWTools\openvino_2019.3.379\deployment_tools\demo\\..\open_model_zoo\tools
\downloader\requirements.in (line 2)) (3.0.4)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\manpasu\appdata\ro
aming\python\python36\site-packages (from requests->-r C:\Program Files (x86)\In
telSWTools\openvino_2019.3.379\deployment_tools\demo\\..\open_model_zoo\tools\do
wnloader\requirements.in (line 2)) (2019.11.28)
Requirement already satisfied: urllib3<1.23,>=1.21.1 in c:\users\manpasu\appdata
\roaming\python\python36\site-packages (from requests->-r C:\Program Files (x86)
\IntelSWTools\openvino_2019.3.379\deployment_tools\demo\\..\open_model_zoo\tools
\downloader\requirements.in (line 2)) (1.22)
Requirement already satisfied: idna<2.7,>=2.5 in c:\users\manpasu\appdata\roamin
g\python\python36\site-packages (from requests->-r C:\Program Files (x86)\IntelS
WTools\openvino_2019.3.379\deployment_tools\demo\\..\open_model_zoo\tools\downlo
ader\requirements.in (line 2)) (2.6)
WARNING: You are using pip version 19.3.1; however, version 20.0.2 is available.
You should consider upgrading via the 'python -m pip install --upgrade pip' comm
and.
Download public squeezenet1.1 model
python "C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools
\open_model_zoo\tools\downloader\downloader.py" --name squeezenet1.1 --output_di
r C:\Users\Manpasu\Documents\Intel\OpenVINO\openvino_models\models --cache_dir C
:\Users\Manpasu\Documents\Intel\OpenVINO\openvino_models\cache
################|| Downloading models ||################
========== Downloading C:\Users\Manpasu\Documents\Intel\OpenVINO\openvino_models
\models\public\squeezenet1.1\squeezenet1.1.prototxt
========== Downloading C:\Users\Manpasu\Documents\Intel\OpenVINO\openvino_models
\models\public\squeezenet1.1\squeezenet1.1.caffemodel
... 100%, 4834 KB, 134 KB/s, 36 seconds passed
################|| Post-processing ||################
========== Replacing text in C:\Users\Manpasu\Documents\Intel\OpenVINO\openvino_
models\models\public\squeezenet1.1\squeezenet1.1.prototxt
squeezenet1.1 model downloading completed
Waiting for 0 seconds, press a key to continue ...
###############|| Install Model Optimizer prerequisites ||###############
Waiting for 0 seconds, press a key to continue ...
Python 3.6.0
ECHO is off.
Requirement already satisfied: networkx<2.4,>=1.11 in c:\users\manpasu\appdata\r
oaming\python\python36\site-packages (from -r ..\requirements_caffe.txt (line 1)
) (2.3)
Requirement already satisfied: numpy>=1.12.0 in c:\users\manpasu\appdata\roaming
\python\python36\site-packages (from -r ..\requirements_caffe.txt (line 2)) (1.1
7.4)
Requirement already satisfied: protobuf==3.6.1 in c:\users\manpasu\appdata\roami
ng\python\python36\site-packages (from -r ..\requirements_caffe.txt (line 3)) (3
.6.1)
Requirement already satisfied: defusedxml>=0.5.0 in c:\users\manpasu\appdata\roa
ming\python\python36\site-packages (from -r ..\requirements_caffe.txt (line 4))
(0.6.0)
Requirement already satisfied: decorator>=4.3.0 in c:\users\manpasu\appdata\roam
ing\python\python36\site-packages (from networkx<2.4,>=1.11->-r ..\requirements_
caffe.txt (line 1)) (4.4.1)
Requirement already satisfied: six>=1.9 in c:\users\manpasu\appdata\roaming\pyth
on\python36\site-packages (from protobuf==3.6.1->-r ..\requirements_caffe.txt (l
ine 3)) (1.13.0)
Requirement already satisfied: setuptools in c:\program files\python36\lib\site-
packages (from protobuf==3.6.1->-r ..\requirements_caffe.txt (line 3)) (39.0.1)
WARNING: You are using pip version 19.3.1; however, version 20.0.2 is available.
You should consider upgrading via the 'python -m pip install --upgrade pip' comm
and.
********************************************************************************
*********
Warning: please expect that Model Optimizer conversion might be slow.
You can boost conversion speed by installing protobuf-*.egg located in the
"model-optimizer\install_prerequisites" folder or building protobuf library from
sources.
For more information please refer to Model Optimizer FAQ, question #80.
Waiting for 0 seconds, press a key to continue ...
