cant run converted keras model in dnn model, get error 'allocateLayers' or 'cv::dnn::dnn4_v20210301::`anonymous-namespace'::addConstNodes'

[Ohad-Multisense]

System information (version)

• OpenCV => 4.5.2 / 4.1.1
• Operating System / Platform => Windows 10 64 Bit / Ubuntu 16.04 64 Bit
• python=> 3.7.9

Detailed description

Hello, i am try to run my trained model with openCV but get errors:

first i am train my model in keras (down will be show summary).
After training i convert keras model to tensorflow model and get pb file
when i try to run the model get error :

// python code example
error: OpenCV(4.5.2) C:\Users\runneradmin\AppData\Local\Temp\pip-req-build-dn5w5exm\opencv\modules\dnn\src\dnn.cpp:3127: error: (-215:Assertion failed) inp.total() in function 'cv::dnn::dnn4_v20210301::Net::Impl::allocateLayers'

if i try to load pb model with pbtxt configuration i am get error:

// python code example
[ERROR:0] global C:\Users\runneradmin\AppData\Local\Temp\pip-req-build-dn5w5exm\opencv\modules\dnn\src\tensorflow\tf_importer.cpp (748) cv::dnn::dnn4_v20210301::`anonymous-namespace'::addConstNodes DNN/TF: Can't handle node='strided_slice/stack_2'. Exception: OpenCV(4.5.2) C:\Users\runneradmin\AppData\Local\Temp\pip-req-build-dn5w5exm\opencv\modules\dnn\src\tensorflow\tf_importer.cpp:742: error: (-215:Assertion failed) const_layers.insert(std::make_pair(name, li)).second in function 'cv::dnn::dnn4_v20210301::`anonymous-namespace'::addConstNodes'

i am try first with openCV version 4.1.1, and after i am try with version 4.5.2 and try different OS (windows 10 and Ubuntu 16.04)

