"IndexError: list index out of range" occurs when trying to convert BigGAN tflite model to ONNX format

[xtrizeShino]

Ask a Question

Question

"IndexError: list index out of range" occurs when I try to convert a BigGAN tflite model to ONNX format. Can you tell me the correct command or source code to convert this model?

https://drive.google.com/file/d/149nTlPi0WlQZn3F_IAUlzfWkZkMnVok1/view?usp=drive_link

Commands I tried

$ python -m tf2onnx.convert --tflite biggan-128.tflite --output biggan-128_nchw_output.onnx --outputs-as-nchw "StatefulPartitionedCall:0" --inputs "serving_default_y:0,serving_default_z:0,serving_default_truncation:0" --outputs "StatefulPartitionedCall:0"

Error that occurred when executing the command

2025-04-11 23:58:33.873525: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2025-04-11 23:58:33.875774: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-11 23:58:33.875787: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
/home/shino/anaconda3/envs/movenet/lib/python3.10/runpy.py:126: RuntimeWarning: 'tf2onnx.convert' found in sys.modules after import of package 'tf2onnx', but prior to execution of 'tf2onnx.convert'; this may result in unpredictable behaviour
  warn(RuntimeWarning(msg))
2025-04-11 23:58:34.619575: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-11 23:58:34.622735: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-11 23:58:34.625844: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-11 23:58:34.625873: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublas.so.11'; dlerror: libcublas.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-11 23:58:34.625896: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublasLt.so.11'; dlerror: libcublasLt.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-11 23:58:34.625918: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcufft.so.10'; dlerror: libcufft.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-11 23:58:34.647138: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusparse.so.11'; dlerror: libcusparse.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-11 23:58:34.647185: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudnn.so.8'; dlerror: libcudnn.so.8: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-11 23:58:34.647189: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2025-04-11 23:58:34.647389: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2025-04-11 23:58:34,649 - INFO - Using tensorflow=2.9.0, onnx=1.14.1, tf2onnx=1.13.0/2c1db5
2025-04-11 23:58:34,649 - INFO - Using opset <onnx, 13>
INFO: Created TensorFlow Lite XNNPACK delegate for CPU.
Traceback (most recent call last):
  File "/home/shino/anaconda3/envs/movenet/lib/python3.10/runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/home/shino/anaconda3/envs/movenet/lib/python3.10/runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "/home/shino/.local/lib/python3.10/site-packages/tf2onnx/convert.py", line 706, in <module>
    main()
  File "/home/shino/.local/lib/python3.10/site-packages/tf2onnx/convert.py", line 269, in main
    model_proto, _ = _convert_common(
  File "/home/shino/.local/lib/python3.10/site-packages/tf2onnx/convert.py", line 164, in _convert_common
    g = process_tf_graph(tf_graph, const_node_values=const_node_values,
  File "/home/shino/.local/lib/python3.10/site-packages/tf2onnx/tfonnx.py", line 453, in process_tf_graph
    main_g, subgraphs = graphs_from_tflite(tflite_path, input_names, output_names)
  File "/home/shino/.local/lib/python3.10/site-packages/tf2onnx/tflite_utils.py", line 153, in graphs_from_tflite
    parse_tflite_graph(tfl_graph, opcodes, model, prefix, tensor_shapes_from_interpreter)
  File "/home/shino/.local/lib/python3.10/site-packages/tf2onnx/tflite_utils.py", line 485, in parse_tflite_graph
    onnx_node = helper.make_node(optype, input_names, output_names, name=output_names[0], **attr)
IndexError: list index out of range

Further information

• Is this issue related to a specific model?
  Model name:

https://drive.google.com/file/d/149nTlPi0WlQZn3F_IAUlzfWkZkMnVok1/view?usp=drive_link

The model downloaded from Kaggle was converted to TFlite using the following source code:

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
import tensorflow_hub as hub

# https://www.kaggle.com/models/deepmind/biggan
model_url = "https://www.kaggle.com/models/deepmind/biggan/TensorFlow1/128/2"
input_truncation = keras.Input(batch_shape=(), name='truncation')
input_y = keras.Input(shape=(1000, ), name='y')
input_z = keras.Input(shape=(120, ), name='z')
hub_layer = hub.KerasLayer(
            model_url,
            trainable=False,
            signature="default", 
            signature_outputs_as_dict=True,
        )

inputs = dict(y=input_y, z=input_z, truncation=input_truncation)
output = hub_layer(inputs)
model = tf.keras.models.Model(inputs=[input_truncation, input_y, input_z], outputs=[output])
model.build(([1], [1, 1000], [1, 120]))
model.summary()
model.save("biggan-128")

converter = tf.lite.TFLiteConverter.from_keras_model(model)
tflite_model = converter.convert()
open("biggan-128.tflite", "wb").write(tflite_model)

Model opset:

Notes

[xtrizeShino]

Thinking that outputs[0] could not be referenced from the line where "IndexError: list index out of range" occurred, I implemented the following source code and was able to generate an ONNX format file.

