【发布时间】:2021-05-09 11:14:06
【问题描述】:
我正在尝试创建变分自动编码器,这意味着我需要自定义损失函数。问题是在损失函数内部我有 2 种不同的损失——mse 和发散。 mse 是 Tensor,divergence 是 KerasTensor(因为色散和 mu,我从编码器中出来)。我得到这样的错误:
TypeError:无法将符号 Keras 输入/输出转换为 numpy 大批。此错误可能表明您正在尝试传递一个符号 不支持 NumPy 调用的值。或者,您可能正在尝试 将 Keras 符号输入/输出传递给不支持的 TF API 注册调度,防止 Keras 自动转换 函数模型中对 lambda 层的 API 调用。
这是我的架构:
import tensorflow.keras as keras
from keras.layers import Input, Dense, Flatten, Reshape
from keras.layers import Conv2D, MaxPooling2D, UpSampling2D, Conv2DTranspose
from keras.models import Model
import tensorflow as tf
import keras.backend as K
encoded_dim = 2
class Sampling(keras.layers.Layer):
"""Uses (z_mean, z_log_var) to sample z, the vector encoding a digit."""
def call(self, inputs):
z_mean, z_log_var = inputs
batch = tf.shape(z_mean)[0]
dim = tf.shape(z_mean)[1]
epsilon = K.random_normal(shape=(batch, dim))
return z_mean + tf.exp(0.5 * z_log_var) * epsilon
img = Input((28,28,1), name='img')
x = Conv2D(32, (3,3), padding='same', activation='relu')(img)
x = MaxPooling2D()(x)
x = Conv2D(64, (3,3), padding='same', activation='relu')(x)
x = MaxPooling2D()(x)
x = Flatten()(x)
x = Dense(16, activation='relu')(x)
mu = Dense(encoded_dim, name='mu')(x)
sigma = Dense(encoded_dim, name='sigma')(x)
z = Sampling()([mu,sigma])
# print(mu)
xx = Input((encoded_dim,))
x = Dense(7*7*64, activation='relu')(xx)
x = Reshape((7,7,64))(x)
x = Conv2DTranspose(64, 3, activation="relu", strides=2, padding="same")(x)
x = Conv2DTranspose(32, 3, activation="relu", strides=2, padding="same")(x)
out = Conv2DTranspose(1, 3, activation="sigmoid", padding="same")(x)
encoder = Model(img,z, name='encoder')
decoder = Model(xx,out,name='decoder')
autoencoder = Model(img,decoder(encoder(img)),name='autoencoder')
还有损失函数:
def vae_loss(x, y):
loss = tf.reduce_mean(K.square(x-y))
kl_loss = -0.5 * tf.reduce_mean(1 + sigma - tf.square(mu) - tf.exp(sigma))
print(type(loss))
print(type(kl_loss))
return loss + kl_loss
autoencoder.compile(optimizer='adam',
loss = vae_loss)
autoencoder.fit(train,train,
epochs=1,
batch_size=60,
shuffle=True,
verbose = 2)
损失类型和lk_loss:
类'tensorflow.python.framework.ops.Tensor'
类'tensorflow.python.keras.engine.keras_tensor.KerasTensor'
【问题讨论】:
标签: python tensorflow keras deep-learning autoencoder