may be someone can help me to load my trained model with openCV

Steps to reproduce

```.py
// python code example
import numpy as np
import cv2
from keras.models import load_model


keras_mdodel = load_model('pb_people_20/sim_mdl_func_v2_pow2_china_people_20.h5')
keras_mdodel.summary()

matcher_path_opencv='pb_people_20/sim_mdl_func_v2_pow2_china_people_20.pb'
matcher_path_opencv_pbtxt='pb_people_20/sim_mdl_func_v2_pow2_china_people_20_constant_graph.pbtxt'

matcher_opencv = cv2.dnn.readNetFromTensorflow(matcher_path_opencv,matcher_path_opencv_pbtxt)

def get_match_opencv(embs_0,embs_1):
    blobs = cv2.dnn.blobFromImages([embs_0,embs_1])
    # Set blob as input to faceNet
    matcher_opencv.setInput(blobs)#,'input_1')
    #opencv_matcher_model.setInput(blobs)
    #opencv_matcher_model.forward() 
    # Runs a forward pass to compute the net output
    return matcher_opencv.forward() 

def match_opencv(embs_0,embs_1):
    res = 0
    for first_emb in embs_0:
        for second_emb in embs_1:
            match_res = np.squeeze(get_match_opencv(first_emb,second_emb))
            res = max(res,match_res)
    return res


emb_comp = np.array([[-0.01565263, -0.03656595, -0.02106497, -0.01172114, -0.03183658,
         0.04423258,  0.07379042, -0.03448544,  0.00917874,  0.10619327,
         0.067057  , -0.04213215, -0.07464422, -0.06878724, -0.09982515,
         0.01137134,  0.04709429,  0.02330536, -0.00764309, -0.14052634,
        -0.17314363, -0.04568923,  0.01383415, -0.00578358, -0.08864444,
         0.07767007, -0.06337555, -0.09800968, -0.04953913,  0.01720826,
        -0.00948498, -0.09592741, -0.10465072, -0.06163599,  0.22899778,
         0.10588204, -0.03106888,  0.06826289,  0.19125976,  0.09685656,
         0.10258153, -0.04890387,  0.07771394,  0.10267048,  0.12820566,
        -0.17984524,  0.10338812,  0.11158907, -0.06663922,  0.05078162,
        -0.0144849 ,  0.03813416, -0.12748823,  0.06191524, -0.09051557,
         0.01777557,  0.14020114, -0.00481044, -0.04369346, -0.06189118,
         0.02411589, -0.2553893 ,  0.04399932, -0.11846916,  0.08772267,
         0.14251669, -0.11032436,  0.09020332, -0.04854635, -0.1081277 ,
        -0.11582536, -0.03723017,  0.03233388,  0.00288666,  0.13660526,
        -0.03048528, -0.10125471,  0.0071717 ,  0.00672208, -0.02564991,
         0.02921752,  0.03457039,  0.01814651, -0.04918699, -0.08088212,
        -0.00733991, -0.0040632 ,  0.12189984, -0.02091988,  0.27729505,
        -0.09878255,  0.01982491,  0.10892831, -0.05085582, -0.08475107,
         0.13422482, -0.04512339,  0.01896015, -0.10464227,  0.04971102,
        -0.07694119, -0.06038413, -0.0517869 ,  0.00917315,  0.00206291,
        -0.02974024, -0.01651836,  0.16789761,  0.07368388, -0.10541685,
         0.14497165,  0.06329738,  0.1521214 , -0.01266835, -0.00671655,
        -0.083036  ,  0.07795148, -0.02421858,  0.17278863, -0.05978476,
        -0.06421243, -0.14550938,  0.00752902,  0.01600735, -0.07240639,
         0.02841103, -0.05694156, -0.11571351]], dtype=np.float32)

img_embs = np.array([[ 0.11008514,  0.01455243,  0.01908959, -0.05834481, -0.14181934,
         0.1882611 ,  0.08232455, -0.11491345,  0.05500843,  0.04011501,
        -0.04828305, -0.03326504, -0.05615507, -0.02345234, -0.0491137 ,
         0.12314314,  0.03639589, -0.02172398,  0.12210201, -0.09503838,
        -0.08331381,  0.01604286,  0.00464164, -0.01787202, -0.0006151 ,
         0.13468699,  0.01733225, -0.10102501, -0.02729505, -0.10542642,
        -0.07366027,  0.02062424, -0.08315776, -0.01222375,  0.18682827,
         0.11723035, -0.16790824,  0.0639649 ,  0.17808685,  0.05183998,
         0.07382489, -0.0351243 ,  0.00582562,  0.01956308,  0.04352709,
        -0.10843045,  0.02012463, -0.05169405, -0.05626178,  0.07084419,
        -0.03077255, -0.02904841, -0.0701267 , -0.04788103,  0.08546986,
         0.05289406,  0.05466829,  0.05425635, -0.0813577 , -0.03403144,
         0.00544576, -0.2616964 ,  0.06646861, -0.05636386,  0.0138996 ,
         0.01612337, -0.00722741, -0.06695902,  0.05770075, -0.05611902,
        -0.00302837, -0.05027693,  0.03430333,  0.02994055,  0.16164383,
        -0.02481532, -0.20324335,  0.0160754 ,  0.06737015, -0.01307072,
         0.14364758,  0.0039455 ,  0.09170939, -0.03038747,  0.02139785,
        -0.0776043 ,  0.07581864,  0.00768354, -0.03795378,  0.11794008,
        -0.10925949, -0.07035936,  0.03584298, -0.05980027, -0.05402324,
         0.11599824, -0.2232277 ,  0.05435819, -0.06722381,  0.04105445,
        -0.12166104,  0.02219271, -0.10503792,  0.04090016,  0.01218935,
        -0.1845343 , -0.00819734,  0.2274523 , -0.0579717 , -0.19095403,
         0.14852127,  0.10555372,  0.11048518,  0.05386249, -0.07497197,
        -0.01151269,  0.15070955,  0.0480066 ,  0.06213563, -0.01823237,
         0.09147692, -0.14006777,  0.072235  ,  0.03428208, -0.08408232,
         0.03943622, -0.07408921, -0.08130765]], dtype=np.float32)


res2 = match_opencv([emb_comp],[img_embs])




