ValueError: Quantizing a tf.keras Model inside another tf.keras Model is not supported

Question

我刚开始使用 Keras/Tensorflow，我正在尝试重新训练和量化为 int8 a MobileNetV2，但我收到此错误：

ValueError: Quantizing a tf.keras Model inside another tf.keras Model is not supported.

我按照这个 guide 来绕过量化步骤，但我不确定我到底做了什么不同。

IMG_SHAPE = (224, 224, 3)
base_model = tf.keras.applications.MobileNetV2(input_shape=IMG_SHAPE,
                                                  include_top=False, 
                                                  weights='imagenet')
base_model.trainable = False
model = tf.keras.Sequential([
  base_model,
  tf.keras.layers.Conv2D(filters=32, kernel_size=3, activation='relu'),
  tf.keras.layers.Dropout(0.5),
  tf.keras.layers.MaxPool2D(pool_size=(2, 2)),
  tf.keras.layers.Flatten(),
  tf.keras.layers.Dense(units=2, activation='softmax')
])

quantize_model = tfmot.quantization.keras.quantize_model
q_aware_model = quantize_model(model)

堆栈跟踪：

ValueError                                Traceback (most recent call last)

<ipython-input-34-b724ad4872a5> in <module>()
      9 
     10 quantize_model = tfmot.quantization.keras.quantize_model
---> 11 q_aware_model = quantize_model(model)

4 frames

/usr/local/lib/python3.7/dist-packages/tensorflow_model_optimization/python/core/quantization/keras/quantize.py in _add_quant_wrapper(layer)
    217     if isinstance(layer, tf.keras.Model):
    218       raise ValueError(
--> 219           'Quantizing a tf.keras Model inside another tf.keras Model is not supported.'
    220       )
    221

Answer 1

在这种情况下，您的 base_model 的行为就好像它是一个图层。为了扩展它，您需要使用 Functional API，而不是 Sequential API:

IMG_SHAPE = (224, 224, 3)
base_model = tf.keras.applications.MobileNetV2(input_shape=IMG_SHAPE,
                                                  include_top=False, 
                                                  weights='imagenet')
base_model.trainable = False
x = tf.keras.layers.Conv2D(filters=32, kernel_size=3, activation='relu')(base_model.output)
x = tf.keras.layers.Dropout(0.5)(x)
x = tf.keras.layers.MaxPool2D(pool_size=(2, 2))(x)
x = tf.keras.layers.Flatten()(x)
x = tf.keras.layers.Dense(units=2, activation='softmax')(x)

model = tf.keras.Model(base_model.input, x)
model.summary()

请注意，模型摘要显示了所有层，包括 base_model's。那么您可以申请：

quantize_model = tfmot.quantization.keras.quantize_model
q_aware_model = quantize_model(model)

ValueError: Quantizing a tf.keras Model inside another tf.keras Model is not supported

ValueError: Quantizing a tf.keras Model inside another tf.keras Model is not supported

python

keras

tensorflow

tensorflow-lite

quantization-aware-training