使用keras和flask加载模型时出错

Question

I'm writing a simple application using flask. First I tried this:

def get_model():
    global model
    model = load_model('mobilenet.h5')
    print("** Model loaded!")

But I got an error:

Traceback (most recent call last):
  File "C:\Users\Agnieszka\Anaconda4\Scripts\flask-script.py", line 10, in <module>
    sys.exit(main())
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\flask\cli.py", line 894, in main
    cli.main(args=args, prog_name=name)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\flask\cli.py", line 557, in main
    return super(FlaskGroup, self).main(*args, **kwargs)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\click\core.py", line 697, in main
    rv = self.invoke(ctx)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\click\core.py", line 1066, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\click\core.py", line 895, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\click\core.py", line 535, in invoke
    return callback(*args, **kwargs)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\click\decorators.py", line 64, in new_func
    return ctx.invoke(f, obj, *args[1:], **kwargs)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\click\core.py", line 535, in invoke
    return callback(*args, **kwargs)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\flask\cli.py", line 767, in run_command
    app = DispatchingApp(info.load_app, use_eager_loading=eager_loading)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\flask\cli.py", line 293, in __init__
    self._load_unlocked()
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\flask\cli.py", line 317, in _load_unlocked
    self._app = rv = self.loader()
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\flask\cli.py", line 372, in load_app
    app = locate_app(self, import_name, name)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\flask\cli.py", line 235, in locate_app
    __import__(module_name)
  File "C:\Users\Agnieszka\Desktop\noc-naukowcow\flask_apps\predict.py", line 32, in <module>
    get_model()
  File "C:\Users\Agnieszka\Desktop\noc-naukowcow\flask_apps\predict.py", line 19, in get_model
    model = tf.keras.models.load_model('mobilenet.h5')
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\tensorflow\python\keras\engine\saving.py", line 249, in load_model
    optimizer_config, custom_objects=custom_objects)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\tensorflow\python\keras\optimizers.py", line 838, in deserialize
    printable_module_name='optimizer')
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\tensorflow\python\keras\utils\generic_utils.py", line 194, in deserialize_keras_object
    return cls.from_config(cls_config)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\tensorflow\python\keras\optimizers.py", line 159, in from_config
    return cls(**config)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\tensorflow\python\keras\optimizers.py", line 471, in __init__
    super(Adam, self).__init__(**kwargs)
  File "C:\Users\Agnieszka\Anaconda4\lib\site-packages\tensorflow\python\keras\optimizers.py", line 68, in __init__
    'passed to optimizer: ' + str(k))
TypeError: Unexpected keyword argument passed to optimizer: name

Then I changed the code using tf.keras.models.load_model but this haven't helped.

The model was built in Colaboratory using keras 2.2.4 version. I've built the model as follows:

from keras import backend as K

inputs = Input(shape=(96, 80, 3), name='input')


x = Conv2D(64, (3, 3), activation='relu')(inputs)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Dropout(0.25)(x)

x = Conv2D(128, (3, 3), activation='relu')(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Dropout(0.25)(x)

x = Conv2D(256, (3, 3), activation='relu')(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Dropout(0.25)(x)

x = Flatten()(x)
x = Dense(256, activation='relu')(x)
x = Dropout(0.5)(x)
x = Dense(35, activation='softmax')(x)

model = Model(inputs=inputs, outputs=x, name='CNN')



model.compile(optimizer=optimizer, loss=loss_fn, metrics=['acc'])

I wonder if the problem is due to optimaizer that I used (it was adam).

Is there any solution for this?

Answer 1

The error suggests that Unexpected keyword argument passed to optimizer: name . Now as we know, the argument name is used in TensorFlow models to name a specific op or tensor. TensorFlow has included Keras, but Keras still exists as a standalone package.

Maybe you have created a model using tf.keras but you are loading it using keras.models.load_model . The Keras packages can't recognize the TF argument.

Maybe you can try to bring the whole process in TensorFlow by using methods from tf.keras .