```

Test_load_with_open_CV.zip

the summary of the model:
```.py
// python code example
Model: "Similarity_Model"
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to

=================================================================================
ImageA_Input (InputLayer)       (None, 128)          0                                            
__________________________________________________________________________________________________
ImageB_Input (InputLayer)       (None, 128)          0                                            
__________________________________________________________________________________________________
lambda_1 (Lambda)               (None, 128)          0           ImageA_Input[0][0]               
                                                                 ImageB_Input[0][0]               
__________________________________________________________________________________________________
dense_1 (Dense)                 (None, 32)           4128        lambda_1[0][0]                   
__________________________________________________________________________________________________
dropout_1 (Dropout)             (None, 32)           0           dense_1[0][0]                    
__________________________________________________________________________________________________
dense_2 (Dense)                 (None, 16)           528         dropout_1[0][0]                  
__________________________________________________________________________________________________
dropout_2 (Dropout)             (None, 16)           0           dense_2[0][0]                    
__________________________________________________________________________________________________
dense_3 (Dense)                 (None, 1)            17          dropout_2[0][0]                  

============================================================================    
Total params: 4,673
Trainable params: 4,673
Non-trainable params: 0
'''

[Ohad-Multisense]

Thanks @alalek

[rogday]

I think you obtained pb and pbtxt incorrectly. This should work:

import tensorflow as tf
from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2
from tensorflow.keras.models import load_model

def write_pb(model, filename):
    func = tf.function(lambda x: model(x))
    func = func.get_concrete_function([tf.TensorSpec(model_input.shape, model_input.dtype) for model_input in model.inputs])

    func = convert_variables_to_constants_v2(func)
    graph = func.graph.as_graph_def()

    tf.io.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=False)

    return graph

def write_pbtxt(model, graph, filename):
    for i in reversed(range(len(graph.node))):
        if graph.node[i].op == 'Const':
            del graph.node[i]

    graph.library.Clear()

    tf.compat.v1.train.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=True)

def saveModel(model, name):
    graph = write_pb(model, name + '.pb')
    write_pbtxt(model, graph, name + '.pbtxt')

model = load_model('./sim_mdl_func_v2_pow2_china_people_20.h5')
saveModel(model, 'model')

[Ohad-Multisense]

I think you obtained pb and pbtxt incorrectly. This should work:

import tensorflow as tf
from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2
from tensorflow.keras.models import load_model

def write_pb(model, filename):
    func = tf.function(lambda x: model(x))
    func = func.get_concrete_function([tf.TensorSpec(model_input.shape, model_input.dtype) for model_input in model.inputs])

    func = convert_variables_to_constants_v2(func)
    graph = func.graph.as_graph_def()

    tf.io.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=False)

    return graph

def write_pbtxt(model, graph, filename):
    for i in reversed(range(len(graph.node))):
        if graph.node[i].op == 'Const':
            del graph.node[i]

    graph.library.Clear()

