Can't convert (EasyOCR) Openvino IR model to tensorflow lite

[MjiS]

Issue Type

Bug, Others

OS

Ubuntu, Other

OS architecture

x86_64, aarch64, Other

Programming Language

Python, Other

Framework

OpenVINO, TensorFlowLite

Download URL for ONNX / OpenVINO IR

OpenVino Files: https://drive.google.com/file/d/1TGpFkPv62mos3ZaxFk0sbhx4cpGP8zuP/view?usp=sharing ONNX File: https://drive.google.com/file/d/1-KgxCUxfXmX-L2vuxSzMZC9WhV2vq_S0/view?usp=sharing Pytorch File: https://drive.google.com/file/d/1-8XmqzoM2kR5IM_lhsAO0yp7rjpLQ4gS/view?usp=sharing

Convert Script

pip install onnxruntime
pip install onnx-simplifier

#PyTorch to ONNX Conversion

import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.nn.init as init
from torchvision import models
from torchvision.models.vgg import model_urls
from collections import namedtuple
from collections import OrderedDict

class BidirectionalLSTM(nn.Module):

    def __init__(self, input_size, hidden_size, output_size):
        super(BidirectionalLSTM, self).__init__()
        self.rnn = nn.LSTM(input_size, hidden_size, bidirectional=True, batch_first=True)
        self.linear = nn.Linear(hidden_size * 2, output_size)

    def forward(self, input):
        """
        input : visual feature [batch_size x T x input_size]
        output : contextual feature [batch_size x T x output_size]
        """
        try: # multi gpu needs this
            self.rnn.flatten_parameters()
        except: # quantization doesn't work with this 
            pass
        recurrent, _ = self.rnn(input)  # batch_size x T x input_size -> batch_size x T x (2*hidden_size)
        output = self.linear(recurrent)  # batch_size x T x output_size
        return output

class VGG_FeatureExtractor(nn.Module):

    def __init__(self, input_channel, output_channel=256):
        super(VGG_FeatureExtractor, self).__init__()
        self.output_channel = [int(output_channel / 8), int(output_channel / 4),
                               int(output_channel / 2), output_channel]
        self.ConvNet = nn.Sequential(
            nn.Conv2d(input_channel, self.output_channel[0], 3, 1, 1), nn.ReLU(True),
            nn.MaxPool2d(2, 2),
            nn.Conv2d(self.output_channel[0], self.output_channel[1], 3, 1, 1), nn.ReLU(True),
            nn.MaxPool2d(2, 2),
            nn.Conv2d(self.output_channel[1], self.output_channel[2], 3, 1, 1), nn.ReLU(True),
            nn.Conv2d(self.output_channel[2], self.output_channel[2], 3, 1, 1), nn.ReLU(True),
            nn.MaxPool2d((2, 1), (2, 1)),
            nn.Conv2d(self.output_channel[2], self.output_channel[3], 3, 1, 1, bias=False),
            nn.BatchNorm2d(self.output_channel[3]), nn.ReLU(True),
            nn.Conv2d(self.output_channel[3], self.output_channel[3], 3, 1, 1, bias=False),
            nn.BatchNorm2d(self.output_channel[3]), nn.ReLU(True),
            nn.MaxPool2d((2, 1), (2, 1)),
            nn.Conv2d(self.output_channel[3], self.output_channel[3], 2, 1, 0), nn.ReLU(True))

    def forward(self, input):
        return self.ConvNet(input)

class Model(nn.Module):

    def __init__(self, input_channel, output_channel, hidden_size, num_class):
        super(Model, self).__init__()
        """ FeatureExtraction """
        self.FeatureExtraction = VGG_FeatureExtractor(input_channel, output_channel)
        self.FeatureExtraction_output = output_channel
        self.AdaptiveAvgPool = nn.AdaptiveAvgPool2d((None, 1))

        """ Sequence modeling"""
        self.SequenceModeling = nn.Sequential(
            BidirectionalLSTM(self.FeatureExtraction_output, hidden_size, hidden_size),
            BidirectionalLSTM(hidden_size, hidden_size, hidden_size))
        self.SequenceModeling_output = hidden_size

        """ Prediction """
        self.Prediction = nn.Linear(self.SequenceModeling_output, num_class)

    def forward(self, input):
        """ Feature extraction stage """
        visual_feature = self.FeatureExtraction(input)
        visual_feature = self.AdaptiveAvgPool(visual_feature.permute(0, 3, 1, 2))
        visual_feature = visual_feature.squeeze(3)

        """ Sequence modeling stage """
        contextual_feature = self.SequenceModeling(visual_feature)

        """ Prediction stage """
        prediction = self.Prediction(contextual_feature.contiguous())

        return prediction

input_channel = 1
output_channel = 256
hidden_size = 256
num_class = 97

model = Model(input_channel, output_channel, hidden_size, num_class)
state_dict = torch.load("/content/custom_example.pth", map_location="cpu")
new_state_dict = OrderedDict()
for key, value in state_dict.items():
    new_key = key[7:]
    new_state_dict[new_key] = value
model.load_state_dict(new_state_dict)

sample_input = torch.rand((32, 1, 64, 600))

torch.onnx.export(
    model,                  # PyTorch Model
    sample_input,                    # Input tensor
    '/content/custom_example.onnx',        # Output file (eg. 'output_model.onnx')
    opset_version=12)       # Operator support version
    # input_names=['input,text'],   # Input tensor name (arbitary)
    # output_names=['prediction'] # Output tensor name (arbitary)
# )

#ONNX Simplification

!python -m onnxsim /content/custom_example.onnx  /content/custom_ocr1.onnx

pip install openvino-dev[caffe,onnx,tensorflow2,pytorch,mxnet]

#ONNX to OpenVino Conversion

!mo --input_model /content/custom_ocr1.onnx --output_dir /content/openvino1 --input_shape [32,1,64,600] --data_type FP32

#OpenVino to Tensorflow Conversion

!git clone https://github.com/PINTO0309/openvino2tensorflow

%cd /content/openvino2tensorflow/openvino2tensorflow
!python openvino2tensorflow.py \
  --model_path /content/openvino1/custom_ocr1.xml \
  --model_output_path /content/tensorflow1 \
  --output_float16_quant_tflite \
  --output_no_quant_float32_tflite \
  --output_full_integer_quant_tflite

Description

Hi, I'm using EasyOCR custom_example.pth model for conversion.It is pytorch based. I can easily convert it to ONNX > ONNX Simplified > OpenVino but I can't convert it to tflite using you openvino2tensorflow.py. I'm using Colab for this. I had trouble getting Openvino mo.py so I converted it manually to ONNX from there on I followed your steps.

Sorry for my mistakes, I'm new to this.

I can share you my colab notebook if you got any issues regarding my approach.

Here is the repos link I'm trying to convert.