###############|| Run Model Optimizer ||###############
Waiting for 0 seconds, press a key to continue ...
python "C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools
\open_model_zoo\tools\downloader\converter.py" --mo "C:\Program Files (x86)\Inte
lSWTools\openvino_2019.3.379\deployment_tools\model_optimizer\mo.py" --name "squ
eezenet1.1" -d "C:\Users\Manpasu\Documents\Intel\OpenVINO\openvino_models\models
" -o "C:\Users\Manpasu\Documents\Intel\OpenVINO\openvino_models\ir" --precisions
"FP16"
========= Converting squeezenet1.1 to IR (FP16)
Conversion command: "C:\Program Files\Python36\python.exe" -- "C:\Program Files
(x86)\IntelSWTools\openvino_2019.3.379\deployment_tools\model_optimizer\mo.py" -
-framework=caffe --data_type=FP16 --output_dir=C:\Users\Manpasu\Documents\Intel\
OpenVINO\openvino_models\ir\public\squeezenet1.1\FP16 --model_name=squeezenet1.1
--input_shape=[1,3,227,227] --input=data --mean_values=data[104.0,117.0,123.0]
--output=prob --input_model=C:\Users\Manpasu\Documents\Intel\OpenVINO\openvino_m
odels\models\public\squeezenet1.1/squeezenet1.1.caffemodel --input_proto=C:\User
s\Manpasu\Documents\Intel\OpenVINO\openvino_models\models\public\squeezenet1.1/s
queezenet1.1.prototxt
Model Optimizer arguments:
Common parameters:
- Path to the Input Model: C:\Users\Manpasu\Documents\Intel\OpenVIN
O\openvino_models\models\public\squeezenet1.1/squeezenet1.1.caffemodel
- Path for generated IR: C:\Users\Manpasu\Documents\Intel\OpenVIN
O\openvino_models\ir\public\squeezenet1.1\FP16
- IR output name: squeezenet1.1
- Log level: ERROR
- Batch: Not specified, inherited from the model
- Input layers: data
- Output layers: prob
- Input shapes: [1,3,227,227]
- Mean values: data[104.0,117.0,123.0]
- Scale values: Not specified
- Scale factor: Not specified
- Precision of IR: FP16
- Enable fusing: True
- Enable grouped convolutions fusing: True
- Move mean values to preprocess section: False
- Reverse input channels: False
Caffe specific parameters:
- Path to Python Caffe* parser generated from caffe.proto: C:\Progr
am Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools\model_optimizer
\mo\front\caffe\proto
- Enable resnet optimization: True
- Path to the Input prototxt: C:\Users\Manpasu\Documents\Intel\OpenVIN
O\openvino_models\models\public\squeezenet1.1/squeezenet1.1.prototxt
- Path to CustomLayersMapping.xml: C:\Program Files (x86)\IntelSWTo
ols\openvino_2019.3.379\deployment_tools\model_optimizer\extensions\front\caffe\
CustomLayersMapping.xml
- Path to a mean file: Not specified
- Offsets for a mean file: Not specified
Model Optimizer version: 2019.3.0-408-gac8584cb7
[ ERROR ] -------------------------------------------------
[ ERROR ] ----------------- INTERNAL ERROR ----------------
[ ERROR ] Unexpected exception happened.
[ ERROR ] Please contact Model Optimizer developers and forward the following i
nformation:
[ ERROR ] DLL load failed: The specified procedure could not be found.
[ ERROR ] Traceback (most recent call last):
File "C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools
\model_optimizer\mo\main.py", line 298, in main
return driver(argv)
File "C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools
\model_optimizer\mo\main.py", line 250, in driver
import mo.pipeline.caffe as mo_caffe
File "C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools
\model_optimizer\mo\pipeline\caffe.py", line 22, in <module>
from mo.front.caffe import custom_layers_mapping, loader
File "C:\Program Files (x86)\IntelSWTools\openvino_2019.3.379\deployment_tools
\model_optimizer\mo\front\caffe\loader.py", line 24, in <module>
from google.protobuf import text_format
File "C:\Users\Manpasu\AppData\Roaming\Python\Python36\site-packages\google\pr
otobuf\text_format.py", line 54, in <module>
from google.protobuf.internal import type_checkers
File "C:\Users\Manpasu\AppData\Roaming\Python\Python36\site-packages\google\pr
otobuf\internal\type_checkers.py", line 55, in <module>
from google.protobuf.internal import decoder
File "C:\Users\Manpasu\AppData\Roaming\Python\Python36\site-packages\google\pr
otobuf\internal\decoder.py", line 90, in <module>
from google.protobuf.internal import encoder
File "C:\Users\Manpasu\AppData\Roaming\Python\Python36\site-packages\google\pr
otobuf\internal\encoder.py", line 73, in <module>
from google.protobuf.internal import wire_format
File "C:\Users\Manpasu\AppData\Roaming\Python\Python36\site-packages\google\pr
otobuf\internal\wire_format.py", line 36, in <module>
from google.protobuf import descriptor
File "C:\Users\Manpasu\AppData\Roaming\Python\Python36\site-packages\google\pr
otobuf\descriptor.py", line 47, in <module>
from google.protobuf.pyext import _message
ImportError: DLL load failed: The specified procedure could not be found.
[ ERROR ] ---------------- END OF BUG REPORT --------------
[ ERROR ] -------------------------------------------------
FAILED:
squeezenet1.1
Error
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Manpasu,
Please upgrade your OpenVINO toolkit to the latest 2020.1 release and ensure your system has the following software component versions.
- Microsoft Visual Studio* with C++ 2019, 2017, or 2015 with MSBuild
- CMake 3.4 or higher 64-bit ( 3.14 if you are using Microsoft Visual Studio 2019)
- Python 3.6.5 64-bit
Regards,
Jesus

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page