添加自定义指标 Keras 子类化 API

Question

我正在关注 "Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition - Aurélien Geron" 第 12 章的 "Losses and Metrics Based on Model Internals" 部分，他在其中展示了如何添加不依赖于标签和预测的自定义损失和指标。

为了说明这一点，我们添加了一个自定义 "reconstruction loss"，方法是在应该重现输入的上层隐藏层之上添加一个层。损失是重建损失和输入之间的均方差。

他展示了添加自定义损失的代码，效果很好，但即使按照他的描述我也无法添加指标，因为它会引发“ValueError”。他说：

Similarly, you can add a custom metric based on model internals by computing it in any way you want, as long as the result is the output of a metric object. For example, you can create a keras.metrics.Mean object in the constructor, then call it in the call() method, passing it the recon_loss, and finally add it to the model by calling the model’s add_metric() method.

这是代码（我已经为我自己添加的行添加了#MINE）

import tensorflow as tf
from tensorflow import keras
class ReconstructingRegressor(keras.models.Model):
    def __init__(self, output_dim, **kwargs):
        super().__init__(**kwargs)
        self.hidden = [keras.layers.Dense(30, activation="selu",
                                          kernel_initializer="lecun_normal")
                       for _ in range(5)]
        self.out = keras.layers.Dense(output_dim)
        self.reconstruction_mean = keras.metrics.Mean(name="reconstruction_error") #MINE

    def build(self, batch_input_shape):
        n_inputs = batch_input_shape[-1]
        self.reconstruct = keras.layers.Dense(n_inputs)
        super().build(batch_input_shape)

    def call(self, inputs, training=None):
        Z = inputs
        for layer in self.hidden:
            Z = layer(Z)
        reconstruction = self.reconstruct(Z)
        recon_loss = tf.reduce_mean(tf.square(reconstruction - inputs))
        self.add_loss(0.05 * recon_loss)
        if training:                                      #MINE
            result = self.reconstruction_mean(recon_loss) #MINE
        else:                                             #MINE
            result = 0.                                   #MINE, I have also tried different things here, 
                                                          #but the help showed a similar sample to this.
        self.add_metric(result, name="foo")               #MINE
        return self.out(Z)

然后编译和拟合模型：

training_set_size=10
X_dummy = np.random.randn(training_set_size, 8) 
y_dummy = np.random.randn(training_set_size, 1)

model = ReconstructingRegressor(1)
model.compile(loss="mse", optimizer="nadam")
history = model.fit(X_dummy, y_dummy, epochs=2)

抛出：


ValueError: in converted code:

    <ipython-input-296-878bdeb30546>:26 call  *
        self.add_metric(result, name="foo")               #MINE
    C:\Users\Kique\Anaconda3\envs\piz3\lib\site-packages\tensorflow_core\python\keras\engine\base_layer.py:1147 add_metric
        self._symbolic_add_metric(value, aggregation, name)
    C:\Users\Kique\Anaconda3\envs\piz3\lib\site-packages\tensorflow_core\python\keras\engine\base_layer.py:1867 _symbolic_add_metric
        'We do not support adding an aggregated metric result tensor that '

    ValueError: We do not support adding an aggregated metric result tensor that is not the output of a `tf.keras.metrics.Metric` metric instance. Without having access to the metric instance we cannot reset the state of a metric after every epoch during training. You can create a `tf.keras.metrics.Metric` instance and pass the result here or pass an un-aggregated result with `aggregation` parameter set as `mean`. For example: `self.add_metric(tf.reduce_sum(inputs), name='mean_activation', aggregation='mean')`

读完之后，我尝试了类似的方法来解决这个问题，但它只是导致了不同的错误。我该如何解决这个问题？ "correct" 的方法是什么？

我在 Windows 上使用 conda，安装了 tensorflow-gpu 2.1.0。

Answer 1

问题就在这里：

def call(self, inputs, training=None):
    Z = inputs
    for layer in self.hidden:
        Z = layer(Z)
    reconstruction = self.reconstruct(Z)
    recon_loss = tf.reduce_mean(tf.square(reconstruction - inputs))
    self.add_loss(0.05 * recon_loss)
    if training:                                      
        result = self.reconstruction_mean(recon_loss) 
    else:                                             
        result = 0.#<---Here!                                          
    self.add_metric(result, name="foo")              
    return self.out(Z)

错误表明 add_metric 仅获取从 tf.keras.metrics.Metric 派生的度量，但 0 是标量，而不是度量类型。

我建议的解决方案是简单地这样做：

def call(self, inputs, training=None):
    Z = inputs
    for layer in self.hidden:
        Z = layer(Z)
    reconstruction = self.reconstruct(Z)
    recon_loss = tf.reduce_mean(tf.square(reconstruction - inputs))
    self.add_loss(0.05 * recon_loss)
    if training:                                      
        result = self.reconstruction_mean(recon_loss)                           
        self.add_metric(result, name="foo")              
    return self.out(Z)

这样，您的平均值 reconstruction_error 将仅在训练时间显示。

由于您使用的是急切模式，因此您应该使用 dynamic=True 创建图层，如下所示：

model = ReconstructingRegressor(1,dynamic=True)
model.compile(loss="mse", optimizer="nadam")
history = model.fit(X_dummy, y_dummy, epochs=2, batch_size=10)

P.S - 注意，在调用 model.fit or model.evaluate 时，您还应该确保批量大小划分您的训练集（因为这是一个有状态的网络）。因此，像这样调用这些函数：model.fit(X_dummy, y_dummy, epochs=2, batch_size=10) 或 model.evaluate(X_dummy,y_dummy, batch_size=10)。祝你好运！

添加自定义指标 Keras 子类化 API

Adding custom metric Keras Subclassing API

deep-learning

keras

tensorflow

keras-layer

tf.keras