Relevant Log Output

Log

```shell Streaming output truncated to the last 5000 lines. [[-2.95811333e-02 -6.77071139e-02 -1.04522124e-01] [ 4.50438075e-02 3.84310223e-02 -9.83352289e-02] [ 4.37608771e-02 4.57786992e-02 -7.72977173e-02]] [[-6.32593483e-02 7.30905402e-03 -9.36140195e-02] [ 1.63864996e-02 -4.28028479e-02 -1.91088334e-01] [ 1.60162479e-01 1.75632522e-01 -4.98583131e-02]]] [[[-4.82106842e-02 -1.06339090e-01 -8.43107030e-02] [ 1.51326098e-02 1.55510278e-02 1.92827508e-02] [-8.71984195e-03 7.28080347e-02 6.41333163e-02]] [[ 1.07924372e-01 -4.57621403e-02 -1.70236379e-02] [ 2.02840250e-02 -8.87788273e-03 8.34525377e-02] [-2.44103864e-01 -1.61868542e-01 8.99570137e-02]] [[ 1.24929979e-01 1.53552860e-01 1.66937292e-01] [-5.87170199e-03 -6.14559185e-03 1.66431397e-01] [-2.27416217e-01 -5.48819005e-02 -6.39001979e-03]] ... [[ 8.06598142e-02 -4.90045287e-02 -1.89861611e-01] [ 2.30177641e-01 -1.67296007e-02 -8.54323953e-02] [ 1.46779940e-01 5.44505455e-02 1.58826753e-01]] [[-6.09425232e-02 1.83403362e-02 -1.21939756e-01] [-9.14027542e-02 -8.44789743e-02 -3.81563120e-02] [-1.05654180e-01 1.36012912e-01 -1.43561978e-02]] [[ 6.86693564e-02 -5.25557511e-02 -7.68397178e-04] [ 8.11080858e-02 -4.18115743e-02 7.09754452e-02] [-2.79048011e-02 3.66497338e-02 9.77664813e-02]]] ... [[[-4.48200516e-02 -2.30259731e-01 -1.22266062e-01] [-9.43673998e-02 -1.17822394e-01 8.41443054e-03] [-9.54898074e-03 -1.31063173e-02 1.40621647e-01]] [[ 1.08646162e-01 -2.47144550e-02 -1.02791920e-01] [ 2.21570004e-02 6.39272332e-02 -2.65616551e-02] [-2.22024117e-02 3.63171846e-02 -2.95106675e-02]] [[ 5.73377050e-02 1.25504017e-01 5.44849737e-03] [ 1.32233351e-01 4.08724509e-02 1.21828197e-02] [-4.23353016e-02 1.91647839e-02 7.05294833e-02]] ... [[ 8.92549977e-02 5.81360199e-02 -4.31775711e-02] [-4.65084054e-02 -2.37391498e-02 -9.96773615e-02] [-5.92803545e-02 -7.04194931e-03 -7.93661848e-02]] [[-9.01751816e-02 -5.98597201e-03 1.18264414e-01] [-1.39788970e-01 -1.08852096e-01 -6.16656840e-02] [-2.61297170e-02 -4.31453362e-02 -7.00454265e-02]] [[-4.61471342e-02 9.62751806e-02 7.06367269e-02] [ 1.06112704e-01 1.07557446e-01 6.53385445e-02] [ 9.82305259e-02 -1.81966219e-02 2.54098214e-02]]] [[[-8.62546265e-02 1.04071222e-01 2.21220385e-02] [-1.55964881e-01 2.97929309e-02 2.33170539e-02] [ 9.56139043e-02 -1.26986757e-01 -1.45588294e-01]] [[ 1.21927828e-01 1.34220095e-02 8.91835839e-02] [-1.21829867e-01 -5.80755956e-02 5.51275946e-02] [-5.50245345e-02 4.36670035e-02 7.95496907e-03]] [[ 3.09066158e-02 2.90900525e-02 -6.52338639e-02] [ 5.25978878e-02 -9.80800465e-02 2.08655978e-03] [ 9.97798238e-03 3.67663689e-02 5.43354638e-02]] ... [[-6.96470663e-02 -5.78866489e-02 -1.49183825e-01] [ 1.62278935e-01 5.08819073e-02 -8.96392670e-03] [ 3.38246450e-02 8.44074339e-02 4.19791192e-02]] [[ 2.34789830e-02 -6.25639185e-02 -1.53599948e-01] [-1.03807673e-02 -1.30833238e-01 -4.82163858e-03] [-1.04984917e-01 8.11734274e-02 3.62244993e-02]] [[-2.19632816e-02 5.01569286e-02 5.35188355e-02] [-1.12485968e-01 -1.26890928e-01 8.34817514e-02] [ 1.79442279e-02 5.71724586e-02 9.80309993e-02]]] [[[ 4.17121686e-02 1.63396895e-02 -2.09202081e-01] [ 7.53018036e-02 -2.59469956e-01 -2.95245312e-02] [ 8.02911818e-02 1.04729988e-01 2.60156035e-01]] [[-1.80374682e-02 3.55196395e-03 -2.68372089e-01] [ 4.83784154e-02 1.31930172e-01 -8.70164558e-02] [ 1.01166606e-01 1.35724425e-01 1.93020254e-02]] [[-1.28046021e-01 -3.12202126e-02 -4.70940247e-02] [-1.92891136e-02 1.84634581e-01 7.59749711e-02] [-7.05002621e-02 4.36586253e-02 1.04314849e-01]] ... [[ 7.57369697e-02 -1.77860297e-02 2.94807237e-02] [ 2.57692952e-03 -4.69644758e-04 -9.68134850e-02] [-9.28373635e-02 4.36916128e-02 3.77617292e-02]] [[-1.12077661e-01 -9.88401696e-02 -1.46991000e-01] [ 3.46258841e-02 1.64831430e-01 -6.40104562e-02] [ 1.96114376e-01 1.35532275e-01 -3.03323045e-02]] [[-1.20795503e-01 -5.18574454e-02 6.75762221e-02] [-1.32524461e-01 -1.61652938e-01 -7.79546872e-02] [ 4.02205512e-02 -1.65308136e-02 1.25296623e-01]]]] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='conv2d_3/Conv2D:0', description="created by layer 'conv2d_3'") ==================================================================================== layer_type: Const layer_id: 21 tf_layers_dict_shape: (1, 128, 1, 1) tf_layers_dict_value: [[[[ 0.06365544]] [[ 0.17440169]] [[ 0.13802384]] [[ 0.1434599 ]] [[ 0.09554026]] [[ 0.19104205]] [[ 0.14794402]] [[ 0.05967022]] [[ 0.14538807]] [[ 0.28102633]] [[ 0.14817083]] [[ 0.06452471]] [[ 0.22792052]] [[ 0.1496419 ]] [[ 0.05057549]] [[ 0.07517863]] [[ 0.18098344]] [[ 0.2277265 ]] [[ 0.10095646]] [[ 0.14097348]] [[ 0.0386198 ]] [[ 0.09742323]] [[ 0.22435041]] [[ 0.02771827]] [[ 0.04130811]] [[ 0.16568539]] [[ 0.01617244]] [[ 0.12669352]] [[ 0.16841334]] [[ 0.22216603]] [[ 0.14352629]] [[ 0.08214001]] [[ 0.01751791]] [[ 0.11086492]] [[ 0.24264213]] [[ 0.2054666 ]] [[ 0.13668662]] [[ 0.21881661]] [[ 0.10328092]] [[ 0.06210288]] [[ 0.2187342 ]] [[ 0.21094921]] [[ 0.2390452 ]] [[ 0.11723056]] [[ 0.1987968 ]] [[ 0.16230391]] [[ 0.14695776]] [[ 0.18023731]] [[ 0.10384943]] [[ 0.17390479]] [[ 0.19256116]] [[ 0.15763739]] [[ 0.19655801]] [[ 0.16949137]] [[ 0.11731795]] [[ 0.13143264]] [[-0.0140219 ]] [[ 0.03496253]] [[ 0.05618617]] [[ 0.11404498]] [[ 0.0296807 ]] [[ 0.09027019]] [[ 0.1132808 ]] [[ 0.08199051]] [[ 0.11314929]] [[ 0.03395184]] [[ 0.06342249]] [[ 0.24081771]] [[ 0.16111426]] [[ 0.1126646 ]] [[ 0.17360877]] [[ 0.08913379]] [[ 0.19635409]] [[ 0.02979483]] [[ 0.01221019]] [[ 0.14273937]] [[ 0.13429476]] [[ 0.11160374]] [[ 0.2897043 ]] [[ 0.21560079]] [[ 0.10185473]] [[ 0.21019731]] [[ 0.18140098]] [[ 0.24455835]] [[ 0.15140295]] [[ 0.17605507]] [[ 0.1170727 ]] [[ 0.07941964]] [[ 0.06998542]] [[ 0.07473787]] [[ 0.15409398]] [[ 0.23048171]] [[ 0.3052191 ]] [[ 0.18031162]] [[ 0.14394389]] [[ 0.033415 ]] [[ 0.10926103]] [[ 0.05425028]] [[-0.0813401 ]] [[ 0.12198608]] [[ 0.12026691]] [[ 0.27022 ]] [[ 0.0826319 ]] [[ 0.01014584]] [[ 0.06637743]] [[ 0.07636148]] [[ 0.21211134]] [[ 0.18471187]] [[ 0.06800966]] [[ 0.14172472]] [[ 0.1784296 ]] [[-0.00446683]] [[ 0.15346543]] [[ 0.23843904]] [[ 0.01337082]] [[ 0.05301565]] [[ 0.25440654]] [[ 0.01492397]] [[ 0.2045006 ]] [[ 0.08787836]] [[ 0.13799052]] [[ 0.1399398 ]] [[ 0.15308578]] [[ 0.17615095]] [[ 0.03357711]] [[ 0.16514328]] [[ 0.21165489]] [[ 0.21479228]]]] ==================================================================================== layer_type: Add layer_id: 22 input_layer0: layer_id=20: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='conv2d_3/Conv2D:0', description="created by layer 'conv2d_3'") input_layer1_value: layer_id=21: [[[[ 0.06365544]] [[ 0.17440169]] [[ 0.13802384]] [[ 0.1434599 ]] [[ 0.09554026]] [[ 0.19104205]] [[ 0.14794402]] [[ 0.05967022]] [[ 0.14538807]] [[ 0.28102633]] [[ 0.14817083]] [[ 0.06452471]] [[ 0.22792052]] [[ 0.1496419 ]] [[ 0.05057549]] [[ 0.07517863]] [[ 0.18098344]] [[ 0.2277265 ]] [[ 0.10095646]] [[ 0.14097348]] [[ 0.0386198 ]] [[ 0.09742323]] [[ 0.22435041]] [[ 0.02771827]] [[ 0.04130811]] [[ 0.16568539]] [[ 0.01617244]] [[ 0.12669352]] [[ 0.16841334]] [[ 0.22216603]] [[ 0.14352629]] [[ 0.08214001]] [[ 0.01751791]] [[ 0.11086492]] [[ 0.24264213]] [[ 0.2054666 ]] [[ 0.13668662]] [[ 0.21881661]] [[ 0.10328092]] [[ 0.06210288]] [[ 0.2187342 ]] [[ 0.21094921]] [[ 0.2390452 ]] [[ 0.11723056]] [[ 0.1987968 ]] [[ 0.16230391]] [[ 0.14695776]] [[ 0.18023731]] [[ 0.10384943]] [[ 0.17390479]] [[ 0.19256116]] [[ 0.15763739]] [[ 0.19655801]] [[ 0.16949137]] [[ 0.11731795]] [[ 0.13143264]] [[-0.0140219 ]] [[ 0.03496253]] [[ 0.05618617]] [[ 0.11404498]] [[ 0.0296807 ]] [[ 0.09027019]] [[ 0.1132808 ]] [[ 0.08199051]] [[ 0.11314929]] [[ 0.03395184]] [[ 0.06342249]] [[ 0.24081771]] [[ 0.16111426]] [[ 0.1126646 ]] [[ 0.17360877]] [[ 0.08913379]] [[ 0.19635409]] [[ 0.02979483]] [[ 0.01221019]] [[ 0.14273937]] [[ 0.13429476]] [[ 0.11160374]] [[ 0.2897043 ]] [[ 0.21560079]] [[ 0.10185473]] [[ 0.21019731]] [[ 0.18140098]] [[ 0.24455835]] [[ 0.15140295]] [[ 0.17605507]] [[ 0.1170727 ]] [[ 0.07941964]] [[ 0.06998542]] [[ 0.07473787]] [[ 0.15409398]] [[ 0.23048171]] [[ 0.3052191 ]] [[ 0.18031162]] [[ 0.14394389]] [[ 0.033415 ]] [[ 0.10926103]] [[ 0.05425028]] [[-0.0813401 ]] [[ 0.12198608]] [[ 0.12026691]] [[ 0.27022 ]] [[ 0.0826319 ]] [[ 0.01014584]] [[ 0.06637743]] [[ 0.07636148]] [[ 0.21211134]] [[ 0.18471187]] [[ 0.06800966]] [[ 0.14172472]] [[ 0.1784296 ]] [[-0.00446683]] [[ 0.15346543]] [[ 0.23843904]] [[ 0.01337082]] [[ 0.05301565]] [[ 0.25440654]] [[ 0.01492397]] [[ 0.2045006 ]] [[ 0.08787836]] [[ 0.13799052]] [[ 0.1399398 ]] [[ 0.15308578]] [[ 0.17615095]] [[ 0.03357711]] [[ 0.16514328]] [[ 0.21165489]] [[ 0.21479228]]]] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.math.add_3/Add:0', description="created by layer 'tf.math.add_3'") ==================================================================================== layer_type: ReLU layer_id: 23 input_layer0: layer_id=22: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.math.add_3/Add:0', description="created by layer 'tf.math.add_3'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.nn.relu_3/Relu:0', description="created by layer 'tf.nn.relu_3'") ==================================================================================== layer_type: MaxPool layer_id: 24 input_layer0: layer_id=23: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.nn.relu_3/Relu:0', description="created by layer 'tf.nn.relu_3'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 128), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_2/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_2'") ==================================================================================== layer_type: Const layer_id: 25 tf_layers_dict_shape: (256, 128, 3, 3) tf_layers_dict_value: [[[[ 1.15246866e-02 7.43661728e-03 7.11896876e-03] [ 7.58890202e-03 -2.35633668e-03 -8.38773791e-03] [-6.47206558e-03 -2.57088803e-03 4.21760976e-03]] [[-2.10882514e-03 1.72332674e-03 4.32542432e-03] [ 5.16221393e-03 -3.48487822e-03 8.78799241e-03] [ 1.10953236e-02 -2.69743544e-03 -5.15680760e-03]] [[-9.53451172e-03 -2.82265991e-03 -1.28406957e-02] [-6.40441896e-03 1.66948466e-03 1.14302104e-03] [-1.60199646e-02 1.40176741e-02 3.06645362e-03]] ... [[-1.16415592e-02 -1.57239903e-02 1.34947170e-02] [-6.02871226e-03 5.74577693e-03 1.67468265e-02] [ 1.61934774e-02 1.12581840e-02 -1.56301931e-02]] [[ 1.72171241e-03 5.09083329e-04 7.56164081e-03] [-1.08633321e-02 5.84405335e-03 4.17543389e-03] [-5.14384033e-03 3.85695486e-03 -2.30960012e-03]] [[-1.64202624e-03 -2.95853504e-04 -2.09731283e-03] [-8.23166966e-03 1.60538091e-03 -1.43687834e-03] [-1.74358822e-02 8.71890597e-03 3.04525113e-03]]] [[[-3.52834305e-03 8.29524593e-04 -1.74687840e-02] [-5.76481177e-03 -3.72369378e-03 -2.27376595e-02] [-4.75500245e-04 -5.82169555e-03 4.61884821e-03]] [[ 2.24112603e-03 3.43121192e-03 6.16211863e-03] [-3.50251235e-03 8.01661517e-03 -6.67249784e-03] [ 9.55544645e-04 -1.85027043e-03 6.02218369e-03]] [[ 7.11276382e-03 2.59197853e-03 -7.43014831e-03] [ 5.00726933e-03 1.10678859e-02 1.89633276e-02] [-1.32207265e-02 6.16477942e-03 8.06163065e-03]] ... [[ 8.65571015e-03 2.62502348e-03 -7.87869561e-03] [-1.51539699e-03 -5.36831981e-03 -5.87983057e-03] [-1.17096433e-03 2.84600956e-03 3.03928624e-03]] [[-5.81907621e-03 -6.98628463e-03 -1.57176238e-02] [-1.28897503e-02 -9.97134950e-03 4.45764605e-03] [ 7.65150553e-03 1.75617244e-02 1.35162361e-02]] [[-4.82016383e-03 7.70875532e-03 5.51228644e-03] [ 7.91868952e-04 1.96585758e-03 -1.30440970e-03] [ 5.33431815e-03 1.84788257e-02 8.61326512e-03]]] [[[ 1.84520311e-03 -4.86603752e-03 8.99988227e-03] [ 5.87106973e-04 -3.48790060e-03 1.37186693e-02] [-1.36622007e-03 3.64021561e-03 1.54278777e-03]] [[-1.20531647e-02 1.03384606e-03 1.98278972e-03] [-8.29347130e-03 -4.19950485e-03 5.80803305e-03] [-5.68737183e-03 -2.01798766e-03 2.46692309e-03]] [[-7.07189972e-03 -3.67173599e-03 -1.08191753e-02] [-1.36312274e-02 -1.26937637e-03 3.07450769e-04] [-2.08825176e-03 1.25343073e-02 7.23117590e-03]] ... [[-3.91222490e-03 1.08885001e-02 -3.35392798e-03] [-9.74506326e-03 4.65849414e-03 1.22959409e-02] [-1.31188892e-02 -9.98106902e-04 1.45298131e-02]] [[ 1.19308673e-03 8.52052215e-03 -3.76744638e-03] [-2.16646516e-03 -2.11952929e-03 -4.98862006e-03] [-5.20573184e-03 3.02712247e-03 -1.65641494e-03]] [[-2.46674358e-03 -4.52469056e-03 1.80805230e-03] [-9.14657430e-04 2.53731944e-03 -3.68000520e-03] [-1.71639584e-03 1.17295394e-02 1.04795853e-02]]] ... [[[-8.33850820e-03 -9.80686676e-03 -6.39130501e-03] [-2.79592979e-03 4.54802020e-03 -1.95195212e-03] [ 1.76193770e-02 1.31955920e-02 7.61975208e-03]] [[ 2.86094774e-03 -6.31535286e-03 3.92993167e-03] [-1.85613055e-03 -1.34525998e-02 -7.90349732e-04] [ 1.44972978e-02 6.75864378e-03 9.67936404e-03]] [[-5.10482397e-03 -9.92135890e-03 -2.94809206e-03] [-2.68083508e-03 -4.50535025e-03 -1.11167515e-02] [-1.00535003e-03 3.08036921e-04 2.88249669e-03]] ... [[ 5.93944546e-03 -2.50258553e-03 -4.41644434e-03] [-4.17348044e-03 5.67906210e-03 -2.43334845e-02] [ 1.35842925e-02 -5.01225377e-03 5.76310034e-04]] [[ 7.25751370e-03 -1.08321160e-02 -5.80613082e-03] [ 3.62618431e-03 3.57717089e-03 -4.92676068e-03] [ 6.23808848e-03 2.22781789e-04 1.07812905e-03]] [[-2.27104109e-02 2.54031015e-03 9.94869880e-03] [ 1.55787263e-02 3.73024563e-03 -2.17096843e-02] [ 7.28548318e-03 6.24256535e-03 -1.25676015e-05]]] [[[ 1.30262226e-03 6.73383661e-03 -2.17128024e-02] [ 5.00287209e-03 7.96902925e-03 4.48334403e-03] [-1.13786934e-02 9.78915254e-04 -9.13041737e-03]] [[-1.84491500e-02 9.65074822e-03 2.66267732e-03] [ 2.43384275e-03 -3.57573875e-03 7.77664501e-03] [-1.05960146e-02 -5.73859084e-03 5.31001063e-03]] [[-8.15782044e-03 -3.53085902e-03 -6.23842003e-03] [-1.22496728e-02 -8.07665847e-03 5.01510128e-03] [-3.10147647e-04 2.26028799e-03 3.77046899e-03]] ... [[-9.19859798e-04 -7.61793088e-03 -7.89463799e-03] [ 5.17135311e-04 -2.78467555e-02 -3.25798281e-02] [-5.69814211e-03 -3.73197696e-03 9.15166270e-03]] [[-1.22190651e-03 -2.61003245e-03 6.19929691e-04] [-4.82921489e-03 -2.35801525e-02 1.39521845e-02] [-1.59817096e-02 3.11788009e-03 -9.53365304e-03]] [[ 7.78590562e-03 4.20127704e-04 -1.63016394e-02] [ 2.69251666e-03 2.19487422e-03 8.89003204e-05] [ 1.40868602e-02 4.79381159e-03 2.79590534e-03]]] [[[ 5.56306506e-04 6.19302457e-03 6.42550923e-03] [-9.78724193e-03 4.22338396e-03 -4.49284865e-03] [ 8.32202422e-05 -2.68504722e-04 4.93027689e-03]] [[ 1.52437240e-02 -1.07793538e-02 5.48798870e-03] [ 5.49978670e-03 1.65854144e-04 7.72294938e-04] [-1.82602240e-03 2.22187908e-03 7.80849997e-03]] [[-1.49104223e-02 -1.12586049e-02 -1.38134323e-02] [-1.15399482e-02 -8.20516609e-03 -7.96749163e-03] [-1.71724660e-03 -1.01414556e-02 2.98672845e-03]] ... [[ 9.98066738e-03 -6.14956813e-03 -9.54485498e-03] [ 7.70576473e-04 -6.31371571e-04 -2.53834436e-03] [ 1.68076660e-02 1.43217323e-02 -3.78193613e-03]] [[ 1.61791705e-02 1.42808734e-02 3.86070576e-03] [-3.00799683e-03 -6.67624502e-03 9.96823888e-03] [ 4.11968853e-04 3.93140828e-03 7.18226330e-03]] [[-1.54953304e-04 3.95501731e-03 1.09885863e-04] [-7.29954429e-03 -3.11365305e-03 -1.62109733e-02] [ 2.18678871e-03 -7.86761940e-03 -6.87157619e-04]]]] ==================================================================================== layer_type: Convolution layer_id: 26 input_layer0: layer_id=24: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 128), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_2/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_2'") input_layer1_value: layer_id=25: [[[[ 1.15246866e-02 7.43661728e-03 7.11896876e-03] [ 7.58890202e-03 -2.35633668e-03 -8.38773791e-03] [-6.47206558e-03 -2.57088803e-03 4.21760976e-03]] [[-2.10882514e-03 1.72332674e-03 4.32542432e-03] [ 5.16221393e-03 -3.48487822e-03 8.78799241e-03] [ 1.10953236e-02 -2.69743544e-03 -5.15680760e-03]] [[-9.53451172e-03 -2.82265991e-03 -1.28406957e-02] [-6.40441896e-03 1.66948466e-03 1.14302104e-03] [-1.60199646e-02 1.40176741e-02 3.06645362e-03]] ... [[-1.16415592e-02 -1.57239903e-02 1.34947170e-02] [-6.02871226e-03 5.74577693e-03 1.67468265e-02] [ 1.61934774e-02 1.12581840e-02 -1.56301931e-02]] [[ 1.72171241e-03 5.09083329e-04 7.56164081e-03] [-1.08633321e-02 5.84405335e-03 4.17543389e-03] [-5.14384033e-03 3.85695486e-03 -2.30960012e-03]] [[-1.64202624e-03 -2.95853504e-04 -2.09731283e-03] [-8.23166966e-03 1.60538091e-03 -1.43687834e-03] [-1.74358822e-02 8.71890597e-03 3.04525113e-03]]] [[[-3.52834305e-03 8.29524593e-04 -1.74687840e-02] [-5.76481177e-03 -3.72369378e-03 -2.27376595e-02] [-4.75500245e-04 -5.82169555e-03 4.61884821e-03]] [[ 2.24112603e-03 3.43121192e-03 6.16211863e-03] [-3.50251235e-03 8.01661517e-03 -6.67249784e-03] [ 9.55544645e-04 -1.85027043e-03 6.02218369e-03]] [[ 7.11276382e-03 2.59197853e-03 -7.43014831e-03] [ 5.00726933e-03 1.10678859e-02 1.89633276e-02] [-1.32207265e-02 6.16477942e-03 8.06163065e-03]] ... [[ 8.65571015e-03 2.62502348e-03 -7.87869561e-03] [-1.51539699e-03 -5.36831981e-03 -5.87983057e-03] [-1.17096433e-03 2.84600956e-03 3.03928624e-03]] [[-5.81907621e-03 -6.98628463e-03 -1.57176238e-02] [-1.28897503e-02 -9.97134950e-03 4.45764605e-03] [ 7.65150553e-03 1.75617244e-02 1.35162361e-02]] [[-4.82016383e-03 7.70875532e-03 5.51228644e-03] [ 7.91868952e-04 1.96585758e-03 -1.30440970e-03] [ 5.33431815e-03 1.84788257e-02 8.61326512e-03]]] [[[ 1.84520311e-03 -4.86603752e-03 8.99988227e-03] [ 5.87106973e-04 -3.48790060e-03 1.37186693e-02] [-1.36622007e-03 3.64021561e-03 1.54278777e-03]] [[-1.20531647e-02 1.03384606e-03 1.98278972e-03] [-8.29347130e-03 -4.19950485e-03 5.80803305e-03] [-5.68737183e-03 -2.01798766e-03 2.46692309e-03]] [[-7.07189972e-03 -3.67173599e-03 -1.08191753e-02] [-1.36312274e-02 -1.26937637e-03 3.07450769e-04] [-2.08825176e-03 1.25343073e-02 7.23117590e-03]] ... [[-3.91222490e-03 1.08885001e-02 -3.35392798e-03] [-9.74506326e-03 4.65849414e-03 1.22959409e-02] [-1.31188892e-02 -9.98106902e-04 1.45298131e-02]] [[ 1.19308673e-03 8.52052215e-03 -3.76744638e-03] [-2.16646516e-03 -2.11952929e-03 -4.98862006e-03] [-5.20573184e-03 3.02712247e-03 -1.65641494e-03]] [[-2.46674358e-03 -4.52469056e-03 1.80805230e-03] [-9.14657430e-04 2.53731944e-03 -3.68000520e-03] [-1.71639584e-03 1.17295394e-02 1.04795853e-02]]] ... [[[-8.33850820e-03 -9.80686676e-03 -6.39130501e-03] [-2.79592979e-03 4.54802020e-03 -1.95195212e-03] [ 1.76193770e-02 1.31955920e-02 7.61975208e-03]] [[ 2.86094774e-03 -6.31535286e-03 3.92993167e-03] [-1.85613055e-03 -1.34525998e-02 -7.90349732e-04] [ 1.44972978e-02 6.75864378e-03 9.67936404e-03]] [[-5.10482397e-03 -9.92135890e-03 -2.94809206e-03] [-2.68083508e-03 -4.50535025e-03 -1.11167515e-02] [-1.00535003e-03 3.08036921e-04 2.88249669e-03]] ... [[ 5.93944546e-03 -2.50258553e-03 -4.41644434e-03] [-4.17348044e-03 5.67906210e-03 -2.43334845e-02] [ 1.35842925e-02 -5.01225377e-03 5.76310034e-04]] [[ 7.25751370e-03 -1.08321160e-02 -5.80613082e-03] [ 3.62618431e-03 3.57717089e-03 -4.92676068e-03] [ 6.23808848e-03 2.22781789e-04 1.07812905e-03]] [[-2.27104109e-02 2.54031015e-03 9.94869880e-03] [ 1.55787263e-02 3.73024563e-03 -2.17096843e-02] [ 7.28548318e-03 6.24256535e-03 -1.25676015e-05]]] [[[ 1.30262226e-03 6.73383661e-03 -2.17128024e-02] [ 5.00287209e-03 7.96902925e-03 4.48334403e-03] [-1.13786934e-02 9.78915254e-04 -9.13041737e-03]] [[-1.84491500e-02 9.65074822e-03 2.66267732e-03] [ 2.43384275e-03 -3.57573875e-03 7.77664501e-03] [-1.05960146e-02 -5.73859084e-03 5.31001063e-03]] [[-8.15782044e-03 -3.53085902e-03 -6.23842003e-03] [-1.22496728e-02 -8.07665847e-03 5.01510128e-03] [-3.10147647e-04 2.26028799e-03 3.77046899e-03]] ... [[-9.19859798e-04 -7.61793088e-03 -7.89463799e-03] [ 5.17135311e-04 -2.78467555e-02 -3.25798281e-02] [-5.69814211e-03 -3.73197696e-03 9.15166270e-03]] [[-1.22190651e-03 -2.61003245e-03 6.19929691e-04] [-4.82921489e-03 -2.35801525e-02 1.39521845e-02] [-1.59817096e-02 3.11788009e-03 -9.53365304e-03]] [[ 7.78590562e-03 4.20127704e-04 -1.63016394e-02] [ 2.69251666e-03 2.19487422e-03 8.89003204e-05] [ 1.40868602e-02 4.79381159e-03 2.79590534e-03]]] [[[ 5.56306506e-04 6.19302457e-03 6.42550923e-03] [-9.78724193e-03 4.22338396e-03 -4.49284865e-03] [ 8.32202422e-05 -2.68504722e-04 4.93027689e-03]] [[ 1.52437240e-02 -1.07793538e-02 5.48798870e-03] [ 5.49978670e-03 1.65854144e-04 7.72294938e-04] [-1.82602240e-03 2.22187908e-03 7.80849997e-03]] [[-1.49104223e-02 -1.12586049e-02 -1.38134323e-02] [-1.15399482e-02 -8.20516609e-03 -7.96749163e-03] [-1.71724660e-03 -1.01414556e-02 2.98672845e-03]] ... [[ 9.98066738e-03 -6.14956813e-03 -9.54485498e-03] [ 7.70576473e-04 -6.31371571e-04 -2.53834436e-03] [ 1.68076660e-02 1.43217323e-02 -3.78193613e-03]] [[ 1.61791705e-02 1.42808734e-02 3.86070576e-03] [-3.00799683e-03 -6.67624502e-03 9.96823888e-03] [ 4.11968853e-04 3.93140828e-03 7.18226330e-03]] [[-1.54953304e-04 3.95501731e-03 1.09885863e-04] [-7.29954429e-03 -3.11365305e-03 -1.62109733e-02] [ 2.18678871e-03 -7.86761940e-03 -6.87157619e-04]]]] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='conv2d_4/Conv2D:0', description="created by layer 'conv2d_4'") ==================================================================================== layer_type: Const layer_id: 27 tf_layers_dict_shape: (1, 256, 1, 1) tf_layers_dict_value: [[[[ 6.71427250e-01]] [[ 5.52937746e-01]] [[ 7.66052127e-01]] [[-1.50864571e-01]] [[ 5.91544032e-01]] [[ 4.46107179e-01]] [[ 4.50249165e-01]] [[ 4.80484456e-01]] [[-8.48442316e-04]] [[ 3.19402277e-01]] [[ 3.37663144e-01]] [[ 3.21049660e-01]] [[ 4.25722212e-01]] [[ 4.89600152e-01]] [[ 6.00574017e-01]] [[ 5.77634931e-01]] [[ 4.58552420e-01]] [[ 5.90682447e-01]] [[ 5.37155509e-01]] [[ 8.26590061e-02]] [[ 2.87096173e-01]] [[ 4.57192779e-01]] [[ 5.01783609e-01]] [[ 4.94036257e-01]] [[ 2.98066676e-01]] [[ 3.79041255e-01]] [[ 2.70980716e-01]] [[ 9.19971764e-02]] [[ 6.48531795e-01]] [[ 4.73688096e-01]] [[ 2.33837843e-01]] [[ 6.99514031e-01]] [[ 4.39027071e-01]] [[ 4.23583329e-01]] [[ 5.00908732e-01]] [[ 4.63739812e-01]] [[ 4.72133547e-01]] [[ 3.79087120e-01]] [[ 5.55197954e-01]] [[ 6.80628717e-01]] [[ 4.86002147e-01]] [[ 7.07101762e-01]] [[ 5.32510459e-01]] [[ 4.52424765e-01]] [[ 4.07454908e-01]] [[ 2.27437794e-01]] [[ 5.30595601e-01]] [[ 6.15455866e-01]] [[ 3.79343510e-01]] [[ 5.80531001e-01]] [[ 5.48276126e-01]] [[ 1.84216946e-01]] [[ 4.57174629e-01]] [[ 5.78682184e-01]] [[ 3.10521096e-01]] [[ 4.82251018e-01]] [[ 3.65526140e-01]] [[ 5.82783699e-01]] [[-3.03257108e-01]] [[ 5.33525109e-01]] [[ 5.82204819e-01]] [[-7.40220547e-02]] [[ 1.01213843e-01]] [[ 4.05030102e-01]] [[ 5.16814113e-01]] [[ 3.75149101e-01]] [[ 1.91623747e-01]] [[ 2.33608648e-01]] [[ 4.92107809e-01]] [[ 5.48987329e-01]] [[ 1.73197269e-01]] [[ 6.76775753e-01]] [[ 4.28486198e-01]] [[ 2.87737876e-01]] [[ 1.24263674e-01]] [[ 4.00610954e-01]] [[ 7.62742281e-01]] [[ 3.53596479e-01]] [[ 4.72269297e-01]] [[ 5.76742589e-01]] [[ 5.00802159e-01]] [[ 2.98448503e-01]] [[ 3.78215253e-01]] [[-2.56993622e-01]] [[ 3.30837369e-01]] [[ 3.34707022e-01]] [[ 8.07584524e-02]] [[ 3.60012054e-01]] [[ 3.65185291e-01]] [[ 6.49379969e-01]] [[ 4.07869339e-01]] [[ 6.97051585e-01]] [[ 9.31154430e-01]] [[ 6.04052424e-01]] [[-2.15828598e-01]] [[ 5.12393534e-01]] [[-1.14431933e-01]] [[ 1.92788869e-01]] [[ 2.21520126e-01]] [[ 2.12098777e-01]] [[ 5.42933583e-01]] [[ 3.56970638e-01]] [[ 1.61276549e-01]] [[ 3.74990493e-01]] [[ 4.43385035e-01]] [[ 5.04419148e-01]] [[ 4.67418939e-01]] [[ 5.02752125e-01]] [[ 4.39406872e-01]] [[ 6.62869930e-01]] [[ 2.96492726e-01]] [[ 1.68411374e-01]] [[ 3.07748020e-01]] [[ 3.50788057e-01]] [[ 5.01036882e-01]] [[ 5.94175339e-01]] [[ 4.53846782e-01]] [[ 2.15376019e-01]] [[ 1.08815730e-02]] [[ 4.31263208e-01]] [[ 5.08058608e-01]] [[ 5.20691216e-01]] [[ 6.31496787e-01]] [[ 2.84592330e-01]] [[-1.13850236e-02]] [[-3.13930362e-01]] [[ 5.67701817e-01]] [[ 7.83751667e-01]] [[ 3.11678886e-01]] [[ 5.66307783e-01]] [[ 3.57379943e-01]] [[ 2.53745079e-01]] [[ 9.71183479e-02]] [[ 4.95564193e-01]] [[ 7.61505783e-01]] [[ 3.66883636e-01]] [[ 2.20622420e-02]] [[ 6.75311446e-01]] [[-1.54321790e-02]] [[ 7.19890058e-01]] [[ 3.36594164e-01]] [[ 3.18491846e-01]] [[ 3.49677861e-01]] [[ 5.74084044e-01]] [[ 3.79650801e-01]] [[ 2.74990588e-01]] [[ 3.60303402e-01]] [[ 5.34847796e-01]] [[ 4.85372007e-01]] [[ 6.71505451e-01]] [[ 1.25199676e-01]] [[ 5.72630048e-01]] [[ 4.40211296e-01]] [[ 5.39865613e-01]] [[ 5.93603194e-01]] [[ 8.80353153e-02]] [[ 2.82854348e-01]] [[ 5.93490958e-01]] [[ 2.87100226e-01]] [[-2.20596179e-01]] [[ 2.88955331e-01]] [[ 1.35036469e-01]] [[ 3.71361554e-01]] [[ 6.36474311e-01]] [[ 5.57818770e-01]] [[ 9.85127091e-02]] [[-2.89166182e-01]] [[ 2.19927639e-01]] [[-3.18051904e-01]] [[ 3.38902473e-01]] [[-9.67534035e-02]] [[ 1.58640265e-01]] [[ 5.45803666e-01]] [[ 4.92637992e-01]] [[ 4.36586648e-01]] [[ 5.27272522e-01]] [[ 4.93104070e-01]] [[ 3.83852094e-01]] [[ 7.15311885e-01]] [[ 4.72791821e-01]] [[ 4.11290467e-01]] [[ 5.72395444e-01]] [[ 4.77737874e-01]] [[ 1.35244697e-01]] [[ 5.73922634e-01]] [[ 4.34546858e-01]] [[ 5.07689655e-01]] [[ 5.15947700e-01]] [[ 6.23539090e-01]] [[ 5.36686420e-01]] [[ 4.98249352e-01]] [[ 4.62986529e-01]] [[ 4.39290196e-01]] [[ 4.41972315e-01]] [[ 7.17141747e-01]] [[ 6.06223226e-01]] [[ 3.14870119e-01]] [[ 1.92293435e-01]] [[ 7.47385085e-01]] [[ 4.24728990e-02]] [[ 9.38877165e-02]] [[ 3.30547333e-01]] [[ 4.03664708e-01]] [[ 2.01823413e-02]] [[ 4.18795347e-01]] [[ 6.45746827e-01]] [[ 1.40906930e-01]] [[ 7.52854824e-01]] [[ 2.85072625e-02]] [[-7.71620274e-02]] [[ 4.54542756e-01]] [[ 6.59634471e-01]] [[ 5.30037165e-01]] [[ 2.24474728e-01]] [[ 6.91716671e-01]] [[ 5.89040399e-01]] [[ 6.85806513e-01]] [[ 4.86260533e-01]] [[ 4.85488415e-01]] [[ 7.51849353e-01]] [[ 6.29359841e-01]] [[ 6.53311372e-01]] [[ 2.12384239e-01]] [[ 4.15704250e-01]] [[ 5.46333909e-01]] [[ 6.56535864e-01]] [[ 6.38723493e-01]] [[ 3.51629227e-01]] [[ 1.08048886e-01]] [[ 5.94680786e-01]] [[ 5.63277602e-01]] [[ 4.65030909e-01]] [[ 4.27132368e-01]] [[ 9.39056575e-02]] [[ 5.57781219e-01]] [[ 5.86885571e-01]] [[ 5.23915172e-01]] [[ 2.76289344e-01]] [[ 3.26565951e-01]] [[ 4.41352129e-01]] [[ 5.23489118e-01]] [[ 5.61501801e-01]] [[ 2.10143775e-01]] [[ 6.14307165e-01]] [[ 7.57386088e-01]] [[ 3.36242080e-01]] [[ 3.05433810e-01]] [[ 3.85679901e-01]] [[ 4.88582820e-01]] [[ 4.55188811e-01]] [[ 4.19952303e-01]] [[ 4.86803889e-01]] [[ 5.12055755e-01]] [[ 6.04412675e-01]] [[ 2.47175395e-01]] [[ 6.71081245e-01]]]] ==================================================================================== layer_type: Add layer_id: 28 input_layer0: layer_id=26: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='conv2d_4/Conv2D:0', description="created by layer 'conv2d_4'") input_layer1_value: layer_id=27: [[[[ 6.71427250e-01]] [[ 5.52937746e-01]] [[ 7.66052127e-01]] [[-1.50864571e-01]] [[ 5.91544032e-01]] [[ 4.46107179e-01]] [[ 4.50249165e-01]] [[ 4.80484456e-01]] [[-8.48442316e-04]] [[ 3.19402277e-01]] [[ 3.37663144e-01]] [[ 3.21049660e-01]] [[ 4.25722212e-01]] [[ 4.89600152e-01]] [[ 6.00574017e-01]] [[ 5.77634931e-01]] [[ 4.58552420e-01]] [[ 5.90682447e-01]] [[ 5.37155509e-01]] [[ 8.26590061e-02]] [[ 2.87096173e-01]] [[ 4.57192779e-01]] [[ 5.01783609e-01]] [[ 4.94036257e-01]] [[ 2.98066676e-01]] [[ 3.79041255e-01]] [[ 2.70980716e-01]] [[ 9.19971764e-02]] [[ 6.48531795e-01]] [[ 4.73688096e-01]] [[ 2.33837843e-01]] [[ 6.99514031e-01]] [[ 4.39027071e-01]] [[ 4.23583329e-01]] [[ 5.00908732e-01]] [[ 4.63739812e-01]] [[ 4.72133547e-01]] [[ 3.79087120e-01]] [[ 5.55197954e-01]] [[ 6.80628717e-01]] [[ 4.86002147e-01]] [[ 7.07101762e-01]] [[ 5.32510459e-01]] [[ 4.52424765e-01]] [[ 4.07454908e-01]] [[ 2.27437794e-01]] [[ 5.30595601e-01]] [[ 6.15455866e-01]] [[ 3.79343510e-01]] [[ 5.80531001e-01]] [[ 5.48276126e-01]] [[ 1.84216946e-01]] [[ 4.57174629e-01]] [[ 5.78682184e-01]] [[ 3.10521096e-01]] [[ 4.82251018e-01]] [[ 3.65526140e-01]] [[ 5.82783699e-01]] [[-3.03257108e-01]] [[ 5.33525109e-01]] [[ 5.82204819e-01]] [[-7.40220547e-02]] [[ 1.01213843e-01]] [[ 4.05030102e-01]] [[ 5.16814113e-01]] [[ 3.75149101e-01]] [[ 1.91623747e-01]] [[ 2.33608648e-01]] [[ 4.92107809e-01]] [[ 5.48987329e-01]] [[ 1.73197269e-01]] [[ 6.76775753e-01]] [[ 4.28486198e-01]] [[ 2.87737876e-01]] [[ 1.24263674e-01]] [[ 4.00610954e-01]] [[ 7.62742281e-01]] [[ 3.53596479e-01]] [[ 4.72269297e-01]] [[ 5.76742589e-01]] [[ 5.00802159e-01]] [[ 2.98448503e-01]] [[ 3.78215253e-01]] [[-2.56993622e-01]] [[ 3.30837369e-01]] [[ 3.34707022e-01]] [[ 8.07584524e-02]] [[ 3.60012054e-01]] [[ 3.65185291e-01]] [[ 6.49379969e-01]] [[ 4.07869339e-01]] [[ 6.97051585e-01]] [[ 9.31154430e-01]] [[ 6.04052424e-01]] [[-2.15828598e-01]] [[ 5.12393534e-01]] [[-1.14431933e-01]] [[ 1.92788869e-01]] [[ 2.21520126e-01]] [[ 2.12098777e-01]] [[ 5.42933583e-01]] [[ 3.56970638e-01]] [[ 1.61276549e-01]] [[ 3.74990493e-01]] [[ 4.43385035e-01]] [[ 5.04419148e-01]] [[ 4.67418939e-01]] [[ 5.02752125e-01]] [[ 4.39406872e-01]] [[ 6.62869930e-01]] [[ 2.96492726e-01]] [[ 1.68411374e-01]] [[ 3.07748020e-01]] [[ 3.50788057e-01]] [[ 5.01036882e-01]] [[ 5.94175339e-01]] [[ 4.53846782e-01]] [[ 2.15376019e-01]] [[ 1.08815730e-02]] [[ 4.31263208e-01]] [[ 5.08058608e-01]] [[ 5.20691216e-01]] [[ 6.31496787e-01]] [[ 2.84592330e-01]] [[-1.13850236e-02]] [[-3.13930362e-01]] [[ 5.67701817e-01]] [[ 7.83751667e-01]] [[ 3.11678886e-01]] [[ 5.66307783e-01]] [[ 3.57379943e-01]] [[ 2.53745079e-01]] [[ 9.71183479e-02]] [[ 4.95564193e-01]] [[ 7.61505783e-01]] [[ 3.66883636e-01]] [[ 2.20622420e-02]] [[ 6.75311446e-01]] [[-1.54321790e-02]] [[ 7.19890058e-01]] [[ 3.36594164e-01]] [[ 3.18491846e-01]] [[ 3.49677861e-01]] [[ 5.74084044e-01]] [[ 3.79650801e-01]] [[ 2.74990588e-01]] [[ 3.60303402e-01]] [[ 5.34847796e-01]] [[ 4.85372007e-01]] [[ 6.71505451e-01]] [[ 1.25199676e-01]] [[ 5.72630048e-01]] [[ 4.40211296e-01]] [[ 5.39865613e-01]] [[ 5.93603194e-01]] [[ 8.80353153e-02]] [[ 2.82854348e-01]] [[ 5.93490958e-01]] [[ 2.87100226e-01]] [[-2.20596179e-01]] [[ 2.88955331e-01]] [[ 1.35036469e-01]] [[ 3.71361554e-01]] [[ 6.36474311e-01]] [[ 5.57818770e-01]] [[ 9.85127091e-02]] [[-2.89166182e-01]] [[ 2.19927639e-01]] [[-3.18051904e-01]] [[ 3.38902473e-01]] [[-9.67534035e-02]] [[ 1.58640265e-01]] [[ 5.45803666e-01]] [[ 4.92637992e-01]] [[ 4.36586648e-01]] [[ 5.27272522e-01]] [[ 4.93104070e-01]] [[ 3.83852094e-01]] [[ 7.15311885e-01]] [[ 4.72791821e-01]] [[ 4.11290467e-01]] [[ 5.72395444e-01]] [[ 4.77737874e-01]] [[ 1.35244697e-01]] [[ 5.73922634e-01]] [[ 4.34546858e-01]] [[ 5.07689655e-01]] [[ 5.15947700e-01]] [[ 6.23539090e-01]] [[ 5.36686420e-01]] [[ 4.98249352e-01]] [[ 4.62986529e-01]] [[ 4.39290196e-01]] [[ 4.41972315e-01]] [[ 7.17141747e-01]] [[ 6.06223226e-01]] [[ 3.14870119e-01]] [[ 1.92293435e-01]] [[ 7.47385085e-01]] [[ 4.24728990e-02]] [[ 9.38877165e-02]] [[ 3.30547333e-01]] [[ 4.03664708e-01]] [[ 2.01823413e-02]] [[ 4.18795347e-01]] [[ 6.45746827e-01]] [[ 1.40906930e-01]] [[ 7.52854824e-01]] [[ 2.85072625e-02]] [[-7.71620274e-02]] [[ 4.54542756e-01]] [[ 6.59634471e-01]] [[ 5.30037165e-01]] [[ 2.24474728e-01]] [[ 6.91716671e-01]] [[ 5.89040399e-01]] [[ 6.85806513e-01]] [[ 4.86260533e-01]] [[ 4.85488415e-01]] [[ 7.51849353e-01]] [[ 6.29359841e-01]] [[ 6.53311372e-01]] [[ 2.12384239e-01]] [[ 4.15704250e-01]] [[ 5.46333909e-01]] [[ 6.56535864e-01]] [[ 6.38723493e-01]] [[ 3.51629227e-01]] [[ 1.08048886e-01]] [[ 5.94680786e-01]] [[ 5.63277602e-01]] [[ 4.65030909e-01]] [[ 4.27132368e-01]] [[ 9.39056575e-02]] [[ 5.57781219e-01]] [[ 5.86885571e-01]] [[ 5.23915172e-01]] [[ 2.76289344e-01]] [[ 3.26565951e-01]] [[ 4.41352129e-01]] [[ 5.23489118e-01]] [[ 5.61501801e-01]] [[ 2.10143775e-01]] [[ 6.14307165e-01]] [[ 7.57386088e-01]] [[ 3.36242080e-01]] [[ 3.05433810e-01]] [[ 3.85679901e-01]] [[ 4.88582820e-01]] [[ 4.55188811e-01]] [[ 4.19952303e-01]] [[ 4.86803889e-01]] [[ 5.12055755e-01]] [[ 6.04412675e-01]] [[ 2.47175395e-01]] [[ 6.71081245e-01]]]] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.math.add_4/Add:0', description="created by layer 'tf.math.add_4'") ==================================================================================== layer_type: ReLU layer_id: 29 input_layer0: layer_id=28: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.math.add_4/Add:0', description="created by layer 'tf.math.add_4'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.nn.relu_4/Relu:0', description="created by layer 'tf.nn.relu_4'") ==================================================================================== layer_type: Const layer_id: 30 tf_layers_dict_shape: (256, 256, 3, 3) tf_layers_dict_value: [[[[-0.00817604 -0.02044623 -0.01221118] [ 0.02048556 -0.05705471 -0.05337912] [ 0.07299346 -0.03238072 0.01427441]] [[-0.04525789 -0.01239105 0.02669756] [-0.02396949 -0.07232359 -0.01619774] [-0.01867786 0.03293019 0.02050117]] [[ 0.01190231 -0.0418215 0.01035193] [-0.01371272 0.01064188 0.0435654 ] [-0.00409764 0.01910046 0.00585798]] ... [[ 0.01318217 0.03015503 0.02078221] [ 0.01455614 0.02829458 0.00480922] [ 0.01947073 -0.00829352 0.0001441 ]] [[-0.06044754 -0.00836264 0.00945898] [-0.05594178 -0.00743945 0.06417853] [-0.01841389 0.01305482 0.00404568]] [[ 0.00411236 -0.0294481 -0.04347574] [-0.04733609 -0.00383368 0.0010358 ] [-0.07031596 -0.01355635 -0.02696834]]] [[[-0.0015259 -0.01570036 0.00339301] [ 0.01451091 0.02548653 -0.00564921] [-0.0417646 0.00587825 -0.01117164]] [[ 0.01301596 -0.05447353 0.00393108] [ 0.00916963 0.02012841 0.02646472] [ 0.02432137 -0.02059802 0.04612806]] [[-0.02037351 0.02127605 0.05511075] [ 0.02101016 0.07367793 0.0166705 ] [-0.01516519 0.02109121 0.02679432]] ... [[ 0.00605948 0.01777631 0.00116807] [ 0.01712194 0.02424101 0.01442361] [-0.02268632 -0.02986166 -0.0124677 ]] [[-0.04928539 0.0139828 0.03449006] [-0.00910583 -0.03168073 0.01045519] [ 0.04498566 0.00275637 0.01457907]] [[-0.00728058 0.01953195 0.00334412] [-0.007175 0.00166223 0.02432836] [ 0.00777044 -0.01479382 -0.02920663]]] [[[ 0.01072372 0.04171617 0.0524292 ] [ 0.0012758 0.00842071 0.08477137] [-0.01212861 -0.00933968 0.02275392]] [[ 0.00667596 0.03612684 0.00620081] [ 0.00856644 -0.00285489 0.00468059] [ 0.03415949 0.01993636 0.03889169]] [[ 0.0083022 -0.01025522 -0.00810084] [-0.04195455 0.00945167 0.00775856] [-0.03982546 -0.01682907 -0.00500113]] ... [[ 0.0523683 -0.02143776 -0.00027111] [ 0.01555412 -0.02447165 0.00816779] [ 0.03409537 -0.00600717 0.01020366]] [[-0.04364026 0.03854291 0.09061941] [ 0.01618193 0.03114691 0.01665231] [ 0.04106982 0.01376748 -0.00913415]] [[ 0.04745869 -0.01693473 0.00735908] [-0.01953194 0.01827185 -0.02144995] [ 0.04200162 -0.00051357 -0.02946531]]] ... [[[ 0.06243844 0.00379756 0.02622754] [ 0.01788783 0.01755145 -0.0121094 ] [-0.0734148 -0.06418753 -0.00187477]] [[-0.01769352 0.01247541 0.04797258] [-0.05271181 -0.0261347 0.01498156] [-0.02252002 -0.01014182 -0.02938397]] [[ 0.0169919 0.04406522 0.01435214] [ 0.00633042 -0.01140661 -0.01721627] [ 0.00665837 -0.01266269 0.00010574]] ... [[-0.0606346 -0.06941258 -0.02495186] [-0.00402789 -0.01831277 0.01157964] [ 0.03419735 0.04042831 0.02629922]] [[-0.01655724 -0.07594325 -0.03325762] [ 0.05151952 0.02892669 -0.05346958] [ 0.04886706 0.05178685 -0.01046246]] [[-0.01669818 -0.06239357 -0.03665224] [ 0.02598093 -0.00367708 0.01141081] [ 0.0098105 -0.01836678 -0.02678548]]] [[[-0.03331951 -0.03929156 0.05841409] [-0.09403543 0.01656366 0.03241038] [ 0.00199312 0.01298139 0.01849559]] [[-0.00221099 -0.00491926 -0.03490087] [ 0.01806375 0.04517895 0.04379812] [-0.03533578 0.00619991 0.00970466]] [[-0.02453962 -0.00173261 -0.05616616] [-0.04846562 0.04943664 0.04187672] [-0.03179381 0.0018821 0.01213892]] ... [[ 0.01174229 0.01859181 -0.06749402] [ 0.06837579 0.02825802 -0.05569838] [ 0.01312342 -0.03074685 -0.02067011]] [[-0.03200166 -0.0430546 0.0323187 ] [ 0.00073249 0.04585096 0.01304206] [-0.00312355 0.06171969 0.05546439]] [[-0.01677481 0.0159171 -0.00532877] [ 0.02343283 0.04526766 0.00491914] [-0.00574355 0.00431968 -0.00480968]]] [[[ 0.06301296 -0.01881102 0.00636145] [-0.01019723 0.01928155 0.01259715] [-0.02451075 -0.00291171 0.01900369]] [[ 0.03026861 -0.00239516 -0.0315201 ] [-0.06779017 -0.02392449 -0.03287339] [-0.01157447 0.00406421 0.0088893 ]] [[ 0.0163357 -0.0104907 -0.02517396] [ 0.01861256 -0.00419083 -0.04265951] [ 0.04121919 0.01102744 -0.02983104]] ... [[ 0.00696781 -0.01912308 0.02528142] [-0.02033334 0.02391466 0.01993305] [-0.01228632 0.00636031 0.03949639]] [[ 0.01490704 0.00737205 0.00534889] [ 0.02859532 -0.02592338 -0.04232856] [-0.03505002 0.02074432 -0.0141375 ]] [[-0.02699802 -0.0072087 0.02267146] [ 0.00468355 0.00598089 0.00386121] [ 0.00638918 -0.00954374 -0.01687566]]]] ==================================================================================== layer_type: Convolution layer_id: 31 input_layer0: layer_id=29: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.nn.relu_4/Relu:0', description="created by layer 'tf.nn.relu_4'") input_layer1_value: layer_id=30: [[[[-0.00817604 -0.02044623 -0.01221118] [ 0.02048556 -0.05705471 -0.05337912] [ 0.07299346 -0.03238072 0.01427441]] [[-0.04525789 -0.01239105 0.02669756] [-0.02396949 -0.07232359 -0.01619774] [-0.01867786 0.03293019 0.02050117]] [[ 0.01190231 -0.0418215 0.01035193] [-0.01371272 0.01064188 0.0435654 ] [-0.00409764 0.01910046 0.00585798]] ... [[ 0.01318217 0.03015503 0.02078221] [ 0.01455614 0.02829458 0.00480922] [ 0.01947073 -0.00829352 0.0001441 ]] [[-0.06044754 -0.00836264 0.00945898] [-0.05594178 -0.00743945 0.06417853] [-0.01841389 0.01305482 0.00404568]] [[ 0.00411236 -0.0294481 -0.04347574] [-0.04733609 -0.00383368 0.0010358 ] [-0.07031596 -0.01355635 -0.02696834]]] [[[-0.0015259 -0.01570036 0.00339301] [ 0.01451091 0.02548653 -0.00564921] [-0.0417646 0.00587825 -0.01117164]] [[ 0.01301596 -0.05447353 0.00393108] [ 0.00916963 0.02012841 0.02646472] [ 0.02432137 -0.02059802 0.04612806]] [[-0.02037351 0.02127605 0.05511075] [ 0.02101016 0.07367793 0.0166705 ] [-0.01516519 0.02109121 0.02679432]] ... [[ 0.00605948 0.01777631 0.00116807] [ 0.01712194 0.02424101 0.01442361] [-0.02268632 -0.02986166 -0.0124677 ]] [[-0.04928539 0.0139828 0.03449006] [-0.00910583 -0.03168073 0.01045519] [ 0.04498566 0.00275637 0.01457907]] [[-0.00728058 0.01953195 0.00334412] [-0.007175 0.00166223 0.02432836] [ 0.00777044 -0.01479382 -0.02920663]]] [[[ 0.01072372 0.04171617 0.0524292 ] [ 0.0012758 0.00842071 0.08477137] [-0.01212861 -0.00933968 0.02275392]] [[ 0.00667596 0.03612684 0.00620081] [ 0.00856644 -0.00285489 0.00468059] [ 0.03415949 0.01993636 0.03889169]] [[ 0.0083022 -0.01025522 -0.00810084] [-0.04195455 0.00945167 0.00775856] [-0.03982546 -0.01682907 -0.00500113]] ... [[ 0.0523683 -0.02143776 -0.00027111] [ 0.01555412 -0.02447165 0.00816779] [ 0.03409537 -0.00600717 0.01020366]] [[-0.04364026 0.03854291 0.09061941] [ 0.01618193 0.03114691 0.01665231] [ 0.04106982 0.01376748 -0.00913415]] [[ 0.04745869 -0.01693473 0.00735908] [-0.01953194 0.01827185 -0.02144995] [ 0.04200162 -0.00051357 -0.02946531]]] ... [[[ 0.06243844 0.00379756 0.02622754] [ 0.01788783 0.01755145 -0.0121094 ] [-0.0734148 -0.06418753 -0.00187477]] [[-0.01769352 0.01247541 0.04797258] [-0.05271181 -0.0261347 0.01498156] [-0.02252002 -0.01014182 -0.02938397]] [[ 0.0169919 0.04406522 0.01435214] [ 0.00633042 -0.01140661 -0.01721627] [ 0.00665837 -0.01266269 0.00010574]] ... [[-0.0606346 -0.06941258 -0.02495186] [-0.00402789 -0.01831277 0.01157964] [ 0.03419735 0.04042831 0.02629922]] [[-0.01655724 -0.07594325 -0.03325762] [ 0.05151952 0.02892669 -0.05346958] [ 0.04886706 0.05178685 -0.01046246]] [[-0.01669818 -0.06239357 -0.03665224] [ 0.02598093 -0.00367708 0.01141081] [ 0.0098105 -0.01836678 -0.02678548]]] [[[-0.03331951 -0.03929156 0.05841409] [-0.09403543 0.01656366 0.03241038] [ 0.00199312 0.01298139 0.01849559]] [[-0.00221099 -0.00491926 -0.03490087] [ 0.01806375 0.04517895 0.04379812] [-0.03533578 0.00619991 0.00970466]] [[-0.02453962 -0.00173261 -0.05616616] [-0.04846562 0.04943664 0.04187672] [-0.03179381 0.0018821 0.01213892]] ... [[ 0.01174229 0.01859181 -0.06749402] [ 0.06837579 0.02825802 -0.05569838] [ 0.01312342 -0.03074685 -0.02067011]] [[-0.03200166 -0.0430546 0.0323187 ] [ 0.00073249 0.04585096 0.01304206] [-0.00312355 0.06171969 0.05546439]] [[-0.01677481 0.0159171 -0.00532877] [ 0.02343283 0.04526766 0.00491914] [-0.00574355 0.00431968 -0.00480968]]] [[[ 0.06301296 -0.01881102 0.00636145] [-0.01019723 0.01928155 0.01259715] [-0.02451075 -0.00291171 0.01900369]] [[ 0.03026861 -0.00239516 -0.0315201 ] [-0.06779017 -0.02392449 -0.03287339] [-0.01157447 0.00406421 0.0088893 ]] [[ 0.0163357 -0.0104907 -0.02517396] [ 0.01861256 -0.00419083 -0.04265951] [ 0.04121919 0.01102744 -0.02983104]] ... [[ 0.00696781 -0.01912308 0.02528142] [-0.02033334 0.02391466 0.01993305] [-0.01228632 0.00636031 0.03949639]] [[ 0.01490704 0.00737205 0.00534889] [ 0.02859532 -0.02592338 -0.04232856] [-0.03505002 0.02074432 -0.0141375 ]] [[-0.02699802 -0.0072087 0.02267146] [ 0.00468355 0.00598089 0.00386121] [ 0.00638918 -0.00954374 -0.01687566]]]] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='conv2d_5/Conv2D:0', description="created by layer 'conv2d_5'") ==================================================================================== layer_type: Const layer_id: 32 tf_layers_dict_shape: (1, 256, 1, 1) tf_layers_dict_value: [[[[-0.66215795]] [[-1.0793598 ]] [[-0.12864324]] [[-0.09210221]] [[-0.02339157]] [[-0.7677838 ]] [[-0.05400363]] [[-0.3274841 ]] [[-0.6937257 ]] [[ 0.19219849]] [[-0.33115944]] [[-0.8551749 ]] [[-0.17466651]] [[-0.8751474 ]] [[-0.4032056 ]] [[ 0.55652404]] [[-0.29099107]] [[-0.21928775]] [[ 0.08200642]] [[-0.61198664]] [[-0.03039339]] [[ 0.05778658]] [[ 0.11089791]] [[ 0.01220737]] [[ 0.4807411 ]] [[-0.29699877]] [[-0.92438686]] [[-0.12716845]] [[-0.0272046 ]] [[ 0.3471931 ]] [[-0.44590586]] [[-0.5044377 ]] [[-0.11681929]] [[-0.07705443]] [[-0.62645006]] [[-0.56179535]] [[-1.3112849 ]] [[-1.1626178 ]] [[-0.38106027]] [[-0.5801429 ]] [[ 0.0471943 ]] [[ 0.6954113 ]] [[-0.34389213]] [[ 0.16190329]] [[-0.42025024]] [[ 0.17143048]] [[-0.09482347]] [[-0.48655728]] [[-1.7620898 ]] [[-0.03739524]] [[-0.8312596 ]] [[-0.5451746 ]] [[-0.3605582 ]] [[ 0.18162513]] [[ 0.33696336]] [[-0.13899365]] [[-0.40744087]] [[ 0.19306198]] [[-0.28280184]] [[ 0.07937172]] [[-0.20320842]] [[-0.38661492]] [[ 0.15558816]] [[ 0.11203393]] [[-0.01244132]] [[ 0.18443158]] [[-0.50237256]] [[-0.39708316]] [[-0.03656378]] [[-0.2807185 ]] [[ 0.13246974]] [[-0.28341812]] [[-0.7360424 ]] [[-0.39937887]] [[-0.2970263 ]] [[-0.15176316]] [[ 0.08337802]] [[-0.72073454]] [[-0.17444229]] [[-0.11208168]] [[-0.0713919 ]] [[ 0.28157902]] [[-0.46663344]] [[ 0.45301664]] [[ 0.26187128]] [[ 0.27085602]] [[ 0.09305693]] [[-0.6595011 ]] [[-0.8165705 ]] [[ 0.52350837]] [[ 0.31793445]] [[-0.09421863]] [[-0.45570958]] [[-0.06857166]] [[-0.18960266]] [[-0.92799324]] [[-0.36780825]] [[-0.57595134]] [[-2.0125442 ]] [[ 0.05105141]] [[-0.11996806]] [[-1.0935168 ]] [[ 0.5372357 ]] [[-0.42254534]] [[ 0.16384278]] [[ 0.00566152]] [[-0.5909584 ]] [[-0.6134763 ]] [[-0.02606541]] [[ 0.17639846]] [[-0.5810833 ]] [[ 0.25826445]] [[-0.14115286]] [[-0.00874014]] [[-0.10728054]] [[-0.28780794]] [[-0.23496751]] [[-0.17677635]] [[-0.34425107]] [[-0.9891344 ]] [[-0.9870744 ]] [[-0.2222833 ]] [[-0.2234739 ]] [[-0.05937833]] [[ 0.20918357]] [[-0.27342367]] [[-0.6505356 ]] [[-0.00870612]] [[ 0.15128176]] [[-0.41350657]] [[-0.3022014 ]] [[ 0.22487459]] [[ 0.62535536]] [[-0.07823944]] [[-0.60741544]] [[-0.12613231]] [[-0.3340531 ]] [[-0.07123522]] [[-0.35885814]] [[-0.02129679]] [[ 0.01930089]] [[ 0.05729164]] [[-0.43685153]] [[-0.02345137]] [[-0.2367326 ]] [[-0.08966972]] [[ 0.147452 ]] [[-0.48264962]] [[ 0.28839582]] [[-0.28916034]] [[-0.01488267]] [[-0.5463208 ]] [[-0.94955844]] [[-0.5922339 ]] [[ 0.2248558 ]] [[ 0.25275943]] [[-0.08149786]] [[-0.3624608 ]] [[-1.1613955 ]] [[ 0.08401096]] [[ 0.05615947]] [[ 0.15570442]] [[-0.66929233]] [[ 0.51896876]] [[-0.7954865 ]] [[-0.6056318 ]] [[-0.07529856]] [[-0.29393137]] [[-0.18998793]] [[ 0.2909674 ]] [[-0.35796243]] [[-0.39805973]] [[ 0.00851597]] [[-0.30597454]] [[ 0.40186498]] [[-0.6756208 ]] [[-0.53287137]] [[-0.49242362]] [[ 0.49377888]] [[ 0.25625998]] [[-0.54078346]] [[-0.14233068]] [[-0.7901596 ]] [[-0.1687132 ]] [[ 0.24133795]] [[-0.3851242 ]] [[-0.3239624 ]] [[-0.13442121]] [[-0.1745669 ]] [[ 0.317284 ]] [[-0.2817645 ]] [[-0.02722016]] [[-0.2613833 ]] [[-1.432364 ]] [[ 0.2481656 ]] [[ 0.21661265]] [[-0.04091816]] [[-0.7128489 ]] [[-0.0957304 ]] [[-0.43171757]] [[-0.45417753]] [[-0.19659334]] [[ 0.24447757]] [[ 0.10078101]] [[-0.21091653]] [[ 0.22909662]] [[ 0.06636876]] [[-0.81062996]] [[ 0.2870627 ]] [[ 0.1527494 ]] [[-0.8694248 ]] [[ 0.03510076]] [[-0.10668851]] [[ 0.15578675]] [[ 0.05992652]] [[-1.310649 ]] [[ 0.09496081]] [[ 0.03324096]] [[ 0.20623878]] [[-0.5072021 ]] [[-0.38462675]] [[-0.47202256]] [[ 0.15750122]] [[-0.5838213 ]] [[-0.6653978 ]] [[-0.42114812]] [[-0.4718194 ]] [[ 0.0907151 ]] [[ 0.12052561]] [[ 0.19420159]] [[-0.30877304]] [[-0.40799406]] [[-0.88151276]] [[-0.18628035]] [[-0.23143287]] [[-0.45156175]] [[ 0.2308112 ]] [[-0.59444296]] [[-0.08405867]] [[-0.5345175 ]] [[ 0.09972532]] [[ 0.16032381]] [[-0.20137715]] [[-0.06198655]] [[-0.5098819 ]] [[ 0.06306146]] [[-0.42783558]] [[-0.8541188 ]] [[-0.4799066 ]] [[ 0.28303087]] [[-0.36655033]] [[ 0.6064741 ]] [[-0.12686744]] [[ 0.08627622]] [[-0.0202141 ]] [[-0.4763366 ]]]] ==================================================================================== layer_type: Add layer_id: 33 input_layer0: layer_id=31: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='conv2d_5/Conv2D:0', description="created by layer 'conv2d_5'") input_layer1_value: layer_id=32: [[[[-0.66215795]] [[-1.0793598 ]] [[-0.12864324]] [[-0.09210221]] [[-0.02339157]] [[-0.7677838 ]] [[-0.05400363]] [[-0.3274841 ]] [[-0.6937257 ]] [[ 0.19219849]] [[-0.33115944]] [[-0.8551749 ]] [[-0.17466651]] [[-0.8751474 ]] [[-0.4032056 ]] [[ 0.55652404]] [[-0.29099107]] [[-0.21928775]] [[ 0.08200642]] [[-0.61198664]] [[-0.03039339]] [[ 0.05778658]] [[ 0.11089791]] [[ 0.01220737]] [[ 0.4807411 ]] [[-0.29699877]] [[-0.92438686]] [[-0.12716845]] [[-0.0272046 ]] [[ 0.3471931 ]] [[-0.44590586]] [[-0.5044377 ]] [[-0.11681929]] [[-0.07705443]] [[-0.62645006]] [[-0.56179535]] [[-1.3112849 ]] [[-1.1626178 ]] [[-0.38106027]] [[-0.5801429 ]] [[ 0.0471943 ]] [[ 0.6954113 ]] [[-0.34389213]] [[ 0.16190329]] [[-0.42025024]] [[ 0.17143048]] [[-0.09482347]] [[-0.48655728]] [[-1.7620898 ]] [[-0.03739524]] [[-0.8312596 ]] [[-0.5451746 ]] [[-0.3605582 ]] [[ 0.18162513]] [[ 0.33696336]] [[-0.13899365]] [[-0.40744087]] [[ 0.19306198]] [[-0.28280184]] [[ 0.07937172]] [[-0.20320842]] [[-0.38661492]] [[ 0.15558816]] [[ 0.11203393]] [[-0.01244132]] [[ 0.18443158]] [[-0.50237256]] [[-0.39708316]] [[-0.03656378]] [[-0.2807185 ]] [[ 0.13246974]] [[-0.28341812]] [[-0.7360424 ]] [[-0.39937887]] [[-0.2970263 ]] [[-0.15176316]] [[ 0.08337802]] [[-0.72073454]] [[-0.17444229]] [[-0.11208168]] [[-0.0713919 ]] [[ 0.28157902]] [[-0.46663344]] [[ 0.45301664]] [[ 0.26187128]] [[ 0.27085602]] [[ 0.09305693]] [[-0.6595011 ]] [[-0.8165705 ]] [[ 0.52350837]] [[ 0.31793445]] [[-0.09421863]] [[-0.45570958]] [[-0.06857166]] [[-0.18960266]] [[-0.92799324]] [[-0.36780825]] [[-0.57595134]] [[-2.0125442 ]] [[ 0.05105141]] [[-0.11996806]] [[-1.0935168 ]] [[ 0.5372357 ]] [[-0.42254534]] [[ 0.16384278]] [[ 0.00566152]] [[-0.5909584 ]] [[-0.6134763 ]] [[-0.02606541]] [[ 0.17639846]] [[-0.5810833 ]] [[ 0.25826445]] [[-0.14115286]] [[-0.00874014]] [[-0.10728054]] [[-0.28780794]] [[-0.23496751]] [[-0.17677635]] [[-0.34425107]] [[-0.9891344 ]] [[-0.9870744 ]] [[-0.2222833 ]] [[-0.2234739 ]] [[-0.05937833]] [[ 0.20918357]] [[-0.27342367]] [[-0.6505356 ]] [[-0.00870612]] [[ 0.15128176]] [[-0.41350657]] [[-0.3022014 ]] [[ 0.22487459]] [[ 0.62535536]] [[-0.07823944]] [[-0.60741544]] [[-0.12613231]] [[-0.3340531 ]] [[-0.07123522]] [[-0.35885814]] [[-0.02129679]] [[ 0.01930089]] [[ 0.05729164]] [[-0.43685153]] [[-0.02345137]] [[-0.2367326 ]] [[-0.08966972]] [[ 0.147452 ]] [[-0.48264962]] [[ 0.28839582]] [[-0.28916034]] [[-0.01488267]] [[-0.5463208 ]] [[-0.94955844]] [[-0.5922339 ]] [[ 0.2248558 ]] [[ 0.25275943]] [[-0.08149786]] [[-0.3624608 ]] [[-1.1613955 ]] [[ 0.08401096]] [[ 0.05615947]] [[ 0.15570442]] [[-0.66929233]] [[ 0.51896876]] [[-0.7954865 ]] [[-0.6056318 ]] [[-0.07529856]] [[-0.29393137]] [[-0.18998793]] [[ 0.2909674 ]] [[-0.35796243]] [[-0.39805973]] [[ 0.00851597]] [[-0.30597454]] [[ 0.40186498]] [[-0.6756208 ]] [[-0.53287137]] [[-0.49242362]] [[ 0.49377888]] [[ 0.25625998]] [[-0.54078346]] [[-0.14233068]] [[-0.7901596 ]] [[-0.1687132 ]] [[ 0.24133795]] [[-0.3851242 ]] [[-0.3239624 ]] [[-0.13442121]] [[-0.1745669 ]] [[ 0.317284 ]] [[-0.2817645 ]] [[-0.02722016]] [[-0.2613833 ]] [[-1.432364 ]] [[ 0.2481656 ]] [[ 0.21661265]] [[-0.04091816]] [[-0.7128489 ]] [[-0.0957304 ]] [[-0.43171757]] [[-0.45417753]] [[-0.19659334]] [[ 0.24447757]] [[ 0.10078101]] [[-0.21091653]] [[ 0.22909662]] [[ 0.06636876]] [[-0.81062996]] [[ 0.2870627 ]] [[ 0.1527494 ]] [[-0.8694248 ]] [[ 0.03510076]] [[-0.10668851]] [[ 0.15578675]] [[ 0.05992652]] [[-1.310649 ]] [[ 0.09496081]] [[ 0.03324096]] [[ 0.20623878]] [[-0.5072021 ]] [[-0.38462675]] [[-0.47202256]] [[ 0.15750122]] [[-0.5838213 ]] [[-0.6653978 ]] [[-0.42114812]] [[-0.4718194 ]] [[ 0.0907151 ]] [[ 0.12052561]] [[ 0.19420159]] [[-0.30877304]] [[-0.40799406]] [[-0.88151276]] [[-0.18628035]] [[-0.23143287]] [[-0.45156175]] [[ 0.2308112 ]] [[-0.59444296]] [[-0.08405867]] [[-0.5345175 ]] [[ 0.09972532]] [[ 0.16032381]] [[-0.20137715]] [[-0.06198655]] [[-0.5098819 ]] [[ 0.06306146]] [[-0.42783558]] [[-0.8541188 ]] [[-0.4799066 ]] [[ 0.28303087]] [[-0.36655033]] [[ 0.6064741 ]] [[-0.12686744]] [[ 0.08627622]] [[-0.0202141 ]] [[-0.4763366 ]]]] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.math.add_5/Add:0', description="created by layer 'tf.math.add_5'") ==================================================================================== layer_type: ReLU layer_id: 34 input_layer0: layer_id=33: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.math.add_5/Add:0', description="created by layer 'tf.math.add_5'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.nn.relu_5/Relu:0', description="created by layer 'tf.nn.relu_5'") ==================================================================================== layer_type: MaxPool layer_id: 35 input_layer0: layer_id=34: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.nn.relu_5/Relu:0', description="created by layer 'tf.nn.relu_5'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 4, 150, 256), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_3/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_3'") ==================================================================================== layer_type: Const layer_id: 36 tf_layers_dict_shape: (256, 256, 2, 2) tf_layers_dict_value: [[[[ 0.01923482 -0.09797887] [-0.28427625 -0.14335105]] [[ 0.04385138 -0.04746195] [ 0.06992514 -0.00083265]] [[ 0.00282467 0.0843595 ] [-0.06250754 0.05707154]] ... [[ 0.16511755 0.04972154] [ 0.19439313 0.10729088]] [[-0.16178958 0.01322846] [-0.24459559 -0.09252056]] [[-0.08417284 -0.03044169] [ 0.05994251 0.2758846 ]]] [[[-0.11698069 -0.05545083] [-0.04084748 -0.11102765]] [[ 0.15095642 0.01255986] [-0.03171324 -0.00615657]] [[ 0.10236792 0.02094265] [ 0.01674829 -0.10912988]] ... [[ 0.11512588 0.14152887] [-0.01542834 0.06969558]] [[-0.07222787 -0.08151515] [-0.06663026 0.05762163]] [[ 0.05068195 -0.13608913] [-0.05490693 0.01509662]]] [[[ 0.06851256 -0.08606968] [-0.07073927 -0.09949067]] [[ 0.01312333 0.04129999] [ 0.08028405 -0.00921386]] [[-0.12406948 0.01902246] [-0.00384236 -0.01211822]] ... [[ 0.11748335 0.13427229] [ 0.06512318 -0.0342935 ]] [[ 0.01043487 -0.09739924] [-0.05052587 -0.09817957]] [[ 0.08511893 -0.02645148] [-0.02729066 -0.03341789]]] ... [[[-0.0237075 0.06058346] [-0.02711316 0.01534354]] [[-0.12176994 -0.11182908] [-0.19502413 -0.02609891]] [[-0.07080888 -0.10789613] [ 0.01132993 0.05115137]] ... [[-0.02952362 -0.05831409] [-0.0169778 0.03331037]] [[ 0.0360941 -0.06874772] [ 0.05930533 -0.0726686 ]] [[-0.05549059 -0.03690162] [-0.01944719 0.05420472]]] [[[ 0.04586334 0.09328008] [ 0.15339714 0.08325125]] [[ 0.10520578 0.08038849] [-0.02984812 0.00555401]] [[-0.08461875 -0.05512154] [-0.10090742 -0.11672983]] ... [[-0.07128152 0.02935277] [ 0.0612449 0.06607781]] [[-0.04948336 -0.04514969] [-0.01840428 -0.04389716]] [[-0.08382327 0.01705424] [-0.02602764 0.00435023]]] [[[-0.02969993 0.09264328] [-0.02397962 0.00969981]] [[-0.07154638 0.11642523] [-0.10606195 0.0796001 ]] [[ 0.03331683 -0.00915385] [ 0.03167675 0.03806659]] ... [[-0.0098352 0.13737035] [ 0.08187705 -0.00996816]] [[-0.0356164 0.02072238] [ 0.07538576 0.05897611]] [[ 0.03374712 0.1995614 ] [ 0.13652338 0.01229248]]]] ==================================================================================== layer_type: Convolution layer_id: 37 input_layer0: layer_id=35: KerasTensor(type_spec=TensorSpec(shape=(32, 4, 150, 256), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_3/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_3'") input_layer1_value: layer_id=36: [[[[ 0.01923482 -0.09797887] [-0.28427625 -0.14335105]] [[ 0.04385138 -0.04746195] [ 0.06992514 -0.00083265]] [[ 0.00282467 0.0843595 ] [-0.06250754 0.05707154]] ... [[ 0.16511755 0.04972154] [ 0.19439313 0.10729088]] [[-0.16178958 0.01322846] [-0.24459559 -0.09252056]] [[-0.08417284 -0.03044169] [ 0.05994251 0.2758846 ]]] [[[-0.11698069 -0.05545083] [-0.04084748 -0.11102765]] [[ 0.15095642 0.01255986] [-0.03171324 -0.00615657]] [[ 0.10236792 0.02094265] [ 0.01674829 -0.10912988]] ... [[ 0.11512588 0.14152887] [-0.01542834 0.06969558]] [[-0.07222787 -0.08151515] [-0.06663026 0.05762163]] [[ 0.05068195 -0.13608913] [-0.05490693 0.01509662]]] [[[ 0.06851256 -0.08606968] [-0.07073927 -0.09949067]] [[ 0.01312333 0.04129999] [ 0.08028405 -0.00921386]] [[-0.12406948 0.01902246] [-0.00384236 -0.01211822]] ... [[ 0.11748335 0.13427229] [ 0.06512318 -0.0342935 ]] [[ 0.01043487 -0.09739924] [-0.05052587 -0.09817957]] [[ 0.08511893 -0.02645148] [-0.02729066 -0.03341789]]] ... [[[-0.0237075 0.06058346] [-0.02711316 0.01534354]] [[-0.12176994 -0.11182908] [-0.19502413 -0.02609891]] [[-0.07080888 -0.10789613] [ 0.01132993 0.05115137]] ... [[-0.02952362 -0.05831409] [-0.0169778 0.03331037]] [[ 0.0360941 -0.06874772] [ 0.05930533 -0.0726686 ]] [[-0.05549059 -0.03690162] [-0.01944719 0.05420472]]] [[[ 0.04586334 0.09328008] [ 0.15339714 0.08325125]] [[ 0.10520578 0.08038849] [-0.02984812 0.00555401]] [[-0.08461875 -0.05512154] [-0.10090742 -0.11672983]] ... [[-0.07128152 0.02935277] [ 0.0612449 0.06607781]] [[-0.04948336 -0.04514969] [-0.01840428 -0.04389716]] [[-0.08382327 0.01705424] [-0.02602764 0.00435023]]] [[[-0.02969993 0.09264328] [-0.02397962 0.00969981]] [[-0.07154638 0.11642523] [-0.10606195 0.0796001 ]] [[ 0.03331683 -0.00915385] [ 0.03167675 0.03806659]] ... [[-0.0098352 0.13737035] [ 0.08187705 -0.00996816]] [[-0.0356164 0.02072238] [ 0.07538576 0.05897611]] [[ 0.03374712 0.1995614 ] [ 0.13652338 0.01229248]]]] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='conv2d_6/Conv2D:0', description="created by layer 'conv2d_6'") ==================================================================================== layer_type: Const layer_id: 38 tf_layers_dict_shape: (1, 256, 1, 1) tf_layers_dict_value: [[[[ 0.03531126]] [[ 0.04153406]] [[-0.01499302]] [[ 0.04892787]] [[ 0.10467708]] [[ 0.05344332]] [[ 0.09078331]] [[ 0.03810004]] [[ 0.09852754]] [[ 0.10886087]] [[ 0.0583949 ]] [[ 0.15019625]] [[ 0.12375277]] [[ 0.0836074 ]] [[ 0.05285143]] [[-0.01451365]] [[ 0.09832055]] [[ 0.05607849]] [[ 0.07759804]] [[ 0.07484994]] [[ 0.14392051]] [[ 0.12323733]] [[ 0.04834333]] [[ 