However, although I am expecting an image as output from this model, it is being output in NHWC format instead of NCHW format. How can I make it into NCHW format?

Also, as my goal is to convert using a command, I would be grateful if you could continue to teach me how to use the command.

Source code for conversion

import numpy as np
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
import tensorflow_hub as hub
import tf2onnx
import onnx
import onnxruntime
from onnxruntime.datasets import get_example

model_url = "https://www.kaggle.com/models/deepmind/biggan/TensorFlow1/128/2"
input_truncation = keras.Input(batch_shape=(), name='truncation')
input_y = keras.Input(shape=(1000, ), name='y')
input_z = keras.Input(shape=(120, ), name='z')
hub_layer = hub.KerasLayer(
            model_url,
            trainable=False,
            signature="default", 
            signature_outputs_as_dict=True,
        )

inputs = dict(y=input_y, z=input_z, truncation=input_truncation)
output = hub_layer(inputs)
model = tf.keras.models.Model(inputs=[input_truncation, input_y, input_z], outputs=[output])
model.build(([1], [1, 1000], [1, 120]))
model.summary()
model.save("biggan-128")

converter = tf.lite.TFLiteConverter.from_keras_model(model)
tflite_model = converter.convert()
open("biggan-128.tflite", "wb").write(tflite_model)

onnx_model, _ = tf2onnx.convert.from_keras(model, outputs_as_nchw=[model.outputs[0].name])
onnx.save(onnx_model, './biggan-128.onnx')

sess = onnxruntime.InferenceSession('./biggan-128.onnx', providers=['CPUExecutionProvider'])

inputs_onnx_model = sess.get_inputs()
for i in range(len(inputs_onnx_model)):
    input_name = inputs_onnx_model[i].name
    print("Input name  :", input_name)
    input_shape = inputs_onnx_model[i].shape
    print("Input shape :", input_shape)
    input_type = inputs_onnx_model[i].type
    print("Input type  :", input_type)
    print()

outputs_onnx_model = sess.get_outputs()
for i in range(len(outputs_onnx_model)):
    output_name = outputs_onnx_model[i].name
    print("Output name  :", output_name)  
    output_shape = outputs_onnx_model[i].shape
    print("Output shape :", output_shape)
    output_type = outputs_onnx_model[i].type
    print("Output type  :", output_type)

output

2025-04-12 11:18:20.264002: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2025-04-12 11:18:20.266246: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-12 11:18:20.266256: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2025-04-12 11:18:22.797739: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:18:22.800921: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:18:22.804080: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-12 11:18:22.804108: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublas.so.11'; dlerror: libcublas.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-12 11:18:22.804129: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublasLt.so.11'; dlerror: libcublasLt.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-12 11:18:22.804150: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcufft.so.10'; dlerror: libcufft.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-12 11:18:22.825051: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusparse.so.11'; dlerror: libcusparse.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-12 11:18:22.825091: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudnn.so.8'; dlerror: libcudnn.so.8: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64:/usr/local/cuda/lib64:
2025-04-12 11:18:22.825095: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2025-04-12 11:18:22.825841: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Model: "model"
__________________________________________________________________________________________________
 Layer (type)                   Output Shape         Param #     Connected to                     
==================================================================================================
 truncation (InputLayer)        [None]               0           []                               
                                                                                                  
 y (InputLayer)                 [(None, 1000)]       0           []                               
                                                                                                  
 z (InputLayer)                 [(None, 120)]        0           []                               
                                                                                                  
 keras_layer (KerasLayer)       {'default': (None,   141028845   ['truncation[0][0]',             
                                128, 128, 3)}                     'y[0][0]',                      
                                                                  'z[0][0]']                      
                                                                                                  
