Model fit / TypeError: 'NoneType' object 不可調用

Question

我正在嘗試使用 ResNet50 和我自己的分類器在分類任務上進行遷移學習，我有 28 個類要預測，我被困在model.fit() 。

我正在使用數據生成器為 model 提供

this is my model:
train_datagen = ImageDataGenerator(
rescale=1./255,

)
test_datagen = ImageDataGenerator(rescale=1./255)

train_dir = "/content/arabic/training"
train_generator = train_datagen.flow_from_directory(
train_dir,
color_mode = "rgb",
target_size=(32, 32),
batch_size=20,
class_mode='categorical')

validation_dir="/content/arabic/validation"
validation_generator = test_datagen.flow_from_directory(
validation_dir,
target_size=(32, 32),
batch_size=20,
class_mode='categorical')


base_model = keras.applications.ResNet50(
    weights='imagenet',  # Load weights pre-trained on ImageNet.
    input_shape=(32, 32, 3),
    include_top=False)  # Do not include the ImageNet classifier at the top.

def modelFunctionTransfer(l2 = 0.001, lr=0.001, drop = 0.2, modelName = None):
  model = models.Sequential()
  if(modelName != None):
    model.add(modelName)

  model.add(layers.Flatten())
  model.add(layers.Dense(512, kernel_regularizer=regularizers.l2(l2), activation='relu'))
  model.add(layers.Dense(28, activation='softmax'))
  opt = keras.optimizers.Adam(learning_rate=lr)
  model.compile(loss="categorical_crossentropy",
  optimizer=opt,
  metrics=['acc'])
  return model

model = modelFunctionTransfer(l2 = 0.001, lr = 0.001, drop = 0.5, modelName=base_model)
model.summary()

Model: "sequential_1"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
resnet50 (Functional)        (None, 1, 1, 2048)        23587712  
_________________________________________________________________
flatten_1 (Flatten)          (None, 2048)              0         
_________________________________________________________________
dense_2 (Dense)              (None, 512)               1049088   
_________________________________________________________________
dense_3 (Dense)              (None, 28)                14364     
=================================================================
Total params: 24,651,164
Trainable params: 1,063,452
Non-trainable params: 23,587,712

這是我得到錯誤的地方：

history = model.fit(train_generator,
                    steps_per_epoch = 672,
                    epochs = 30,
                    validation_data = validation_generator,
                    validation_steps = 30,
                    )

作為參考，這是我收到的完整錯誤消息：

Epoch 1/30
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-46-1dfe32a08202> in <module>()
     10 history = model.fit(train_generator,
     11                     steps_per_epoch = 100,
---> 12                     epochs = 30,
     13 
     14                     )

2 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)
   1098                 _r=1):
   1099               callbacks.on_train_batch_begin(step)
-> 1100               tmp_logs = self.train_function(iterator)
   1101               if data_handler.should_sync:
   1102                 context.async_wait()

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py in __call__(self, *args, **kwds)
    826     tracing_count = self.experimental_get_tracing_count()
    827     with trace.Trace(self._name) as tm:
--> 828       result = self._call(*args, **kwds)
    829       compiler = "xla" if self._experimental_compile else "nonXla"
    830       new_tracing_count = self.experimental_get_tracing_count()

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py in _call(self, *args, **kwds)
    853       # In this case we have created variables on the first call, so we run the
    854       # defunned version which is guaranteed to never create variables.
--> 855       return self._stateless_fn(*args, **kwds)  # pylint: disable=not-callable
    856     elif self._stateful_fn is not None:
    857       # Release the lock early so that multiple threads can perform the call

TypeError: 'NoneType' object is not callable

Answer 1

我復制了你的代碼，稍作改動。 我沒有你的數據集，所以我使用了只有 2 個類的數據集。 我使用的代碼在下面並且運行沒有錯誤。

train_datagen = ImageDataGenerator(
rescale=1./255,

)
test_datagen = ImageDataGenerator(rescale=1./255)

train_dir = r'c:\temp\people\train'
train_generator = train_datagen.flow_from_directory(
train_dir,
color_mode = "rgb",
target_size=(32, 32),
batch_size=20,
class_mode='categorical')

validation_dir=r'c:\temp\people\test'
validation_generator = test_datagen.flow_from_directory(
validation_dir,
target_size=(32, 32),
batch_size=20,
class_mode='categorical')


base_model = keras.applications.ResNet50(
    weights='imagenet',  # Load weights pre-trained on ImageNet.
    input_shape=(32, 32, 3),
    include_top=False)  # Do not include the ImageNet classifier at the top.

def modelFunctionTransfer(l2 = 0.001, lr=0.001, drop = 0.2, modelName = None):
  model = Sequential()
  if(modelName != None):
    model.add(modelName)

  model.add(tf.keras.layers.Flatten())
  model.add(tf.keras.layers.Dense(512, kernel_regularizer=regularizers.l2(l2), 
                               activation='relu'))
  model.add(tf.keras.layers.Dense(2, activation='softmax'))
  opt = keras.optimizers.Adam(learning_rate=.001)
  model.compile(loss="categorical_crossentropy",
  optimizer=opt,
  metrics=['acc'])
  return model

model = modelFunctionTransfer(l2 = 0.001, lr = 0.001, drop = 0.5, modelName=base_model)
model.summary()
history = model.fit(train_generator, epochs = 30, validation_data = validation_generator)

在 model.fit 中，我通常不使用 steps_per_epoch 和 validation_steps，讓 model.fit 在內部計算這些值。