    tf.compat.v1.train.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=True)

def saveModel(model, name):
    graph = write_pb(model, name + '.pb')
    write_pbtxt(model, graph, name + '.pbtxt')

model = load_model('./sim_mdl_func_v2_pow2_china_people_20.h5')
saveModel(model, 'model')

thanks i will try it,
witch version of tensorflow you are using?

[rogday]

I think you obtained pb and pbtxt incorrectly. This should work:

import tensorflow as tf
from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2
from tensorflow.keras.models import load_model

def write_pb(model, filename):
    func = tf.function(lambda x: model(x))
    func = func.get_concrete_function([tf.TensorSpec(model_input.shape, model_input.dtype) for model_input in model.inputs])

    func = convert_variables_to_constants_v2(func)
    graph = func.graph.as_graph_def()

    tf.io.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=False)

    return graph

def write_pbtxt(model, graph, filename):
    for i in reversed(range(len(graph.node))):
        if graph.node[i].op == 'Const':
            del graph.node[i]

    graph.library.Clear()

    tf.compat.v1.train.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=True)

def saveModel(model, name):
    graph = write_pb(model, name + '.pb')
    write_pbtxt(model, graph, name + '.pbtxt')

model = load_model('./sim_mdl_func_v2_pow2_china_people_20.h5')
saveModel(model, 'model')

thanks i will try it,
witch version of tensorflow you are using?

I was using 2.5.0 IIRC

[Ohad-Multisense]

I think you obtained pb and pbtxt incorrectly. This should work:

import tensorflow as tf
from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2
from tensorflow.keras.models import load_model

def write_pb(model, filename):
    func = tf.function(lambda x: model(x))
    func = func.get_concrete_function([tf.TensorSpec(model_input.shape, model_input.dtype) for model_input in model.inputs])

    func = convert_variables_to_constants_v2(func)
    graph = func.graph.as_graph_def()

    tf.io.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=False)

    return graph

def write_pbtxt(model, graph, filename):
    for i in reversed(range(len(graph.node))):
        if graph.node[i].op == 'Const':
            del graph.node[i]

    graph.library.Clear()

    tf.compat.v1.train.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=True)

def saveModel(model, name):
    graph = write_pb(model, name + '.pb')
    write_pbtxt(model, graph, name + '.pbtxt')

model = load_model('./sim_mdl_func_v2_pow2_china_people_20.h5')
saveModel(model, 'model')

thanks i will try it,
witch version of tensorflow you are using?

I was using 2.5.0 IIRC

thanks, i will try it with tensorflow 2.5.0 because with 1.14.0 it not working

[Ohad-Multisense]

I think you obtained pb and pbtxt incorrectly. This should work:

import tensorflow as tf
from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2
from tensorflow.keras.models import load_model

def write_pb(model, filename):
    func = tf.function(lambda x: model(x))
    func = func.get_concrete_function([tf.TensorSpec(model_input.shape, model_input.dtype) for model_input in model.inputs])

    func = convert_variables_to_constants_v2(func)
    graph = func.graph.as_graph_def()

    tf.io.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=False)

    return graph

def write_pbtxt(model, graph, filename):
    for i in reversed(range(len(graph.node))):
        if graph.node[i].op == 'Const':
            del graph.node[i]

    graph.library.Clear()

    tf.compat.v1.train.write_graph(graph_or_graph_def=graph, logdir=".", name=filename, as_text=True)

def saveModel(model, name):
    graph = write_pb(model, name + '.pb')
    write_pbtxt(model, graph, name + '.pbtxt')

model = load_model('./sim_mdl_func_v2_pow2_china_people_20.h5')
saveModel(model, 'model')

thanks i will try it,
witch version of tensorflow you are using?

