如何在 Keras/Theano 中进行反卷积？

Question

我正在尝试在 Keras 中实现反卷积。我的模型定义如下：

model=Sequential()


model.add(Convolution2D(32, 3, 3, border_mode='same',
                        input_shape=X_train.shape[1:]))
model.add(Activation('relu'))
model.add(Convolution2D(32, 3, 3, border_mode='same'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))

model.add(Convolution2D(64, 3, 3, border_mode='same'))
model.add(Activation('relu'))
model.add(Convolution2D(64, 3, 3,border_mode='same'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))

model.add(Flatten())
model.add(Dense(512))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))

我想对第一个卷积层给出的输出执行反卷积或转置卷积，即convolution2d_1。

假设我们在第一个卷积层之后拥有的特征图是 X，它属于 (9, 32, 32, 32)，其中 9 是我已经通过该层的 32x32 维度图像的编号.通过Keras的get_weights()函数得到的第一层的权重矩阵。权重矩阵的维度是(32, 3, 3, 2)。

我用于执行转置卷积的代码是

 conv_out = K.deconv2d(self.x, W, (9,3,32,32), dim_ordering = "th")
 deconv_func = K.function([self.x, K.learning_phase()], conv_out)
 X_deconv = deconv_func([X, 0 ])

但是出现错误：

 CorrMM shape inconsistency:
  bottom shape: 9 32 34 34
  weight shape: 3 32 3 3
  top shape: 9 32 32 32 (expected 9 3 32 32)

谁能告诉我哪里出错了？

Answer 1

您可以轻松使用 Deconvolution2D 图层。

这是您要实现的目标：

batch_sz = 1
output_shape = (batch_sz, ) + X_train.shape[1:]
conv_out = Deconvolution2D(3, 3, 3, output_shape, border_mode='same')(model.layers[0].output)

deconv_func = K.function([model.input, K.learning_phase()], [conv_out])

test_x = np.random.random(output_shape)
X_deconv = deconv_func([test_x, 0 ])

但最好创建一个功能模型，这将有助于训练和预测..

batch_sz = 10
output_shape = (batch_sz, ) + X_train.shape[1:]
conv_out = Deconvolution2D(3, 3, 3, output_shape, border_mode='same')(model.layers[0].output)

model2 = Model(model.input, [model.output, conv_out])
model2.summary()
model2.compile(loss=['categorical_crossentropy', 'mse'], optimizer='adam')
model2.fit(X_train, [Y_train, X_train], batch_size=batch_sz)

Answer 2

在 Keras 中，Conv2DTranspose 层执行转置卷积，也就是反卷积。它支持后端库，即 Theano 和 Keras。

Keras Documentation 说：

Conv2DTranspose

Transposed convolution layer (sometimes called Deconvolution).

The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution.

如何在 Keras/Theano 中进行反卷积？

How to perform deconvolution in Keras/ Theano?

python

theano

deep-learning

keras

keras-layer