0.04026539]] [[ 0.0206848 ]] [[ 0.0345908 ]] [[ 0.09243347]] [[ 0.04752847]] [[ 0.09716211]] [[ 0.02665334]] [[ 0.07137176]] [[ 0.06183958]] [[ 0.05770827]] [[ 0.04015264]] [[ 0.05639784]] [[ 0.05093274]] [[ 0.06614189]] [[ 0.04954747]] [[ 0.01887337]] [[ 0.08641648]] [[ 0.03211017]] [[ 0.13757695]] [[ 0.02274888]] [[ 0.01190568]] [[ 0.02532989]] [[ 0.0563163 ]] [[ 0.04885947]] [[ 0.06898033]] [[ 0.01896106]] [[ 0.10011526]] [[ 0.05837568]] [[ 0.06790939]] [[ 0.09529823]] [[-0.0412923 ]] [[ 0.13273457]] [[ 0.07322548]] [[ 0.09249222]] [[ 0.04539789]] [[ 0.08955366]] [[ 0.03735309]] [[ 0.04273967]] [[ 0.09724385]] [[ 0.01885115]] [[ 0.00743112]] [[ 0.03441679]] [[ 0.01052186]] [[-0.00418794]] [[ 0.04830207]] [[ 0.09452251]] [[ 0.09627661]] [[ 0.05013204]] [[ 0.13239259]] [[ 0.1103457 ]] [[ 0.04369677]] [[ 0.01661588]] [[ 0.05327062]] [[ 0.06146394]] [[ 0.05842484]] [[ 0.03206206]] [[ 0.10954275]] [[ 0.0323467 ]] [[ 0.05475442]] [[ 0.10349695]] [[ 0.08188834]] [[ 0.04716377]] [[ 0.07654594]] [[ 0.08036467]] [[ 0.08638818]] [[ 0.08552508]] [[ 0.10098928]] [[ 0.05385393]] [[ 0.01202287]] [[ 0.06818848]] [[ 0.07115225]] [[-0.00766839]] [[-0.0496178 ]] [[ 0.05881699]] [[ 0.05345987]] [[ 0.11750553]] [[ 0.09980227]] [[ 0.07972913]] [[ 0.08672011]] [[ 0.04515148]] [[ 0.06909422]] [[ 0.05746065]] [[-0.01622215]] [[ 0.0576393 ]] [[ 0.17703095]] [[ 0.04814146]] [[ 0.07529981]] [[ 0.08827124]] [[ 0.05045927]] [[ 0.06269133]] [[ 0.08623993]] [[ 0.01814328]] [[ 0.11289588]] [[ 0.04944252]] [[ 0.10802261]] [[ 0.11637259]] [[ 0.01122058]] [[ 0.04731656]] [[ 0.05897205]] [[ 0.09767184]] [[ 0.00135915]] [[ 0.04978014]] [[ 0.05766164]] [[ 0.08329739]] [[ 0.07507744]] [[ 0.04028618]] [[ 0.04261505]] [[ 0.13034639]] [[ 0.07776155]] [[ 0.00332735]] [[ 0.14093177]] [[ 0.16749053]] [[ 0.10557273]] [[ 0.08042369]] [[ 0.08202975]] [[ 0.09170167]] [[ 0.1356161 ]] [[ 0.08357933]] [[ 0.0974719 ]] [[ 0.11223293]] [[-0.00038942]] [[ 0.05731443]] [[ 0.02905753]] [[-0.04610949]] [[ 0.01388604]] [[ 0.08600491]] [[ 0.03778464]] [[ 0.11034977]] [[ 0.04955809]] [[ 0.08919523]] [[ 0.06298601]] [[ 0.00079616]] [[ 0.08868147]] [[ 0.11750155]] [[ 0.12066914]] [[ 0.06959121]] [[ 0.07425694]] [[ 0.07394385]] [[ 0.07509953]] [[ 0.06012011]] [[ 0.05826611]] [[ 0.04602192]] [[ 0.08241433]] [[ 0.02783399]] [[ 0.02495469]] [[ 0.09342177]] [[ 0.06165688]] [[ 0.05896452]] [[ 0.02253314]] [[ 0.07780281]] [[ 0.07161339]] [[-0.0221484 ]] [[-0.03266057]] [[ 0.06453755]] [[ 0.07277562]] [[ 0.05789132]] [[ 0.0713466 ]] [[ 0.06435218]] [[ 0.07362403]] [[ 0.08777972]] [[ 0.12356211]] [[ 0.02017886]] [[ 0.10155802]] [[ 0.08033946]] [[ 0.06478995]] [[ 0.05630787]] [[ 0.1496954 ]] [[ 0.05970615]] [[ 0.02488095]] [[ 0.09522102]] [[ 0.11198034]] [[ 0.02205544]] [[ 0.06242434]] [[ 0.15488677]] [[ 0.02909784]] [[ 0.05858369]] [[ 0.06075398]] [[ 0.01913048]] [[ 0.01297504]] [[ 0.05418276]] [[ 0.03741093]] [[ 0.12530588]] [[ 0.01182354]] [[ 0.08590037]] [[ 0.07205859]] [[ 0.09614597]] [[ 0.02901293]] [[ 0.0280707 ]] [[ 0.04857701]] [[ 0.13908157]] [[ 0.02994158]] [[ 0.127237 ]] [[ 0.05313 ]] [[ 0.00637886]] [[ 0.1148726 ]] [[ 0.04164571]] [[ 0.04220049]] [[ 0.07257959]] [[ 0.08117146]] [[ 0.0093219 ]] [[ 0.05521437]] [[ 0.04970867]] [[ 0.05983538]] [[ 0.0710875 ]] [[ 0.11061664]] [[-0.02379834]] [[ 0.06210491]] [[ 0.11894427]] [[ 0.14475995]] [[ 0.06255442]] [[ 0.09570058]] [[ 0.03538594]] [[ 0.07302488]] [[ 0.07556595]] [[ 0.03807254]] [[ 0.11827453]] [[ 0.09984626]] [[ 0.07600728]] [[ 0.0467343 ]] [[ 0.01499407]] [[ 0.09875156]] [[ 0.0671856 ]] [[ 0.00813012]] [[ 0.00043575]] [[ 0.13457415]] [[ 0.11626634]] [[ 0.10812641]] [[ 0.06989877]] [[ 0.09068255]] [[ 0.05036348]] [[ 0.08673097]] [[ 0.04698934]] [[ 0.05146983]]]] ==================================================================================== layer_type: Add layer_id: 39 input_layer0: layer_id=37: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='conv2d_6/Conv2D:0', description="created by layer 'conv2d_6'") input_layer1_value: layer_id=38: [[[[ 0.03531126]] [[ 0.04153406]] [[-0.01499302]] [[ 0.04892787]] [[ 0.10467708]] [[ 0.05344332]] [[ 0.09078331]] [[ 0.03810004]] [[ 0.09852754]] [[ 0.10886087]] [[ 0.0583949 ]] [[ 0.15019625]] [[ 0.12375277]] [[ 0.0836074 ]] [[ 0.05285143]] [[-0.01451365]] [[ 0.09832055]] [[ 0.05607849]] [[ 0.07759804]] [[ 0.07484994]] [[ 0.14392051]] [[ 0.12323733]] [[ 0.04834333]] [[ 0.04026539]] [[ 0.0206848 ]] [[ 0.0345908 ]] [[ 0.09243347]] [[ 0.04752847]] [[ 0.09716211]] [[ 0.02665334]] [[ 0.07137176]] [[ 0.06183958]] [[ 0.05770827]] [[ 0.04015264]] [[ 0.05639784]] [[ 0.05093274]] [[ 0.06614189]] [[ 0.04954747]] [[ 0.01887337]] [[ 0.08641648]] [[ 0.03211017]] [[ 0.13757695]] [[ 0.02274888]] [[ 0.01190568]] [[ 0.02532989]] [[ 0.0563163 ]] [[ 0.04885947]] [[ 0.06898033]] [[ 0.01896106]] [[ 0.10011526]] [[ 0.05837568]] [[ 0.06790939]] [[ 0.09529823]] [[-0.0412923 ]] [[ 0.13273457]] [[ 0.07322548]] [[ 0.09249222]] [[ 0.04539789]] [[ 0.08955366]] [[ 0.03735309]] [[ 0.04273967]] [[ 0.09724385]] [[ 0.01885115]] [[ 0.00743112]] [[ 0.03441679]] [[ 0.01052186]] [[-0.00418794]] [[ 0.04830207]] [[ 0.09452251]] [[ 0.09627661]] [[ 0.05013204]] [[ 0.13239259]] [[ 0.1103457 ]] [[ 0.04369677]] [[ 0.01661588]] [[ 0.05327062]] [[ 0.06146394]] [[ 0.05842484]] [[ 0.03206206]] [[ 0.10954275]] [[ 0.0323467 ]] [[ 0.05475442]] [[ 0.10349695]] [[ 0.08188834]] [[ 0.04716377]] [[ 0.07654594]] [[ 0.08036467]] [[ 0.08638818]] [[ 0.08552508]] [[ 0.10098928]] [[ 0.05385393]] [[ 0.01202287]] [[ 0.06818848]] [[ 0.07115225]] [[-0.00766839]] [[-0.0496178 ]] [[ 0.05881699]] [[ 0.05345987]] [[ 0.11750553]] [[ 0.09980227]] [[ 0.07972913]] [[ 0.08672011]] [[ 0.04515148]] [[ 0.06909422]] [[ 0.05746065]] [[-0.01622215]] [[ 0.0576393 ]] [[ 0.17703095]] [[ 0.04814146]] [[ 0.07529981]] [[ 0.08827124]] [[ 0.05045927]] [[ 0.06269133]] [[ 0.08623993]] [[ 0.01814328]] [[ 0.11289588]] [[ 0.04944252]] [[ 0.10802261]] [[ 0.11637259]] [[ 0.01122058]] [[ 0.04731656]] [[ 0.05897205]] [[ 0.09767184]] [[ 0.00135915]] [[ 0.04978014]] [[ 0.05766164]] [[ 0.08329739]] [[ 0.07507744]] [[ 0.04028618]] [[ 0.04261505]] [[ 0.13034639]] [[ 0.07776155]] [[ 0.00332735]] [[ 0.14093177]] [[ 0.16749053]] [[ 0.10557273]] [[ 0.08042369]] [[ 0.08202975]] [[ 0.09170167]] [[ 0.1356161 ]] [[ 0.08357933]] [[ 0.0974719 ]] [[ 0.11223293]] [[-0.00038942]] [[ 0.05731443]] [[ 0.02905753]] [[-0.04610949]] [[ 0.01388604]] [[ 0.08600491]] [[ 0.03778464]] [[ 0.11034977]] [[ 0.04955809]] [[ 0.08919523]] [[ 0.06298601]] [[ 0.00079616]] [[ 0.08868147]] [[ 0.11750155]] [[ 0.12066914]] [[ 0.06959121]] [[ 0.07425694]] [[ 0.07394385]] [[ 0.07509953]] [[ 0.06012011]] [[ 0.05826611]] [[ 0.04602192]] [[ 0.08241433]] [[ 0.02783399]] [[ 0.02495469]] [[ 0.09342177]] [[ 0.06165688]] [[ 0.05896452]] [[ 0.02253314]] [[ 0.07780281]] [[ 0.07161339]] [[-0.0221484 ]] [[-0.03266057]] [[ 0.06453755]] [[ 0.07277562]] [[ 0.05789132]] [[ 0.0713466 ]] [[ 0.06435218]] [[ 0.07362403]] [[ 0.08777972]] [[ 0.12356211]] [[ 0.02017886]] [[ 0.10155802]] [[ 0.08033946]] [[ 0.06478995]] [[ 0.05630787]] [[ 0.1496954 ]] [[ 0.05970615]] [[ 0.02488095]] [[ 0.09522102]] [[ 0.11198034]] [[ 0.02205544]] [[ 0.06242434]] [[ 0.15488677]] [[ 0.02909784]] [[ 0.05858369]] [[ 0.06075398]] [[ 0.01913048]] [[ 0.01297504]] [[ 0.05418276]] [[ 0.03741093]] [[ 0.12530588]] [[ 0.01182354]] [[ 0.08590037]] [[ 0.07205859]] [[ 0.09614597]] [[ 0.02901293]] [[ 0.0280707 ]] [[ 0.04857701]] [[ 0.13908157]] [[ 0.02994158]] [[ 0.127237 ]] [[ 0.05313 ]] [[ 0.00637886]] [[ 0.1148726 ]] [[ 0.04164571]] [[ 0.04220049]] [[ 0.07257959]] [[ 0.08117146]] [[ 0.0093219 ]] [[ 0.05521437]] [[ 0.04970867]] [[ 0.05983538]] [[ 0.0710875 ]] [[ 0.11061664]] [[-0.02379834]] [[ 0.06210491]] [[ 0.11894427]] [[ 0.14475995]] [[ 0.06255442]] [[ 0.09570058]] [[ 0.03538594]] [[ 0.07302488]] [[ 0.07556595]] [[ 0.03807254]] [[ 0.11827453]] [[ 0.09984626]] [[ 0.07600728]] [[ 0.0467343 ]] [[ 0.01499407]] [[ 0.09875156]] [[ 0.0671856 ]] [[ 0.00813012]] [[ 0.00043575]] [[ 0.13457415]] [[ 0.11626634]] [[ 0.10812641]] [[ 0.06989877]] [[ 0.09068255]] [[ 0.05036348]] [[ 0.08673097]] [[ 0.04698934]] [[ 0.05146983]]]] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.math.add_6/Add:0', description="created by layer 'tf.math.add_6'") ==================================================================================== layer_type: ReLU layer_id: 40 input_layer0: layer_id=39: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.math.add_6/Add:0', description="created by layer 'tf.math.add_6'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.nn.relu_6/Relu:0', description="created by layer 'tf.nn.relu_6'") ==================================================================================== layer_type: Const layer_id: 41 tf_layers_dict_shape: (4,) tf_layers_dict_value: [0 3 1 2] ==================================================================================== layer_type: Transpose layer_id: 42 input_layer0: layer_id=40: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.nn.relu_6/Relu:0', description="created by layer 'tf.nn.relu_6'") input_layer1_value: layer_id=41: [0 3 1 2] tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.compat.v1.transpose/transpose:0', description="created by layer 'tf.compat.v1.transpose'") ==================================================================================== layer_type: AvgPool layer_id: 43 input_layer0: layer_id=42: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.compat.v1.transpose/transpose:0', description="created by layer 'tf.compat.v1.transpose'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 49, 256), dtype=tf.float32, name=None), name='average_pooling2d/AvgPool:0', description="created by layer 'average_pooling2d'") ==================================================================================== layer_type: Const layer_id: 44 tf_layers_dict_shape: (1,) tf_layers_dict_value: [3] ==================================================================================== ERROR: Can not squeeze dim[3], expected a dimension of 1, got 256 for '{{node tf.squeeze_1/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[-1]](Placeholder)' with input shapes: [32,3,49,256]. ERROR: model_path : /content/openvino1/custom_ocr1.xml ERROR: weights_path: /content/openvino1/custom_ocr1.bin ERROR: layer_id : 45 ERROR: input_layer0 layer_id=43: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 49, 256), dtype=tf.float32, name=None), name='average_pooling2d/AvgPool:0', description="created by layer 'average_pooling2d'") ERROR: input_layer1 layer_id=44: Const(ndarray).shape (1,) array([3]) ERROR: The trace log is below. Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/deprecation.py", line 538, in new_func return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4379, in squeeze return gen_array_ops.squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/gen_array_ops.py", line 10161, in squeeze input, axis=axis, name=name, ctx=_ctx) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/gen_array_ops.py", line 10195, in squeeze_eager_fallback _attr_T, (input,) = _execute.args_to_matching_eager([input], ctx, []) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/execute.py", line 274, in args_to_matching_eager t, dtype, preferred_dtype=default_dtype, ctx=ctx) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/profiler/trace.py", line 163, in wrapped return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1540, in convert_to_tensor ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 339, in _constant_tensor_conversion_function return constant(v, dtype=dtype, name=name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 265, in constant allow_broadcast=True) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 276, in _constant_impl return _constant_eager_impl(ctx, value, dtype, shape, verify_shape) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 301, in _constant_eager_impl t = convert_to_eager_tensor(value, ctx, dtype) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 98, in convert_to_eager_tensor return ops.EagerTensor(value, ctx.device_name, dtype) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/keras_tensor.py", line 274, in __array__ 'Cannot convert a symbolic Keras input/output to a numpy array. ' TypeError: Cannot convert a symbolic Keras input/output to a numpy array. This error may indicate that you're trying to pass a symbolic value to a NumPy call, which is not supported. Or, you may be trying to pass Keras symbolic inputs/outputs to a TF API that does not register dispatching, preventing Keras from automatically converting the API call to a lambda layer in the Functional Model. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1853, in _create_c_op c_op = pywrap_tf_session.TF_FinishOperation(op_desc) tensorflow.python.framework.errors_impl.InvalidArgumentError: Can not squeeze dim[2], expected a dimension of 1, got 49 for '{{node tf.compat.v1.squeeze/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[2]](Placeholder)' with input shapes: [32,3,49,256]. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4431, in squeeze_v2 return squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 205, in wrapper result = dispatch(wrapper, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 122, in dispatch result = dispatcher.handle(op, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1450, in handle return TFOpLambda(op)(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 952, in __call__ input_list) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 1091, in _functional_construction_call inputs, input_masks, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 822, in _keras_tensor_symbolic_call return self._infer_output_signature(inputs, args, kwargs, input_masks) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 863, in _infer_output_signature outputs = call_fn(inputs, *args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1327, in _call_wrapper return self._call_wrapper(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1359, in _call_wrapper result = self.function(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/deprecation.py", line 538, in new_func return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4379, in squeeze return gen_array_ops.squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/gen_array_ops.py", line 10173, in squeeze "Squeeze", input=input, squeeze_dims=axis, name=name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/op_def_library.py", line 750, in _apply_op_helper attrs=attr_protos, op_def=op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py", line 592, in _create_op_internal compute_device) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 3536, in _create_op_internal op_def=op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 2016, in __init__ control_input_ops, op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1856, in _create_c_op raise ValueError(str(e)) ValueError: Can not squeeze dim[2], expected a dimension of 1, got 49 for '{{node tf.compat.v1.squeeze/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[2]](Placeholder)' with input shapes: [32,3,49,256]. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1853, in _create_c_op c_op = pywrap_tf_session.TF_FinishOperation(op_desc) tensorflow.python.framework.errors_impl.InvalidArgumentError: Can not squeeze dim[2], expected a dimension of 1, got 49 for '{{node tf.squeeze/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[2]](Placeholder)' with input shapes: [32,3,49,256]. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "openvino2tensorflow.py", line 3191, in convert axis=axis File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 205, in wrapper result = dispatch(wrapper, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 122, in dispatch result = dispatcher.handle(op, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1450, in handle return TFOpLambda(op)(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 952, in __call__ input_list) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 1091, in _functional_construction_call inputs, input_masks, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 822, in _keras_tensor_symbolic_call return self._infer_output_signature(inputs, args, kwargs, input_masks) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 863, in _infer_output_signature outputs = call_fn(inputs, *args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1327, in _call_wrapper return self._call_wrapper(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1359, in _call_wrapper result = self.function(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4431, in squeeze_v2 return squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/deprecation.py", line 538, in new_func return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4379, in squeeze return gen_array_ops.squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/gen_array_ops.py", line 10173, in squeeze "Squeeze", input=input, squeeze_dims=axis, name=name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/op_def_library.py", line 750, in _apply_op_helper attrs=attr_protos, op_def=op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py", line 592, in _create_op_internal compute_device) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 3536, in _create_op_internal op_def=op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 2016, in __init__ control_input_ops, op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1856, in _create_c_op raise ValueError(str(e)) ValueError: Can not squeeze dim[2], expected a dimension of 1, got 49 for '{{node tf.squeeze/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[2]](Placeholder)' with input shapes: [32,3,49,256]. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/deprecation.py", line 538, in new_func return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4379, in squeeze return gen_array_ops.squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/gen_array_ops.py", line 10161, in squeeze input, axis=axis, name=name, ctx=_ctx) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/gen_array_ops.py", line 10195, in squeeze_eager_fallback _attr_T, (input,) = _execute.args_to_matching_eager([input], ctx, []) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/execute.py", line 274, in args_to_matching_eager t, dtype, preferred_dtype=default_dtype, ctx=ctx) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/profiler/trace.py", line 163, in wrapped return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1540, in convert_to_tensor ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 339, in _constant_tensor_conversion_function return constant(v, dtype=dtype, name=name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 265, in constant allow_broadcast=True) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 276, in _constant_impl return _constant_eager_impl(ctx, value, dtype, shape, verify_shape) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 301, in _constant_eager_impl t = convert_to_eager_tensor(value, ctx, dtype) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/constant_op.py", line 98, in convert_to_eager_tensor return ops.EagerTensor(value, ctx.device_name, dtype) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/keras_tensor.py", line 274, in __array__ 'Cannot convert a symbolic Keras input/output to a numpy array. ' TypeError: Cannot convert a symbolic Keras input/output to a numpy array. This error may indicate that you're trying to pass a symbolic value to a NumPy call, which is not supported. Or, you may be trying to pass Keras symbolic inputs/outputs to a TF API that does not register dispatching, preventing Keras from automatically converting the API call to a lambda layer in the Functional Model. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1853, in _create_c_op c_op = pywrap_tf_session.TF_FinishOperation(op_desc) tensorflow.python.framework.errors_impl.InvalidArgumentError: Can not squeeze dim[3], expected a dimension of 1, got 256 for '{{node tf.compat.v1.squeeze_1/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[-1]](Placeholder)' with input shapes: [32,3,49,256]. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4431, in squeeze_v2 return squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 205, in wrapper result = dispatch(wrapper, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 122, in dispatch result = dispatcher.handle(op, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1450, in handle return TFOpLambda(op)(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 952, in __call__ input_list) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 1091, in _functional_construction_call inputs, input_masks, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 822, in _keras_tensor_symbolic_call return self._infer_output_signature(inputs, args, kwargs, input_masks) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 863, in _infer_output_signature outputs = call_fn(inputs, *args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1327, in _call_wrapper return self._call_wrapper(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1359, in _call_wrapper result = self.function(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/deprecation.py", line 538, in new_func return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4379, in squeeze return gen_array_ops.squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/gen_array_ops.py", line 10173, in squeeze "Squeeze", input=input, squeeze_dims=axis, name=name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/op_def_library.py", line 750, in _apply_op_helper attrs=attr_protos, op_def=op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py", line 592, in _create_op_internal compute_device) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 3536, in _create_op_internal op_def=op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 2016, in __init__ control_input_ops, op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1856, in _create_c_op raise ValueError(str(e)) ValueError: Can not squeeze dim[3], expected a dimension of 1, got 256 for '{{node tf.compat.v1.squeeze_1/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[-1]](Placeholder)' with input shapes: [32,3,49,256]. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1853, in _create_c_op c_op = pywrap_tf_session.TF_FinishOperation(op_desc) tensorflow.python.framework.errors_impl.InvalidArgumentError: Can not squeeze dim[3], expected a dimension of 1, got 256 for '{{node tf.squeeze_1/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[-1]](Placeholder)' with input shapes: [32,3,49,256]. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "openvino2tensorflow.py", line 3218, in convert axis=-1 File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 205, in wrapper result = dispatch(wrapper, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 122, in dispatch result = dispatcher.handle(op, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1450, in handle return TFOpLambda(op)(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 952, in __call__ input_list) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 1091, in _functional_construction_call inputs, input_masks, args, kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 822, in _keras_tensor_symbolic_call return self._infer_output_signature(inputs, args, kwargs, input_masks) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 863, in _infer_output_signature outputs = call_fn(inputs, *args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1327, in _call_wrapper return self._call_wrapper(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/layers/core.py", line 1359, in _call_wrapper result = self.function(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4431, in squeeze_v2 return squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py", line 201, in wrapper return target(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/deprecation.py", line 538, in new_func return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/array_ops.py", line 4379, in squeeze return gen_array_ops.squeeze(input, axis, name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/ops/gen_array_ops.py", line 10173, in squeeze "Squeeze", input=input, squeeze_dims=axis, name=name) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/op_def_library.py", line 750, in _apply_op_helper attrs=attr_protos, op_def=op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py", line 592, in _create_op_internal compute_device) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 3536, in _create_op_internal op_def=op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 2016, in __init__ control_input_ops, op_def) File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/ops.py", line 1856, in _create_c_op raise ValueError(str(e)) ValueError: Can not squeeze dim[3], expected a dimension of 1, got 256 for '{{node tf.squeeze_1/Squeeze}} = Squeeze[T=DT_FLOAT, squeeze_dims=[-1]](Placeholder)' with input shapes: [32,3,49,256]. ```

Source code for simple inference testing code

No response

[PINTO0309]

Unfortunately, there are no plans to implement a recurrent layer. LSTM = LSTMCell = TensorIterator + α

• replace.json

  {
  "format_version": 2,
  "layers": [
      {
          "layer_id": "41",
          "type": "Const",
          "replace_mode": "direct",
          "values": [
              0,
              3,
              1,
              2
          ]
      }
  ]
  }

• Command

  
  docker run -it --rm \
  -v `pwd`:/home/user/workdir \
  ghcr.io/pinto0309/openvino2tensorflow:latest

$INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo.py \ --input_model custom_ocr1.onnx \ --data_type FP32 \ --output_dir openvino/FP32 \ --model_name custom_ocr1

openvino2tensorflow \ --model_path openvino/FP32/custom_ocr1.xml \ --output_saved_model \ --output_pb \ --output_no_quant_float32_tflite \ --non_verbose \ --weight_replacement_config replace.json

- Log

TensorFlow/Keras model building process starts ====================================== weight_replacement_config format_version: 2 Replace the value of Const for each layer_id with the value below. {'41': {'layer_id': '41', 'replace_mode': 'direct', 'type': 'Const', 'values': [0, 3, 1, 2]}} Layer structure ===================================================================== layer_type: Input layer_id: 0 tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 64, 600, 1), dtype=tf.float32, name='input_1'), name='input_1', description="created by layer 'input_1'")

layer_type: Const layer_id: 1 tf_layers_dict_shape: (1, 1, 256)

layer_type: Const layer_id: 2 tf_layers_dict_shape: (32, 1, 3, 3)

layer_type: Convolution layer_id: 3 input_layer0: layer_id=0: KerasTensor(type_spec=TensorSpec(shape=(32, 64, 600, 1), dtype=tf.float32, name='input_1'), name='input_1', description="created by layer 'input_1'") input_layer1_shape: layer_id=2: (32, 1, 3, 3) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 64, 600, 32), dtype=tf.float32, name=None), name='conv2d/Conv2D:0', description="created by layer 'conv2d'")

layer_type: Const layer_id: 4 tf_layers_dict_shape: (1, 32, 1, 1)

layer_type: Add layer_id: 5 input_layer0: layer_id=3: KerasTensor(type_spec=TensorSpec(shape=(32, 64, 600, 32), dtype=tf.float32, name=None), name='conv2d/Conv2D:0', description="created by layer 'conv2d'") input_layer1_shape: layer_id=4: (1, 32, 1, 1) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 64, 600, 32), dtype=tf.float32, name=None), name='tf.math.add/Add:0', description="created by layer 'tf.math.add'")

