Tensorflow keras到tf.keras转换：未定义致密层尺寸？

因此，我使用纯

keras

构建了一个convnet。它完全按照预期进行编译和操作，但我需要将其转换为使用

tf.keras

，以便我可以使用

tfmot

。阅读了文档后，我试图转换它，但出现以下错误：

应定义密集输入的最后一个维度。未找到任何文件。

知道我做错了什么吗

谢谢

原装

keras

型号：

input_layer = keras.layers.Input(shape=(100,))
reshape_layer = keras.layers.Reshape((-1, 100, 1))(input_layer)
conv_layer_1 = keras.layers.Convolution2D(filters=30, kernel_size=(10, 1), strides=(1, 1), padding="same", activation="relu")(reshape_layer)
conv_layer_2 = keras.layers.Convolution2D(filters=30, kernel_size=(8, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_1)
conv_layer_3 = keras.layers.Convolution2D(filters=40, kernel_size=(6, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_2)
conv_layer_4 = keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_3)
conv_layer_5 = keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_4)
flatten_layer = keras.layers.Flatten()(conv_layer_5)
label_layer = keras.layers.Dense(200, activation="relu")(flatten_layer)
output_layer = keras.layers.Dense(1, activation="linear")(label_layer)

model = keras.Model(inputs=input_layer, outputs=output_layer)

input_layer = tf.keras.layers.InputLayer(input_shape=(100,))
reshape_layer = tf.keras.layers.Reshape((-1, 100, 1))(input_layer)
conv_layer_1 = tf.keras.layers.Convolution2D(filters=30, kernel_size=(10, 1), strides=(1, 1), padding="same", activation="relu")(reshape_layer)
conv_layer_2 = tf.keras.layers.Convolution2D(filters=30, kernel_size=(8, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_1)
conv_layer_3 = tf.keras.layers.Convolution2D(filters=40, kernel_size=(6, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_2)
conv_layer_4 = tf.keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_3)
conv_layer_5 = tf.keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_4)
flatten_layer = tf.keras.layers.Flatten()(conv_layer_5)
label_layer = tf.keras.layers.Dense(200, activation="relu")(flatten_layer)
output_layer = tf.keras.layers.Dense(1, activation="linear")(label_layer)

model = tf.keras.Model(inputs=input_layer, outputs=output_layer)

转换后的

tf.keras

型号：

input_layer = keras.layers.Input(shape=(100,))
reshape_layer = keras.layers.Reshape((-1, 100, 1))(input_layer)
conv_layer_1 = keras.layers.Convolution2D(filters=30, kernel_size=(10, 1), strides=(1, 1), padding="same", activation="relu")(reshape_layer)
conv_layer_2 = keras.layers.Convolution2D(filters=30, kernel_size=(8, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_1)
conv_layer_3 = keras.layers.Convolution2D(filters=40, kernel_size=(6, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_2)
conv_layer_4 = keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_3)
conv_layer_5 = keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_4)
flatten_layer = keras.layers.Flatten()(conv_layer_5)
label_layer = keras.layers.Dense(200, activation="relu")(flatten_layer)
output_layer = keras.layers.Dense(1, activation="linear")(label_layer)

model = keras.Model(inputs=input_layer, outputs=output_layer)

input_layer = tf.keras.layers.InputLayer(input_shape=(100,))
reshape_layer = tf.keras.layers.Reshape((-1, 100, 1))(input_layer)
conv_layer_1 = tf.keras.layers.Convolution2D(filters=30, kernel_size=(10, 1), strides=(1, 1), padding="same", activation="relu")(reshape_layer)
conv_layer_2 = tf.keras.layers.Convolution2D(filters=30, kernel_size=(8, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_1)
conv_layer_3 = tf.keras.layers.Convolution2D(filters=40, kernel_size=(6, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_2)
conv_layer_4 = tf.keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_3)
conv_layer_5 = tf.keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_4)
flatten_layer = tf.keras.layers.Flatten()(conv_layer_5)
label_layer = tf.keras.layers.Dense(200, activation="relu")(flatten_layer)
output_layer = tf.keras.layers.Dense(1, activation="linear")(label_layer)

model = tf.keras.Model(inputs=input_layer, outputs=output_layer)

编辑1：

---

我想也许我可以在创建

keras

模型后保存它，并在编译/培训之前立即将其加载为

tf.keras

模型，从而绕过这个问题。这也会犯同样的错误

您的代码几乎没有问题。修好了，你就可以走了

问题1：使用

Input

而不是

InputLayer

标准是使用

Input

层而不是

InputLayer

（实际上在内部使用

InputLayer

）。如果使用的是

input

图层，则还需要将

input\u shape

更改为

shape

input_layer = tf.keras.layers.Input(shape=(100,))

问题2:2

None

输出中的维度 执行以下行时，输出中会出现两个

None

维度

reshape_layer = tf.keras.layers.Reshape((-1, 100, 1))(input_layer)

这就是为什么会出现上述错误。定义

重塑

层时，不定义

批次

维度，该维度将为无。如果要使用

密集

层，这是唯一可以作为

None

的维度。否则，

稠密

层无法推断其权重的形状（这就是为什么会出现错误）。那么换成,

reshape_layer = tf.keras.layers.Reshape((1, 100, 1))(input_layer)

其余的保持不变

conv_layer_1 = tf.keras.layers.Convolution2D(filters=30, kernel_size=(10, 1), strides=(1, 1), padding="same", activation="relu")(reshape_layer)
conv_layer_2 = tf.keras.layers.Convolution2D(filters=30, kernel_size=(8, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_1)
conv_layer_3 = tf.keras.layers.Convolution2D(filters=40, kernel_size=(6, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_2)
conv_layer_4 = tf.keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_3)
conv_layer_5 = tf.keras.layers.Convolution2D(filters=50, kernel_size=(5, 1), strides=(1, 1), padding="same", activation="relu")(conv_layer_4)
flatten_layer = tf.keras.layers.Flatten()(conv_layer_5)
label_layer = tf.keras.layers.Dense(200, activation="relu")(flatten_layer)
output_layer = tf.keras.layers.Dense(1, activation="linear")(label_layer)

model = tf.keras.Model(inputs=input_layer, outputs=output_layer)
model.summary()