将 tensorflow/keras 模型转换为 tensorflow lite 模型时出现的问题

Question

While experimenting with tensorflow i've come accross an issue when converting keras model into tensorflow lite .

This is my setup:

model = models.Sequential()
model.add(layers.Dense(128, activation='relu', input_shape=(1,)))
model.add(layers.Dense(64, activation='relu'))
model.add(layers.Dense(17, activation='softmax'))
model.compile(optimizer='adam',
              loss='categorical_crossentropy',
              metrics=['accuracy'])
...
model.save('Foo.h5')
...
converter = tf.contrib.lite.TFLiteConverter.from_keras_model_file("Foo.h5")
tflite_model = converter.convert()
open("Foo_converted.tflite", "wb").write(tflite_model)

And here is the output:

INFO:tensorflow:Froze 6 variables.
INFO:tensorflow:Converted 6 variables to const ops.

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-6-f0613acb22f5> in <module>()
      2 
      3 converter = tf.contrib.lite.TFLiteConverter.from_keras_model_file("Foo.h5")
----> 4 tflite_model = converter.convert()
      5 open("converted_model.tflite", "wb").write(tflite_model)

c:\users\dev\appdata\local\programs\python\python36\lib\site-packages\tensorflow\contrib\lite\python\lite.py in convert(self)
    451           input_tensors=self._input_tensors,
    452           output_tensors=self._output_tensors,
--> 453           **converter_kwargs)
    454     else:
    455       # Graphs without valid tensors cannot be loaded into tf.Session since they

c:\users\dev\appdata\local\programs\python\python36\lib\site-packages\tensorflow\contrib\lite\python\convert.py in toco_convert_impl(input_data, input_tensors, output_tensors, *args, **kwargs)
    340   data = toco_convert_protos(model_flags.SerializeToString(),
    341                              toco_flags.SerializeToString(),
--> 342                              input_data.SerializeToString())
    343   return data
    344 

c:\users\dev\appdata\local\programs\python\python36\lib\site-packages\tensorflow\contrib\lite\python\convert.py in toco_convert_protos(model_flags_str, toco_flags_str, input_data_str)
    133     else:
    134       raise RuntimeError("TOCO failed see console for info.\n%s\n%s\n" %
--> 135                          (stdout, stderr))
    136 
    137 

RuntimeError: TOCO failed see console for info.
b'Traceback (most recent call last):\r\n  File "c:\\users\\dev\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\contrib\\lite\\toco\\python\\tensorflow_wrap_toco.py", line 18, in swig_import_helper\r\n    fp, pathname, description = imp.find_module(\'_tensorflow_wrap_toco\', [dirname(__file__)])\r\n  File "c:\\users\\dev\\appdata\\local\\programs\\python\\python36\\lib\\imp.py", line 297, in find_module\r\n    raise ImportError(_ERR_MSG.format(name), name=name)\r\nImportError: No module named \'_tensorflow_wrap_toco\'\r\n\r\nDuring handling of the above exception, another exception occurred:\r\n\r\nTraceback (most recent call last):\r\n  File "c:\\users\\dev\\appdata\\local\\programs\\python\\python36\\lib\\runpy.py", line 193, in _run_module_as_main\r\n    "__main__", mod_spec)\r\n  File "c:\\users\\dev\\appdata\\local\\programs\\python\\python36\\lib\\runpy.py", line 85, in _run_code\r\n    exec(code, run_globals)\r\n  File "C:\\Users\\Dev\\AppData\\Local\\Programs\\Python\\Python36\\Scripts\\toco_from_protos.exe\\__main__.py", line 5, in <module>\r\n  File "c:\\users\\dev\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\contrib\\lite\\toco\\python\\toco_from_protos.py", line 22, in <module>\r\n    from tensorflow.contrib.lite.toco.python import tensorflow_wrap_toco\r\n  File "c:\\users\\dev\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\contrib\\lite\\toco\\python\\tensorflow_wrap_toco.py", line 28, in <module>\r\n    _tensorflow_wrap_toco = swig_import_helper()\r\n  File "c:\\users\\dev\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\contrib\\lite\\toco\\python\\tensorflow_wrap_toco.py", line 20, in swig_import_helper\r\n    import _tensorflow_wrap_toco\r\nModuleNotFoundError: No module named \'_tensorflow_wrap_toco\'\r\n'
None

If someone could explain why the conversion fails that would be great.

Answer 1

The TfliteConverter and TocoConverter seem to be very problematic on all OS except Mac. You can still convert the model to TensorFlow Lite using the following steps:

Create a new Google Colab notebook

Write the code to convert the model. Import TFLiteConverter and other stuff.

Upload the model in the notebook.

Run the notebook.

Download the generated TFLite file.

You can use this notebook .

Answer 2

This looks like https://github.com/tensorflow/tensorflow/issues/22617 where the _tensorflow_wrap_toco library is missing in Windows version. I would suggest you to try installing the latest Tensorflow version.