I was using 2.5.0 IIRC

it test with tensorflow 2.5.0
but get error: ValueError: bad marshal data (unknown type code)

Python 3.8.10 (default, Jun  4 2021, 15:09:15) 
Type "copyright", "credits" or "license" for more information.

IPython 7.22.0 -- An enhanced Interactive Python.

runfile('/media/maxim/Windows/Maxim/FaceVerification/Matcher/k2tf_convert__form_forum.py', wdir='/media/maxim/Windows/Maxim/FaceVerification/Matcher')
2021-07-07 15:31:46.369922: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /opt/tensorrt/lib:/usr/local/cuda-9.0/lib64:/lib/x86_64-linux-gnu:/usr/lib/x86_64-linux-gnu:
2021-07-07 15:31:46.369944: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Using TensorFlow backend.
WARNING:tensorflow:From /home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/tensorflow/python/compat/v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
Instructions for updating:
non-resource variables are not supported in the long term
Traceback (most recent call last):

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/utils/generic_utils.py", line 231, in func_load
    code = marshal.loads(raw_code)

ValueError: bad marshal data (unknown type code)


During handling of the above exception, another exception occurred:

Traceback (most recent call last):

  File "/media/maxim/Windows/Maxim/FaceVerification/Matcher/k2tf_convert__form_forum.py", line 50, in <module>
    keras_model = load_model(mdl_path)

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/engine/saving.py", line 419, in load_model
    model = _deserialize_model(f, custom_objects, compile)

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/engine/saving.py", line 225, in _deserialize_model
    model = model_from_config(model_config, custom_objects=custom_objects)

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/engine/saving.py", line 458, in model_from_config
    return deserialize(config, custom_objects=custom_objects)

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/layers/__init__.py", line 52, in deserialize
    return deserialize_keras_object(config,

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/utils/generic_utils.py", line 142, in deserialize_keras_object
    return cls.from_config(

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/engine/network.py", line 1022, in from_config
    process_layer(layer_data)

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/engine/network.py", line 1007, in process_layer
    layer = deserialize_layer(layer_data,

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/layers/__init__.py", line 52, in deserialize
    return deserialize_keras_object(config,

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/utils/generic_utils.py", line 142, in deserialize_keras_object
    return cls.from_config(

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/layers/core.py", line 735, in from_config
    function = func_load(config['function'], globs=globs)

  File "/home/maxim/anaconda2/envs/flask/lib/python3.8/site-packages/keras/utils/generic_utils.py", line 236, in func_load
    code = marshal.loads(raw_code)

ValueError: bad marshal data (unknown type code)

do you know how i can solve it? thanks a lot

[rogday]

I tested my code for saving with python 3.7.10, tensorflow 1.15 and opencv 4.5.2. Everything works fine if I insert tf.compat.v1.enable_eager_execution() at the start.

[Ohad-Multisense]

I tested my code for saving with python 3.7.10, tensorflow 1.15 and opencv 4.5.2. Everything works fine if I insert tf.compat.v1.enable_eager_execution() at the start.
Thank you very much, i try this and get both files: .pb and .pbtxt

converted model according to you instructions:
sim_mdl_func_v2_pow2_china_people_20.zip

but when i try to test it with opench get error:

emb_comp = [np.array([[-2.07526069e-02, -3.76138873e-02, -1.81949530e-02,
        -1.27303926e-03, -2.80939788e-02,  3.59291025e-02,