layer_type: ReLU layer_id: 6 input_layer0: layer_id=5: KerasTensor(type_spec=TensorSpec(shape=(32, 64, 600, 32), dtype=tf.float32, name=None), name='tf.math.add/Add:0', description="created by layer 'tf.math.add'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 64, 600, 32), dtype=tf.float32, name=None), name='tf.nn.relu/Relu:0', description="created by layer 'tf.nn.relu'")

layer_type: MaxPool layer_id: 7 input_layer0: layer_id=6: KerasTensor(type_spec=TensorSpec(shape=(32, 64, 600, 32), dtype=tf.float32, name=None), name='tf.nn.relu/Relu:0', description="created by layer 'tf.nn.relu'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 32, 300, 32), dtype=tf.float32, name=None), name='tf.nn.max_pool2d/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d'")

layer_type: Const layer_id: 8 tf_layers_dict_shape: (64, 32, 3, 3)

layer_type: Convolution layer_id: 9 input_layer0: layer_id=7: KerasTensor(type_spec=TensorSpec(shape=(32, 32, 300, 32), dtype=tf.float32, name=None), name='tf.nn.max_pool2d/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d'") input_layer1_shape: layer_id=8: (64, 32, 3, 3) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 32, 300, 64), dtype=tf.float32, name=None), name='conv2d_1/Conv2D:0', description="created by layer 'conv2d_1'")

layer_type: Const layer_id: 10 tf_layers_dict_shape: (1, 64, 1, 1)

layer_type: Add layer_id: 11 input_layer0: layer_id=9: KerasTensor(type_spec=TensorSpec(shape=(32, 32, 300, 64), dtype=tf.float32, name=None), name='conv2d_1/Conv2D:0', description="created by layer 'conv2d_1'") input_layer1_shape: layer_id=10: (1, 64, 1, 1) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 32, 300, 64), dtype=tf.float32, name=None), name='tf.math.add_1/Add:0', description="created by layer 'tf.math.add_1'")

layer_type: ReLU layer_id: 12 input_layer0: layer_id=11: KerasTensor(type_spec=TensorSpec(shape=(32, 32, 300, 64), dtype=tf.float32, name=None), name='tf.math.add_1/Add:0', description="created by layer 'tf.math.add_1'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 32, 300, 64), dtype=tf.float32, name=None), name='tf.nn.relu_1/Relu:0', description="created by layer 'tf.nn.relu_1'")

layer_type: MaxPool layer_id: 13 input_layer0: layer_id=12: KerasTensor(type_spec=TensorSpec(shape=(32, 32, 300, 64), dtype=tf.float32, name=None), name='tf.nn.relu_1/Relu:0', description="created by layer 'tf.nn.relu_1'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 64), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_1/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_1'")

layer_type: Const layer_id: 14 tf_layers_dict_shape: (128, 64, 3, 3)

layer_type: Convolution layer_id: 15 input_layer0: layer_id=13: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 64), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_1/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_1'") input_layer1_shape: layer_id=14: (128, 64, 3, 3) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='conv2d_2/Conv2D:0', description="created by layer 'conv2d_2'")

layer_type: Const layer_id: 16 tf_layers_dict_shape: (1, 128, 1, 1)

layer_type: Add layer_id: 17 input_layer0: layer_id=15: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='conv2d_2/Conv2D:0', description="created by layer 'conv2d_2'") input_layer1_shape: layer_id=16: (1, 128, 1, 1) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.math.add_2/Add:0', description="created by layer 'tf.math.add_2'")

layer_type: ReLU layer_id: 18 input_layer0: layer_id=17: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.math.add_2/Add:0', description="created by layer 'tf.math.add_2'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.nn.relu_2/Relu:0', description="created by layer 'tf.nn.relu_2'")

layer_type: Const layer_id: 19 tf_layers_dict_shape: (128, 128, 3, 3)

layer_type: Convolution layer_id: 20 input_layer0: layer_id=18: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.nn.relu_2/Relu:0', description="created by layer 'tf.nn.relu_2'") input_layer1_shape: layer_id=19: (128, 128, 3, 3) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='conv2d_3/Conv2D:0', description="created by layer 'conv2d_3'")

layer_type: Const layer_id: 21 tf_layers_dict_shape: (1, 128, 1, 1)

layer_type: Add layer_id: 22 input_layer0: layer_id=20: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='conv2d_3/Conv2D:0', description="created by layer 'conv2d_3'") input_layer1_shape: layer_id=21: (1, 128, 1, 1) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.math.add_3/Add:0', description="created by layer 'tf.math.add_3'")

layer_type: ReLU layer_id: 23 input_layer0: layer_id=22: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.math.add_3/Add:0', description="created by layer 'tf.math.add_3'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.nn.relu_3/Relu:0', description="created by layer 'tf.nn.relu_3'")

layer_type: MaxPool layer_id: 24 input_layer0: layer_id=23: KerasTensor(type_spec=TensorSpec(shape=(32, 16, 150, 128), dtype=tf.float32, name=None), name='tf.nn.relu_3/Relu:0', description="created by layer 'tf.nn.relu_3'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 128), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_2/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_2'")

layer_type: Const layer_id: 25 tf_layers_dict_shape: (256, 128, 3, 3)

layer_type: Convolution layer_id: 26 input_layer0: layer_id=24: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 128), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_2/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_2'") input_layer1_shape: layer_id=25: (256, 128, 3, 3) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='conv2d_4/Conv2D:0', description="created by layer 'conv2d_4'")

layer_type: Const layer_id: 27 tf_layers_dict_shape: (1, 256, 1, 1)

layer_type: Add layer_id: 28 input_layer0: layer_id=26: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='conv2d_4/Conv2D:0', description="created by layer 'conv2d_4'") input_layer1_shape: layer_id=27: (1, 256, 1, 1) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.math.add_4/Add:0', description="created by layer 'tf.math.add_4'")

layer_type: ReLU layer_id: 29 input_layer0: layer_id=28: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.math.add_4/Add:0', description="created by layer 'tf.math.add_4'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.nn.relu_4/Relu:0', description="created by layer 'tf.nn.relu_4'")

layer_type: Const layer_id: 30 tf_layers_dict_shape: (256, 256, 3, 3)

layer_type: Convolution layer_id: 31 input_layer0: layer_id=29: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.nn.relu_4/Relu:0', description="created by layer 'tf.nn.relu_4'") input_layer1_shape: layer_id=30: (256, 256, 3, 3) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='conv2d_5/Conv2D:0', description="created by layer 'conv2d_5'")

layer_type: Const layer_id: 32 tf_layers_dict_shape: (1, 256, 1, 1)

layer_type: Add layer_id: 33 input_layer0: layer_id=31: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='conv2d_5/Conv2D:0', description="created by layer 'conv2d_5'") input_layer1_shape: layer_id=32: (1, 256, 1, 1) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.math.add_5/Add:0', description="created by layer 'tf.math.add_5'")

layer_type: ReLU layer_id: 34 input_layer0: layer_id=33: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.math.add_5/Add:0', description="created by layer 'tf.math.add_5'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.nn.relu_5/Relu:0', description="created by layer 'tf.nn.relu_5'")

layer_type: MaxPool layer_id: 35 input_layer0: layer_id=34: KerasTensor(type_spec=TensorSpec(shape=(32, 8, 150, 256), dtype=tf.float32, name=None), name='tf.nn.relu_5/Relu:0', description="created by layer 'tf.nn.relu_5'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 4, 150, 256), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_3/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_3'")

layer_type: Const layer_id: 36 tf_layers_dict_shape: (256, 256, 2, 2)

layer_type: Convolution layer_id: 37 input_layer0: layer_id=35: KerasTensor(type_spec=TensorSpec(shape=(32, 4, 150, 256), dtype=tf.float32, name=None), name='tf.nn.max_pool2d_3/MaxPool2d:0', description="created by layer 'tf.nn.max_pool2d_3'") input_layer1_shape: layer_id=36: (256, 256, 2, 2) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='conv2d_6/Conv2D:0', description="created by layer 'conv2d_6'")

layer_type: Const layer_id: 38 tf_layers_dict_shape: (1, 256, 1, 1)

layer_type: Add layer_id: 39 input_layer0: layer_id=37: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='conv2d_6/Conv2D:0', description="created by layer 'conv2d_6'") input_layer1_shape: layer_id=38: (1, 256, 1, 1) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.math.add_6/Add:0', description="created by layer 'tf.math.add_6'")

layer_type: ReLU layer_id: 40 input_layer0: layer_id=39: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.math.add_6/Add:0', description="created by layer 'tf.math.add_6'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.nn.relu_6/Relu:0', description="created by layer 'tf.nn.relu_6'")

layer_type: Const layer_id: 41 tf_layers_dict_shape: (4,)

layer_type: Transpose layer_id: 42 input_layer0: layer_id=40: KerasTensor(type_spec=TensorSpec(shape=(32, 3, 149, 256), dtype=tf.float32, name=None), name='tf.nn.relu_6/Relu:0', description="created by layer 'tf.nn.relu_6'") input_layer1_shape: layer_id=41: (4,) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 256, 3, 149), dtype=tf.float32, name=None), name='tf.compat.v1.transpose/transpose:0', description="created by layer 'tf.compat.v1.transpose'")

layer_type: AvgPool layer_id: 43 input_layer0: layer_id=42: KerasTensor(type_spec=TensorSpec(shape=(32, 256, 3, 149), dtype=tf.float32, name=None), name='tf.compat.v1.transpose/transpose:0', description="created by layer 'tf.compat.v1.transpose'") tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 256, 1, 149), dtype=tf.float32, name=None), name='average_pooling2d/AvgPool:0', description="created by layer 'average_pooling2d'")

layer_type: Const layer_id: 44 tf_layers_dict_shape: (1,)

layer_type: Squeeze layer_id: 45 input_layer0: layer_id=43: KerasTensor(type_spec=TensorSpec(shape=(32, 256, 1, 149), dtype=tf.float32, name=None), name='average_pooling2d/AvgPool:0', description="created by layer 'average_pooling2d'") input_layer1_shape: layer_id=44: (1,) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(32, 256, 149), dtype=tf.float32, name=None), name='tf.compat.v1.squeeze/Squeeze:0', description="created by layer 'tf.compat.v1.squeeze'") axis: 2

layer_type: Const layer_id: 46 tf_layers_dict_shape: (3,)

layer_type: Transpose layer_id: 47 input_layer0: layer_id=45: KerasTensor(type_spec=TensorSpec(shape=(32, 256, 149), dtype=tf.float32, name=None), name='tf.compat.v1.squeeze/Squeeze:0', description="created by layer 'tf.compat.v1.squeeze'") input_layer1_shape: layer_id=46: (3,) tf_layers_dict: KerasTensor(type_spec=TensorSpec(shape=(256, 32, 149), dtype=tf.float32, name=None), name='tf.compat.v1.transpose_1/transpose:0', description="created by layer 'tf.compat.v1.transpose_1'")

layer_type: Const layer_id: 48 tf_layers_dict_shape: (1, 32, 256)

layer_type: ShapeOf layer_id: 49 input_layer0_shape: layer_id=47: Const(ndarray).shape (256, 32, 149) tf_layers_dict_shape: (3,) tf_layers_dict: [256 32 149]

layer_type: Const layer_id: 50 tf_layers_dict_shape: (1,)

layer_type: Const layer_id: 51 tf_layers_dict_shape: (1,)

layer_type: Gather layer_id: 52 input_layer0_shape: layer_id=49: Const(ndarray).shape (3,) input_layer1_shape: layer_id=50: Const(ndarray).shape (1,) input_layer2_shape: layer_id=51: Const(ndarray).shape (1,) tf_layers_dict_shape: (1,) tf_layers_dict: [32]

layer_type: Const layer_id: 53 tf_layers_dict_shape: (1,)

layer_type: Concat layer_id: 54 input_layer0_shape: layer_id=52: Const(ndarray).shape (1,) input_layer1_shape: layer_id=53: Const(ndarray).shape (1,) tf_layers_dict_shape: (2,) tf_layers_dict: [ 32 256] axis: 0

layer_type: Reshape layer_id: 55 input_layer0_shape: layer_id=48: Const(ndarray).shape (1, 32, 256) input_layer1_shape: layer_id=54: Const(ndarray).shape (2,) tf_layers_dict_shape: (32, 256) tf_layers_dict: [[0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.] ... [0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.]]

layer_type: Const layer_id: 56 tf_layers_dict_shape: (1, 32, 256)

layer_type: Reshape layer_id: 57 input_layer0_shape: layer_id=56: Const(ndarray).shape (1, 32, 256) input_layer1_shape: layer_id=54: Const(ndarray).shape (2,) tf_layers_dict_shape: (32, 256) tf_layers_dict: [[0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.] ... [0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.] [0. 0. 0. ... 0. 0. 0.]]

The TensorIterator layer is not yet implemented.

[MjiS]

Thank you for your reply. Any solution to this or any other way I can convert this model to tflite. I tried PyTorch > ONNX > PB > tflite. I had issues converting PB to tflite, I tried

pip install toco

!toco --graph_def_file='/content/export1/saved_model.pb' --output_file='/content/custom_ocr.tflite' --input_format=TENSORFLOW_GRAPHDEF --output_format=TFLITE --input_shape=32,1,256,256 --input_array=ResizeBilinear --output_array=final_result --inference_type=FLOAT --input_type=FLOAT --saved_model_dir '/content/export1'

It did converted the file to tflite but I was unable to properly utilize it. Here is the link for tflite file https://drive.google.com/file/d/17YJNfiYegeUvBgLKbZkKNNNS3orAhlDw/view?usp=sharing Can you look into it. Is this tflite converted properly ?

[PINTO0309]

I've tried every easy conversion method you can think of. To sum up, there is no easy way to convert.

You can extract the weights one by one from ONNX and read them into the Keras model to regenerate them. I am not going to implement it as it is a high work cost. However, I think it's a much more meaningful way to spend your time than wasting it on worries.

https://github.com/onnx/onnx-tensorflow