'Model'对象没有属性'_output_tensor_cache'

Question

我尝试从教程运行此代码：

from keras.layers import Input, Dense 
from keras.layers import BatchNormalization, Dropout, Flatten, Reshape, Lambda
from keras.models import Model
from keras.objectives import binary_crossentropy
from keras.layers.advanced_activations import LeakyReLU
from keras import backend as K
def create_vae():
    models = {}
    # Добавим Dropout и BatchNormalization
    def apply_bn_and_dropout(x):
        return Dropout(dropout_rate)(BatchNormalization()(x))
    # Энкодер
    input_img = Input(batch_shape=(batch_size, 28, 28, 1))
    x = Flatten()(input_img)
    x = Dense(256, activation='relu')(x)
    x = apply_bn_and_dropout(x)
    x = Dense(128, activation='relu')(x)
    x = apply_bn_and_dropout(x)
    # Предсказываем параметры распределений
    # Вместо того, чтобы предсказывать стандартное отклонение, предсказываем логарифм вариации
    z_mean = Dense(latent_dim)(x)
    z_log_var = Dense(latent_dim)(x)
    # Сэмплирование из Q с трюком репараметризации
    def sampling(args):
        z_mean, z_log_var = args
        epsilon = K.random_normal(shape=(batch_size, latent_dim), mean=0., stddev=1.0)
        return z_mean + K.exp(z_log_var / 2) * epsilon
    l = Lambda(sampling, output_shape=(latent_dim,))([z_mean, z_log_var])
    models["encoder"]  = Model(input_img, l, 'Encoder') 
    models["z_meaner"] = Model(input_img, z_mean, 'Enc_z_mean')
    models["z_lvarer"] = Model(input_img, z_log_var, 'Enc_z_log_var')
    # Декодер
    z = Input(shape=(latent_dim, ))
    x = Dense(128)(z)
    x = LeakyReLU()(x)
    x = apply_bn_and_dropout(x)
    x = Dense(256)(x)
    x = LeakyReLU()(x)
    x = apply_bn_and_dropout(x)
    x = Dense(28*28, activation='sigmoid')(x)
    decoded = Reshape((28, 28, 1))(x)
    models["decoder"] = Model(z, decoded, name='Decoder')
    models["vae"]     = Model(input_img, models["decoder"](models["encoder"](input_img)), name="VAE")
    def vae_loss(x, decoded):
        x = K.reshape(x, shape=(batch_size, 28*28))
        decoded = K.reshape(decoded, shape=(batch_size, 28*28))
        xent_loss = 28*28*binary_crossentropy(x, decoded)
        kl_loss = -0.5 * K.sum(1 + z_log_var - K.square(z_mean) - K.exp(z_log_var), axis=-1)
        return (xent_loss + kl_loss)/2/28/28
    return models, vae_loss
models, vae_loss = create_vae()
vae = models["vae"]

但是Colab向我展示了一个错误，怎么了？

AttributeError                            Traceback (most recent call last)
<ipython-input-108-b96af0c31b2f> in <module>()
     61     return models, vae_loss
     62 
---> 63 models, vae_loss = create_vae()
     64 vae = models["vae"]
2 frames
/usr/local/lib/python3.6/dist-packages/keras/engine/network.py in call(self, inputs, mask)
    559         cache_key = object_list_uid(inputs)
    560         cache_key += '_' + object_list_uid(masks)
--> 561         if cache_key in self._output_tensor_cache:
    562             return self._output_tensor_cache[cache_key]
    563         else:
AttributeError: 'Model' object has no attribute '_output_tensor_cache'

我正在尝试研究VAE，无法找到如何解决该问题。该行生成错误：

models["vae"]     = Model(input_img, models["decoder"](models["encoder"](input_img)), name="VAE")

，但对我来说看起来完全不错。不要阅读我在下面写的内容，这个网站不允许我没有额外的文字发布问题：看来您的帖子主要是代码；请添加更多详细信息。

Answer 1

一些相关讨论请参阅那里

随后在我的情况下有帮助(python 3.7(：

pip install tensorflow==1.14.0
pip install keras==2.0