         7.89560154e-02, -3.59906815e-02, -2.18113977e-03,
         1.24354869e-01,  7.09442645e-02, -3.73696014e-02,
        -7.89967924e-02, -7.08339959e-02, -1.08526520e-01,
        -5.16016176e-03,  4.82012965e-02,  1.79995038e-02,
        -8.50073341e-03, -1.43060327e-01, -1.79133609e-01,
        -5.06920442e-02,  1.30563295e-02, -1.70601644e-02,
        -8.52133632e-02,  7.94231817e-02, -4.59608659e-02,
        -9.71303508e-02, -5.04867733e-02,  2.24852767e-02,
        -1.70053616e-02, -8.70493948e-02, -1.04366392e-01,
        -8.27868059e-02,  2.25306377e-01,  9.93802696e-02,
        -2.76333746e-02,  6.41410127e-02,  1.88372731e-01,
         8.63860697e-02,  9.89582837e-02, -4.50619422e-02,
         7.15015009e-02,  1.19220816e-01,  1.34420380e-01,
        -1.83178157e-01,  9.51981917e-02,  1.14982471e-01,
        -4.97261807e-02,  5.97340763e-02, -1.05344653e-02,
         4.21733148e-02, -1.17164105e-01,  4.82109375e-02,
        -9.32606086e-02,  1.34737287e-02,  1.39889300e-01,
        -3.43040272e-04, -3.76531482e-02, -6.32638708e-02,
         1.92606598e-02, -2.68090129e-01,  4.57393005e-02,
        -1.21325538e-01,  9.15667936e-02,  1.50711983e-01,
        -1.18540570e-01,  9.71808583e-02, -4.71785925e-02,
        -9.60603356e-02, -1.25995740e-01, -3.34104635e-02,
         2.29550730e-02,  8.03949032e-03,  1.38658345e-01,
        -3.91187556e-02, -8.95357132e-02,  1.62388720e-02,
         7.89797457e-04, -1.24329710e-02,  1.69271659e-02,
         4.36490513e-02,  1.87081564e-02, -5.01879491e-02,
        -8.33383203e-02,  1.45936632e-04, -1.64315850e-02,
         1.07943699e-01, -3.75260226e-02,  2.72473335e-01,
        -1.00223340e-01,  2.67054718e-02,  1.03623085e-01,
        -4.34564687e-02, -8.17834139e-02,  1.23940527e-01,
        -4.41038907e-02,  2.31757257e-02, -1.09333508e-01,
         3.59940566e-02, -8.47578347e-02, -6.52976036e-02,
        -3.89938951e-02,  3.28362011e-03, -1.21749649e-02,
        -2.25499980e-02, -1.25503661e-02,  1.66699931e-01,
         7.29326755e-02, -1.01906836e-01,  1.43347621e-01,
         6.38188198e-02,  1.49770677e-01,  2.11979705e-03,
         2.86606583e-03, -6.54566661e-02,  8.51759166e-02,
        -3.50150280e-02,  1.66765362e-01, -4.12612893e-02,
        -5.51314838e-02, -1.58738792e-01,  6.95668627e-04,
         1.28341401e-02, -8.43253508e-02,  2.57490240e-02,
        -5.32488748e-02, -1.07368752e-01]], dtype=np.float32)]

img_embs = [np.array([[ 0.08299512,  0.04560803,  0.03116382, -0.03930001, -0.08403633,
         0.14950442,  0.0729406 , -0.10549465,  0.10704392,  0.06648093,
        -0.06004801, -0.02714206, -0.07316988, -0.08157247, -0.04738788,
         0.14541239, -0.02736695,  0.0721011 ,  0.10346765, -0.05763485,
        -0.06799629,  0.01061096,  0.05501751,  0.07780259, -0.01651143,
         0.14266111, -0.03180892, -0.13978282, -0.07177536, -0.09723921,
        -0.04592864, -0.04917077, -0.09854629,  0.0072313 ,  0.1446461 ,
         0.10760631, -0.10153292,  0.05217709,  0.22002015,  0.05179396,
         0.11984765, -0.04360949, -0.00775165, -0.0418541 ,  0.06799418,
         0.01069122,  0.05947721, -0.06431236, -0.10907257,  0.07055985,
        -0.08934348, -0.04483014, -0.05907475,  0.01866955,  0.06799407,
        -0.00205917,  0.09784131, -0.02283142, -0.05978012,  0.02966089,
         0.13560592, -0.21287084,  0.06778804, -0.08625988,  0.06007105,
         0.02196306,  0.03125143, -0.03363482,  0.09186722, -0.04559114,
        -0.0246855 ,  0.00345274, -0.01052804,  0.06904986,  0.17674777,
        -0.04841331, -0.22535895,  0.022614  ,  0.13221958, -0.01430765,
         0.1128735 , -0.0665758 ,  0.10571466, -0.05914513,  0.00428186,
         0.01118833,  0.06930657,  0.0920514 , -0.02113676,  0.16240422,
        -0.08114551, -0.02897336,  0.09336429, -0.01132694, -0.03403351,
         0.06171956, -0.16830747,  0.09194379, -0.0766651 ,  0.06004078,
