Question

我在keras中构建自动编码器时遇到了很多麻烦，这使我可以加载自定义数据集进行培训和测试。我能够找到一些应该起作用的代码，但是尝试运行它时却不断出错。这是我的代码，将在树莓派上运行：

  from keras.layers import Input, Dense
  from keras.models import Model
  import numpy as np
  from PIL import Image 
  from keras.preprocessing.image import. ImageDataGenerator
  import matplotlib.pyplot as plt
  image = Image.open('/home/pi/Downloads/neural-network-master/data/train/class_a/test(2chunk0.wav).png.jpg')
  encoding_dim = 28
  input_img = Input(shape=(65536,))
  encoded = Dense(encoding_dim, activation='relu')(input_img)
  decoded = Dense(65536, activation='sigmoid')(encoded)
  autoencoder = Model(input_img, decoded)
  encoder = Model(input_img, encoded)
  encoded_input = Input(shape=(encoding_dim,))
  decoder_layer = autoencoder.layers[-1]
  decoder = Model(encoded_input, decoder_layer(encoded_input))
  autoencoder.compile(optimizer='adadelta', loss='binary_crossentropy')  
train_datagen=ImageDataGenerator(rescale=1./255)
  train_generator = train_datagen.flow_from_directory(
     directory=r"/home/pi/Downloads/neural-network-master/data/train",
     batch_size=32,
     class_mode="categorical",
     shuffle=True,
     seed=42
  )
  autoencoder.fit_generator(train_generator,
        epochs=2,
        steps_per_epoch=256,
        shuffle=True)
  encoded_img = encoder.predict(np.array(image))
  print (encoded_img)
  decoded_img = decoder.predict(encoded_img)
  plt.imshow(decoded_img)
  plt.imshow(image)

我得到的错误是：

    ValueError: Error when checking target: expected dense_2 to have 4 dimensions, but got array with shape (262, 1)

我认为这是由于喀拉拉邦人不喜欢输入的形状引起的。任何帮助将非常感激。谢谢。另外，如果有人可以向我解释density_2的意思...

如何为密集_2修复keras ValueError

0 个答案: