我在Keras中尝试构建一个基本的变分自编码器，以下是我的尝试：

import numpy as npfrom keras import backend as Kimport matplotlib.pyplot as pltfrom keras.models import Modelfrom keras.layers import Input,Dense,Lambdafrom keras.datasets import mnistfrom keras.objectives import binary_crossentropy(xTrain,yTrain),(xTest,yTest)=mnist.load_data()xTrain=xTrain.astype('float32')/255.xTest=xTest.astype('float32')/255.xTrain=xTrain.reshape(xTrain.shape[0],-1)xTest=xTest.reshape(xTest.shape[0],-1)xTrain.shape,yTrain.shape,xTest.shape,yTest.shapehidden_dim1=256hidden_dim2=128z_dim=2n_epoch=100n_batch=512xTrain.shape[1:]def sampleZ(args):  mu,log_var=args  eps=K.random_normal(shape=(z_dim,),mean=0.0,stddev=1.0)  return mu+K.exp(log_var/2)*epsdef vae_loss(yPred,yTrue):  recon_loss=K.sum(K.binary_crossentropy(yPred,yTrue),axis=1)  kl_loss=0.5*K.sum(K.square(mu)+K.exp(log_var)-1-log_var,axis=1)  return recon_loss+kl_lossencoderInput=Input(shape=(xTrain.shape[1:]))encoder=Dense(hidden_dim1,activation='relu')(encoderInput)encoder=Dense(hidden_dim2,activation='relu')(encoder)mu=Dense(z_dim,activation='linear')(encoder)log_var=Dense(z_dim,activation='linear')(encoder)encoderModel=Model(encoderInput,encoder)zLayer=Lambda(sampleZ,output_shape=(z_dim,)([mu,log_var])decoder=Dense(hidden_dim2,activation='relu')(zLayer)decoder=Dense(hidden_dim1,activation='relu')(decoder)decoder=Dense(xTrain.shape[1],activation='sigmoid')(decoder)vaeModel=Model(encoderInput,decoder)vaeModel.compile(optimizer='adam',loss=vae_loss)vaeModel.summary()vaeModel.fit(xTrain,xTrain,epochs=n_epoch,batch_size=n_batch)

这是模型的摘要：

Model: "functional_58"__________________________________________________________________________________________________Layer (type)                    Output Shape         Param #     Connected to                     ==================================================================================================input_12 (InputLayer)           [(None, 784)]        0                                            __________________________________________________________________________________________________dense_90 (Dense)                (None, 256)          200960      input_12[0][0]                   __________________________________________________________________________________________________dense_91 (Dense)                (None, 128)          32896       dense_90[0][0]                   __________________________________________________________________________________________________dense_92 (Dense)                (None, 2)            258         dense_91[0][0]                   __________________________________________________________________________________________________dense_93 (Dense)                (None, 2)            258         dense_91[0][0]                   __________________________________________________________________________________________________lambda_20 (Lambda)              (None, 2)            0           dense_92[0][0]                                                                                    dense_93[0][0]                   __________________________________________________________________________________________________dense_103 (Dense)               (None, 128)          384         lambda_20[0][0]                  __________________________________________________________________________________________________dense_104 (Dense)               (None, 256)          33024       dense_103[0][0]                  __________________________________________________________________________________________________dense_105 (Dense)               (None, 784)          201488      dense_104[0][0]                  ==================================================================================================Total params: 469,268Trainable params: 469,268Non-trainable params: 0

当开始训练时，我遇到了以下错误：

Epoch 1/100---------------------------------------------------------------------------TypeError                                 Traceback (most recent call last)/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)     59     tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,---> 60                                         inputs, attrs, num_outputs)     61   except core._NotOkStatusException as e:TypeError: An op outside of the function building code is being passeda "Graph" tensor. It is possible to have Graph tensorsleak out of the function building context by including atf.init_scope in your function building code.For example, the following function will fail:  @tf.function  def has_init_scope():    my_constant = tf.constant(1.)    with tf.init_scope():      added = my_constant * 2The graph tensor has name: dense_92/BiasAdd:0During handling of the above exception, another exception occurred:_SymbolicException                        Traceback (most recent call last)9 frames/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)     72       raise core._SymbolicException(     73           "Inputs to eager execution function cannot be Keras symbolic "---> 74           "tensors, but found {}".format(keras_symbolic_tensors))     75     raise e     76   # pylint: enable=protected-access_SymbolicException: Inputs to eager execution function cannot be Keras symbolic tensors, but found [<tf.Tensor 'dense_92/BiasAdd:0' shape=(None, 2) dtype=float32>, <tf.Tensor 'dense_93/BiasAdd:0' shape=(None, 2) dtype=float32>]

有谁能指导我如何解决这个问题？

回答：

错误是由于损失函数访问了模型生成的张量，即mu和log_var。你需要创建一个自定义训练循环（例如使用tf.GradientTape），或者使用模型的add_loss方法，如下所示：

vaeModel=Model(encoderInput,decoder)vaeModel.add_loss(vae_loss(encoderInput, decoder))vaeModel.compile(optimizer='adam') # 无需向compile方法传递任何损失函数