时间戳

首页 > 解决方案 > 我收到错误此错误：“张量”对象没有属性“log_prob”

python - 我收到错误此错误：“张量”对象没有属性“log_prob”

问题描述

我收到此错误

AttributeError：“张量”对象没有属性“log_prob”

同时使用 TensorFlow Probability 实现变分自动编码器。

def dense_layers(sizes):
    return tfk.Sequential([tfkl.Dense(size, activation=tf.nn.leaky_relu) for size in sizes])
encoder = tfk.Sequential([
    tfkl.InputLayer(input_shape=input_shape, name='encoder_input'),
    dense_layers(intermediary_dims),
    tfkl.Dense(latent_dim, activation = tf.nn.leaky_relu),
    tfkl.Dense(tfpl.MultivariateNormalTriL.params_size(latent_dim), activation=None),
    tfpl.MultivariateNormalTriL(latent_dim,activity_regularizer=tfpl.KLDivergenceRegularizer(prior)),
], name='encoder')

encoder.summary()
plot_model(encoder, to_file='vae_mlp_encoder.png', show_shapes=True)

decoder = tfk.Sequential([
    tfkl.InputLayer(input_shape=[latent_dim]),
    dense_layers(reversed(intermediary_dims)),
    tfkl.Dense(tfpl.IndependentNormal.params_size(original_dim), activation=None),
    tfpl.IndependentNormal(original_dim),
], name='decoder')

decoder.summary()
plot_model(decoder, to_file='vae_mlp_decoder.png', show_shapes=True)

vae = tfk.Model(inputs=encoder.inputs,
                outputs=decoder(encoder.outputs[0]),
                name='vae_mlp') # fitting the model 

negloglik = lambda x, rv_x: -rv_x.log_prob(x)

vae.compile(optimizer=tf.keras.optimizers.Nadam(), 
            loss=negloglik)

vae.summary() # summary
plot_model(vae,
           to_file='vae_mlp.png',
           show_shapes=True) #plotting the model 

#I was expecting the Vae.summary

标签： pythontensorflow

解决方案

我找到了解决方案。我安装了最新版本的 Tensorflow Probability：

! pip install --upgrade tfp-nightly
! pip install tf_nightly
! pip install tf_estimator_nightly

推荐阅读