        -0.09434112,  0.03440706, -0.13617109,  0.00559863,  0.03040126,
        -0.17688258, -0.12178319,  0.24337214, -0.12474367, -0.13440283,
         0.10498703,  0.11160351,  0.10737723, -0.03220632, -0.12789217,
        -0.00120715,  0.02936077,  0.03229442,  0.09272455,  0.00215561,
         0.00268578, -0.1165002 ,  0.02916852, -0.0132524 , -0.07830146,
        -0.01242953, -0.05116267, -0.10529439]], dtype=np.float32)]



# load model
matcher_path_opencv = './sim_mdl_func_v2_pow2_china_people_20.pb'
matcher_opencv = cv2.dnn.readNetFromTensorflow(matcher_path_opencv,matcher_path_opencv+'txt')

def get_match_opencv(matcher_opencv,embs_0,embs_1):
    blobs_feat = np.squeeze(np.stack([embs_0,embs_1]).astype(np.float32))
    matcher_opencv.setInput(blobs_feat)#,'input_1')
    # Runs a forward pass to compute the net output
    return matcher_opencv.forward()     

res_opencv =  get_match_opencv(matcher_opencv,emb_comp,img_embs) 

and the error is:

error: OpenCV(4.5.2) ../modules/dnn/src/dnn.cpp:3127: error: (-215:Assertion failed) inp.total() in function 'allocateLayers'

did you get this error?

[Ohad-Multisense]

I tested my code for saving with python 3.7.10, tensorflow 1.15 and opencv 4.5.2. Everything works fine if I insert tf.compat.v1.enable_eager_execution() at the start.
Thank you very much, i try this and get both files: .pb and .pbtxt

converted model according to you instructions:
sim_mdl_func_v2_pow2_china_people_20.zip

but when i try to test it with opench get error:

emb_comp = [np.array([[-2.07526069e-02, -3.76138873e-02, -1.81949530e-02,
        -1.27303926e-03, -2.80939788e-02,  3.59291025e-02,
         7.89560154e-02, -3.59906815e-02, -2.18113977e-03,
         1.24354869e-01,  7.09442645e-02, -3.73696014e-02,
        -7.89967924e-02, -7.08339959e-02, -1.08526520e-01,
        -5.16016176e-03,  4.82012965e-02,  1.79995038e-02,
        -8.50073341e-03, -1.43060327e-01, -1.79133609e-01,
        -5.06920442e-02,  1.30563295e-02, -1.70601644e-02,
        -8.52133632e-02,  7.94231817e-02, -4.59608659e-02,
        -9.71303508e-02, -5.04867733e-02,  2.24852767e-02,
        -1.70053616e-02, -8.70493948e-02, -1.04366392e-01,
        -8.27868059e-02,  2.25306377e-01,  9.93802696e-02,
        -2.76333746e-02,  6.41410127e-02,  1.88372731e-01,
         8.63860697e-02,  9.89582837e-02, -4.50619422e-02,
         7.15015009e-02,  1.19220816e-01,  1.34420380e-01,
        -1.83178157e-01,  9.51981917e-02,  1.14982471e-01,
        -4.97261807e-02,  5.97340763e-02, -1.05344653e-02,
         4.21733148e-02, -1.17164105e-01,  4.82109375e-02,
        -9.32606086e-02,  1.34737287e-02,  1.39889300e-01,
        -3.43040272e-04, -3.76531482e-02, -6.32638708e-02,
         1.92606598e-02, -2.68090129e-01,  4.57393005e-02,
        -1.21325538e-01,  9.15667936e-02,  1.50711983e-01,
        -1.18540570e-01,  9.71808583e-02, -4.71785925e-02,
        -9.60603356e-02, -1.25995740e-01, -3.34104635e-02,
         2.29550730e-02,  8.03949032e-03,  1.38658345e-01,
        -3.91187556e-02, -8.95357132e-02,  1.62388720e-02,
         7.89797457e-04, -1.24329710e-02,  1.69271659e-02,
         4.36490513e-02,  1.87081564e-02, -5.01879491e-02,
        -8.33383203e-02,  1.45936632e-04, -1.64315850e-02,
         1.07943699e-01, -3.75260226e-02,  2.72473335e-01,
        -1.00223340e-01,  2.67054718e-02,  1.03623085e-01,
        -4.34564687e-02, -8.17834139e-02,  1.23940527e-01,
        -4.41038907e-02,  2.31757257e-02, -1.09333508e-01,
         3.59940566e-02, -8.47578347e-02, -6.52976036e-02,
        -3.89938951e-02,  3.28362011e-03, -1.21749649e-02,
        -2.25499980e-02, -1.25503661e-02,  1.66699931e-01,
         7.29326755e-02, -1.01906836e-01,  1.43347621e-01,
         6.38188198e-02,  1.49770677e-01,  2.11979705e-03,
         2.86606583e-03, -6.54566661e-02,  8.51759166e-02,
        -3.50150280e-02,  1.66765362e-01, -4.12612893e-02,
