Keras model.fit() 发出 TypeError: 'NoneType' object is not callable

Question

我正在尝试使用 Keras 来实现 Siamese Network 以实现一次性人脸识别模型。但是我遇到了一个我无法理解的错误，请求帮助。

我使用的模型是一个编码器模型，它接收(299,299,3) 图像（锚图像、正图像以及负图像）并输出每个图像的1000 维度编码向量。这类似于带有分类头的 InceptionV3 模型。我也在使用自定义的三元组损失函数。我的模型如下：

class SiameseNet(tf.keras.layers.Layer):

  def __init__(self, model):
    self.model = model  # This is the image feature extraction model (similar to InceptionV3)
    super().__init__()

  def call(self, feat):
    feats = self.model(feat[0])
    nfeats = self.model(feat[1])

    return [feats, nfeats]

损失函数如下

def triplet_loss(y_true, y_pred, alpha=1e-2):
  return max(tf.reduce_sum((y_pred[0]-y_true)**2 - (y_pred[0]-y_pred[1])**2) + alpha, 0)

共有三个数组，分别名为 images（锚定图像）和 negatives（负图像），形状均为 (500,299,299,3)（其中 500 是训练示例的数量）和 positives（正图像特征），形状均为(500,1000)。所有这些都是 numpy 数组。

我的型号代码如下图

image_input = tf.keras.layers.Input(shape=(299,299,3), name='image_input')
negative_input = tf.keras.layers.Input(shape=(299,299,3), name='negative_input')

siamese = SiameseNet(image_features_extract_model)([image_input, negative_input])

model = tf.keras.Model(inputs=[image_input, negative_input], outputs=siamese)

model.compile(optimizer=tf.keras.optimizers.Adam(), loss=triplet_loss, metrics=['accuracy'])

编译与输出效果很好

Model: "functional_3"
__________________________________________________________________________________________________
Layer (type)                    Output Shape         Param #     Connected to                     
==================================================================================================
image_input (InputLayer)        [(None, 299, 299, 3) 0                                            
__________________________________________________________________________________________________
negative_input (InputLayer)     [(None, 299, 299, 3) 0                                            
__________________________________________________________________________________________________
siamese_net (SiameseNet)        [(None, 1000), (None 23851784    image_input[0][0]                
                                                                 negative_input[0][0]             
==================================================================================================
Total params: 23,851,784
Trainable params: 23,817,352
Non-trainable params: 34,432

但在运行时

model.fit([images, negatives], positives, epochs=10, batch_size=8, verbose=2)

我收到以下错误，需要帮助

Epoch 1/10
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-16-83443f79d005> in <module>()
----> 1 model.fit([images, negatives], positives, epochs=10, batch_size=8, verbose=2)
      2 # model.fit(train, epochs=10, verbose=2)

3 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py in _method_wrapper(self, *args, **kwargs)
    106   def _method_wrapper(self, *args, **kwargs):
    107     if not self._in_multi_worker_mode():  # pylint: disable=protected-access
--> 108       return method(self, *args, **kwargs)
    109 
    110     # Running inside `run_distribute_coordinator` already.

/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)
   1096                 batch_size=batch_size):
   1097               callbacks.on_train_batch_begin(step)
-> 1098               tmp_logs = train_function(iterator)
   1099               if data_handler.should_sync:
   1100                 context.async_wait()

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py in __call__(self, *args, **kwds)
    778       else:
    779         compiler = "nonXla"
--> 780         result = self._call(*args, **kwds)
    781 
    782       new_tracing_count = self._get_tracing_count()

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py in _call(self, *args, **kwds)
    805       # In this case we have created variables on the first call, so we run the
    806       # defunned version which is guaranteed to never create variables.
--> 807       return self._stateless_fn(*args, **kwds)  # pylint: disable=not-callable
    808     elif self._stateful_fn is not None:
    809       # Release the lock early so that multiple threads can perform the call

TypeError: 'NoneType' object is not callable

我在 CPU 上运行 Google Colab 上的代码。
请帮我解决这个问题。谢谢。

Answer 1

我不知道它是否正是你要找的，但我可以得到一个形状相似的模型，如下所示。

主要区别在于：

1. return np.stack([feats,nfeats])。
2. 整个模型是 Sequential 模型，所以我不依赖 keras 来构建基于 Input 张量的任何东西（第一层除外）。
3. SiameseNet 的输入是卷积层的输出，因此我对其进行切片以获得正确的形状。
4. 我对@9​​87654325@ 使用了一个简单的 MLP，但这只是为了让它更易于使用。

class SiameseNet(tf.keras.layers.Layer):

  def __init__(self, model):
    super().__init__()
    self.model = model  # This is the image feature extraction model (similar to InceptionV3)

  def call(self, feat):
    # I'm slicing the input this way because I'm expecting input from some convolutional layer
    feats = self.model(feat[:,:,0])
    nfeats = self.model(feat[:,:,1])

    # keras want this output insted of [feats, nfeats]
    return tf.stack([feats, nfeats])


def triplet_loss(y_true, y_pred, alpha=1e-2):
  return max(tf.reduce_sum((y_pred[0]-y_true)**2 - (y_pred[0]-y_pred[1])**2) + alpha, 0)


if __name__ == '__main__':
    # basic MLP for the siamese model
    mlp = tf.keras.models.Sequential()
    mlp.add(tf.keras.Input(shape=(10)))
    mlp.add(tf.keras.layers.Dense(units=10, activation='relu'))

    siamese = SiameseNet(mlp)

    # the "entire model"
    model = tf.keras.Sequential()
    # convolution is used here purely for convenience, to split an input into two channels
    model.add(tf.keras.layers.Conv1D(2,1))
    model.add(siamese)

    # your stuff
    model.compile(optimizer=tf.keras.optimizers.Adam(), loss=triplet_loss, metrics=['accuracy'])
    model.build(input_shape=[1,10,1])
    model.summary()

Answer 2

我想在这里回答我自己的问题，因为我所面临的问题完全不同，后来才弄清楚。特定于我的情况的问题是由于 triplet_loss 函数引起的。损失函数期望根据张量进行计算，而代码根据 Numpy 数组进行计算。更改此设置为我修复了错误，并且代码现在运行良好。

函数实现应该是

def triplet_loss(y_true, y_pred, alpha=0.2):
  return tf.maximum(tf.reduce_sum((y_pred[0]-y_true)**2) - tf.reduce_sum((y_pred[0]-y_pred[1])**2)) + tf.constant(alpha), tf.constant(0.0))

它适用于我的情况。无需进行其他更改。

P.S.：这里，alpha 需要比我之前选择的值 1e-2 更大，并且学习率需要通过使用降低到 1e-5

optimizer = tf.keras.optimizers.Adam(lr=1e-5)

在model.compile().