python - 在构建编码器/解码器模型之前训练自动编码器是否有效？

Question

我正在关注教程https://blog.keras.io/building-autoencoders-in-keras.html构建我的自动编码器。为此，我有两个策略:

A) 步骤1:构建自动编码器；步骤2:构建编码器；步骤3:构建解码器；步骤4:编译自动编码器；第5步:训练自动编码器。

B) 第 1 步:构建自动编码器；步骤2:编译自动编码器；步骤3:训练自动编码器；步骤4:构建编码器；第五步:构建解码器。

对于这两种情况，模型收敛到损失 0.100。然而，对于教程中所述的策略 ​​A，重建效果非常差。在策略 B 的情况下，重建要好得多。

在我看来，这是有道理的，因为在策略 A 中，编码器和解码器模型的权重是在未经训练的层上构建的，并且结果是随机的。另一方面，在策略 B 中，我在训练后更好地定义了权重，因此重建效果更好。

我的问题是，策略 B 有效还是我在重建过程中作弊？在策略 A 的情况下，Keras 是否应该自动更新编码器和解码器模型的权重，因为它们的模型是基于自动编码器层构建的？

###### Code for Strategy A

# Step 1
features = Input(shape=(x_train.shape[1],))

encoded = Dense(1426, activation='relu')(features)
encoded = Dense(732, activation='relu')(encoded)
encoded = Dense(328, activation='relu')(encoded)

encoded = Dense(encoding_dim, activation='relu')(encoded)

decoded = Dense(328, activation='relu')(encoded)
decoded = Dense(732, activation='relu')(decoded)
decoded = Dense(1426, activation='relu')(decoded)
decoded = Dense(x_train.shape[1], activation='relu')(decoded)

autoencoder = Model(inputs=features, outputs=decoded)

# Step 2
encoder = Model(features, encoded)

# Step 3
encoded_input = Input(shape=(encoding_dim,))
decoder_layer = autoencoder.layers[-4](encoded_input)
decoder_layer = autoencoder.layers[-3](decoder_layer)
decoder_layer = autoencoder.layers[-2](decoder_layer)
decoder_layer = autoencoder.layers[-1](decoder_layer)

decoder = Model(encoded_input, decoder_layer)

# Step 4
autoencoder.compile(optimizer='adam', loss='mse')

# Step 5
history = autoencoder.fit(x_train, 
                         x_train,
                         epochs=150,
                         batch_size=256,
                         shuffle=True,
                         verbose=1,
                         validation_split=0.2)

# Testing encoding
encoded_fts = encoder.predict(x_test)
decoded_fts = decoder.predict(encoded_fts)

###### Code for Strategy B

# Step 1
features = Input(shape=(x_train.shape[1],))

encoded = Dense(1426, activation='relu')(features)
encoded = Dense(732, activation='relu')(encoded)
encoded = Dense(328, activation='relu')(encoded)

encoded = Dense(encoding_dim, activation='relu')(encoded)

decoded = Dense(328, activation='relu')(encoded)
decoded = Dense(732, activation='relu')(decoded)
decoded = Dense(1426, activation='relu')(decoded)
decoded = Dense(x_train.shape[1], activation='relu')(decoded)

autoencoder = Model(inputs=features, outputs=decoded)

# Step 2
autoencoder.compile(optimizer='adam', loss='mse')

# Step 3
history = autoencoder.fit(x_train, 
                         x_train,
                         epochs=150,
                         batch_size=256,
                         shuffle=True,
                         verbose=1,
                         validation_split=0.2)
# Step 4
encoder = Model(features, encoded)

# Step 5
encoded_input = Input(shape=(encoding_dim,))
decoder_layer = autoencoder.layers[-4](encoded_input)
decoder_layer = autoencoder.layers[-3](decoder_layer)
decoder_layer = autoencoder.layers[-2](decoder_layer)
decoder_layer = autoencoder.layers[-1](decoder_layer)

decoder = Model(encoded_input, decoder_layer)

# Testing encoding
encoded_fts = encoder.predict(x_test)
decoded_fts = decoder.predict(encoded_fts)

Answer 1

My questions are, is strategy B valid or I am cheating on the reconstruction?

A 和 B 是等效的；不，你没有作弊。

In case of strategy A, is Keras supposed to update the weights of the encoder and decoder models automatically since their models were built based on the autoencoder layers?

解码器模型仅使用自动编码器层。如果A:

decoder.layers
Out:
[<keras.engine.input_layer.InputLayer at 0x7f8a44d805c0>,
 <keras.layers.core.Dense at 0x7f8a44e58400>,
 <keras.layers.core.Dense at 0x7f8a44e746d8>,
 <keras.layers.core.Dense at 0x7f8a44e14940>,
 <keras.layers.core.Dense at 0x7f8a44e2dba8>]

autoencoder.layers
Out:[<keras.engine.input_layer.InputLayer at 0x7f8a44e91c18>,
 <keras.layers.core.Dense at 0x7f8a44e91c50>,
 <keras.layers.core.Dense at 0x7f8a44e91ef0>,
 <keras.layers.core.Dense at 0x7f8a44e89080>,
 <keras.layers.core.Dense at 0x7f8a44e89da0>,
 <keras.layers.core.Dense at 0x7f8a44e58400>,
 <keras.layers.core.Dense at 0x7f8a44e746d8>,
 <keras.layers.core.Dense at 0x7f8a44e14940>,
 <keras.layers.core.Dense at 0x7f8a44e2dba8>]

每个列表最后 4 行的十六进制数字(对象 ID)完全相同 - 因为它们是相同的对象。当然，他们也分享自己的体重。

如果B:

decoder.layers
Out:
[<keras.engine.input_layer.InputLayer at 0x7f8a41de05f8>,
 <keras.layers.core.Dense at 0x7f8a41ee4828>,
 <keras.layers.core.Dense at 0x7f8a41eaceb8>,
 <keras.layers.core.Dense at 0x7f8a41e50ac8>,
 <keras.layers.core.Dense at 0x7f8a41e5d780>]

autoencoder.layers
Out:
[<keras.engine.input_layer.InputLayer at 0x7f8a41da3940>,
 <keras.layers.core.Dense at 0x7f8a41da3978>,
 <keras.layers.core.Dense at 0x7f8a41da3a90>,
 <keras.layers.core.Dense at 0x7f8a41da3b70>,
 <keras.layers.core.Dense at 0x7f8a44720cf8>,
 <keras.layers.core.Dense at 0x7f8a41ee4828>,
 <keras.layers.core.Dense at 0x7f8a41eaceb8>,
 <keras.layers.core.Dense at 0x7f8a41e50ac8>,
 <keras.layers.core.Dense at 0x7f8a41e5d780>]

-层是相同的。

因此，A 和 B 的训练顺序是等效的。更一般地说，如果共享层(以及权重)，则在大多数情况下构建、编译和训练的顺序并不重要，因为它们位于同一个 tensorflow 图中。

我在 mnist 数据集上运行这个示例，它们显示出相同的性能并很好地重建图像。我想，如果您遇到了 case A 的麻烦，那么您已经错过了其他事情(不知道是怎么回事，因为我复制粘贴了您的代码，一切都正常)。

如果您使用 jupyter，有时重新启动并从上到下运行会有所帮助。