==================================================================================================
Total params: 141,028,845
Trainable params: 0
Non-trainable params: 141,028,845
__________________________________________________________________________________________________
WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.
WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.
2025-04-12 11:18:38.667775: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.
2025-04-12 11:18:38.667802: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.
2025-04-12 11:18:38.668199: I tensorflow/cc/saved_model/reader.cc:43] Reading SavedModel from: /tmp/tmp21v40ktt
2025-04-12 11:18:38.674642: I tensorflow/cc/saved_model/reader.cc:81] Reading meta graph with tags { serve }
2025-04-12 11:18:38.674660: I tensorflow/cc/saved_model/reader.cc:122] Reading SavedModel debug info (if present) from: /tmp/tmp21v40ktt
2025-04-12 11:18:38.699266: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:354] MLIR V1 optimization pass is not enabled
2025-04-12 11:18:38.706981: I tensorflow/cc/saved_model/loader.cc:228] Restoring SavedModel bundle.
2025-04-12 11:18:39.108084: I tensorflow/cc/saved_model/loader.cc:212] Running initialization op on SavedModel bundle at path: /tmp/tmp21v40ktt
2025-04-12 11:18:39.350189: I tensorflow/cc/saved_model/loader.cc:301] SavedModel load for tags { serve }; Status: success: OK. Took 681991 microseconds.
2025-04-12 11:18:39.624210: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:263] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.
2025-04-12 11:19:37.612637: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:37.616963: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:37.619823: I tensorflow/core/grappler/devices.cc:66] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 2
2025-04-12 11:19:37.619905: I tensorflow/core/grappler/clusters/single_machine.cc:358] Starting new session
2025-04-12 11:19:37.779830: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:37.781055: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:37.782165: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2025-04-12 11:19:39.151057: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:39.152276: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:39.153436: I tensorflow/core/grappler/devices.cc:66] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 2
2025-04-12 11:19:39.153526: I tensorflow/core/grappler/clusters/single_machine.cc:358] Starting new session
2025-04-12 11:19:39.153939: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:39.155052: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:39.156199: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
WARNING:tf2onnx.tf_loader:TF freezing failed. Attempting to fix freezing errors.
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_64
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_59
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_38
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_63
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_58
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_46
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_44
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_62
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_52
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_45
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_42
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_34
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_61
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_56
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_50
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_41
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_33
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_60
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_57
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_49
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_40
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_36
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_55
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_48
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_43
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_54
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_51
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_39
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_53
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_47
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_37
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_35
WARNING:tf2onnx.tf_loader:Removed Identity model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_65
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_28
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_21
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_11
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_4
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_27
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_24
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_18
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_32
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_26
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_20
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_16
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_10
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_25
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_2
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_15
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_29
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_19
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_30
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_3
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_23
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_14
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_1
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_9
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_8
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_17
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_7
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_13
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_6
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_31
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_22
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_12
WARNING:tf2onnx.tf_loader:Removed Mul model/keras_layer/StatefulPartitionedCall/cond/cond/AssignVariableOp_5
WARNING:tf2onnx.tf_loader:Removed Const model/keras_layer/StatefulPartitionedCall/cond/AssignVariableOp
2025-04-12 11:19:40.783754: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:40.784985: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:40.786056: I tensorflow/core/grappler/devices.cc:66] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 2
2025-04-12 11:19:40.786112: I tensorflow/core/grappler/clusters/single_machine.cc:358] Starting new session
2025-04-12 11:19:40.786373: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:40.787459: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2025-04-12 11:19:40.788598: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
WARNING:tf2onnx.shape_inference:Shape of placeholder 'truncation' is unknown, treated it as a scalar. Please use the --inputs flag and append the shape to the input name if this input is not a scalar.
Input name  : truncation
Input shape : []
Input type  : tensor(float)

Input name  : y
Input shape : ['unk__234', 1000]
Input type  : tensor(float)

Input name  : z
Input shape : ['unk__235', 120]
Input type  : tensor(float)

Output name  : keras_layer
Output shape : ['unk__236', 128, 128, 3]
Output type  : tensor(float)

requitrements.txt

absl-py==2.2.1
astunparse==1.6.3
cachetools==5.5.2
certifi==2025.1.31
charset-normalizer==3.4.1
coloredlogs==15.0.1
flatbuffers==1.12
gast==0.4.0
google-auth==2.38.0
google-auth-oauthlib==0.4.6
google-pasta==0.2.0
grpcio==1.71.0
h5py==3.13.0
humanfriendly==10.0
idna==3.10
keras==2.9.0
Keras-Preprocessing==1.1.2
libclang==18.1.1
Markdown==3.7
MarkupSafe==3.0.2
mpmath==1.3.0
numpy==1.23.5
oauthlib==3.2.2
onnx==1.14.1
onnx-graphsurgeon==0.5.7
onnx2tf==1.26.9
onnxruntime==1.21.0
opt_einsum==3.4.0
packaging==24.2
protobuf==3.20.3
psutil==7.0.0
pyasn1==0.6.1
pyasn1_modules==0.4.2
requests==2.32.3
requests-oauthlib==2.0.0
rsa==4.9
six==1.17.0
sng4onnx==1.0.4
sympy==1.13.3
tensorboard==2.9.0
tensorboard-data-server==0.6.1
tensorboard-plugin-wit==1.8.1
tensorflow==2.9.0
tensorflow-estimator==2.9.0
tensorflow-hub==0.16.1
tensorflow-io-gcs-filesystem==0.37.1
tensorflow-neuron==1.0
termcolor==3.0.0
tf-keras==2.14.1
tf2onnx==1.13.0
typing_extensions==4.13.0
urllib3==2.3.0
Werkzeug==3.1.3
wrapt==1.17.2