        -5.51314838e-02, -1.58738792e-01,  6.95668627e-04,
         1.28341401e-02, -8.43253508e-02,  2.57490240e-02,
        -5.32488748e-02, -1.07368752e-01]], dtype=np.float32)]

img_embs = [np.array([[ 0.08299512,  0.04560803,  0.03116382, -0.03930001, -0.08403633,
         0.14950442,  0.0729406 , -0.10549465,  0.10704392,  0.06648093,
        -0.06004801, -0.02714206, -0.07316988, -0.08157247, -0.04738788,
         0.14541239, -0.02736695,  0.0721011 ,  0.10346765, -0.05763485,
        -0.06799629,  0.01061096,  0.05501751,  0.07780259, -0.01651143,
         0.14266111, -0.03180892, -0.13978282, -0.07177536, -0.09723921,
        -0.04592864, -0.04917077, -0.09854629,  0.0072313 ,  0.1446461 ,
         0.10760631, -0.10153292,  0.05217709,  0.22002015,  0.05179396,
         0.11984765, -0.04360949, -0.00775165, -0.0418541 ,  0.06799418,
         0.01069122,  0.05947721, -0.06431236, -0.10907257,  0.07055985,
        -0.08934348, -0.04483014, -0.05907475,  0.01866955,  0.06799407,
        -0.00205917,  0.09784131, -0.02283142, -0.05978012,  0.02966089,
         0.13560592, -0.21287084,  0.06778804, -0.08625988,  0.06007105,
         0.02196306,  0.03125143, -0.03363482,  0.09186722, -0.04559114,
        -0.0246855 ,  0.00345274, -0.01052804,  0.06904986,  0.17674777,
        -0.04841331, -0.22535895,  0.022614  ,  0.13221958, -0.01430765,
         0.1128735 , -0.0665758 ,  0.10571466, -0.05914513,  0.00428186,
         0.01118833,  0.06930657,  0.0920514 , -0.02113676,  0.16240422,
        -0.08114551, -0.02897336,  0.09336429, -0.01132694, -0.03403351,
         0.06171956, -0.16830747,  0.09194379, -0.0766651 ,  0.06004078,
        -0.09434112,  0.03440706, -0.13617109,  0.00559863,  0.03040126,
        -0.17688258, -0.12178319,  0.24337214, -0.12474367, -0.13440283,
         0.10498703,  0.11160351,  0.10737723, -0.03220632, -0.12789217,
        -0.00120715,  0.02936077,  0.03229442,  0.09272455,  0.00215561,
         0.00268578, -0.1165002 ,  0.02916852, -0.0132524 , -0.07830146,
        -0.01242953, -0.05116267, -0.10529439]], dtype=np.float32)]



# load model
matcher_path_opencv = './sim_mdl_func_v2_pow2_china_people_20.pb'
matcher_opencv = cv2.dnn.readNetFromTensorflow(matcher_path_opencv,matcher_path_opencv+'txt')

def get_match_opencv(matcher_opencv,embs_0,embs_1):
    blobs_feat = np.squeeze(np.stack([embs_0,embs_1]).astype(np.float32))
    matcher_opencv.setInput(blobs_feat)#,'input_1')
    # Runs a forward pass to compute the net output
    return matcher_opencv.forward()     

res_opencv =  get_match_opencv(matcher_opencv,emb_comp,img_embs) 

and the error is:

error: OpenCV(4.5.2) ../modules/dnn/src/dnn.cpp:3127: error: (-215:Assertion failed) inp.total() in function 'allocateLayers'

did you get this error?

resolve this error... i am set not correct the input to the network.
correct input set is:

def get_match_opencv(matcher_opencv,embs_0,embs_1):
    matcher_opencv.setInput(embs_0,'x')
    matcher_opencv.setInput(embs_1,'x_1')
    # Runs a forward pass to compute the net output
    return matcher_opencv.forward()   

but when i try to load my old network that has different in second layer i get error:

cv2.error: OpenCV(4.5.2) ../modules/dnn/src/dnn.cpp:621: error: (-2:Unspecified error) Can't create layer "Similarity_Model/lambda_1/pow" of type "Pow" in function 'getLayerInstance'

the different is that in old network i am use pow function instead multiplication in new network

may be you know how i can solve it?
thanks a lot

[rogday]

the different is that in old network i am use pow function instead multiplication in new network

Do you mean A^B element-wise? I do not believe we support this layer. ONNX importer supports A^c, where A, B - tensors and c is a constant scalar. If the original problem was resolved, feel free to close this issue and open a new one, concerning Pow.