我试图将Keras网站上的2D卷积自编码器示例调整为适合我自己的情况，我使用的是1D输入：https://blog.keras.io/building-autoencoders-in-keras.html
以下是我的代码：
from keras.layers import Input, Dense, Conv1D, MaxPooling1D, UpSampling1Dfrom keras.models import Modelfrom keras import backend as Kimport scipy as scipyimport numpy as np mat = scipy.io.loadmat('edata.mat')emat = mat['edata']input_img = Input(shape=(64,1)) # adapt this if using `channels_first` image data formatx = Conv1D(32, (9), activation='relu', padding='same')(input_img)x = MaxPooling1D((4), padding='same')(x)x = Conv1D(16, (9), activation='relu', padding='same')(x)x = MaxPooling1D((4), padding='same')(x)x = Conv1D(8, (9), activation='relu', padding='same')(x)encoded = MaxPooling1D(4, padding='same')(x)x = Conv1D(8, (9), activation='relu', padding='same')(encoded)x = UpSampling1D((4))(x)x = Conv1D(16, (9), activation='relu', padding='same')(x)x = UpSampling1D((4))(x)x = Conv1D(32, (9), activation='relu')(x)x = UpSampling1D((4))(x)decoded = Conv1D(1, (9), activation='sigmoid', padding='same')(x)autoencoder = Model(input_img, decoded)autoencoder.compile(optimizer='adadelta', loss='binary_crossentropy')x_train = emat[:,0:80000]x_train = np.reshape(x_train, (x_train.shape[1], 64, 1))x_test = emat[:,80000:120000]x_test = np.reshape(x_test, (x_test.shape[1], 64, 1))from keras.callbacks import TensorBoardautoencoder.fit(x_train, x_train, epochs=50, batch_size=128, shuffle=True, validation_data=(x_test, x_test), callbacks=[TensorBoard(log_dir='/tmp/autoencoder')])然而，当我尝试运行autoencoder.fit()时，我收到了以下错误：
ValueError: 检查目标时出错：期望conv1d_165的形状为(None, 32, 1)，但得到的数组形状为(80000, 64, 1)
我知道我设置层的时候可能做错了什么，我只是将maxpool和conv2d的大小改成了1D形式…我对Keras或自编码器的经验很少，有人能看出我哪里做错了么？
谢谢
编辑：在新的控制台上运行时，错误如下：
ValueError: 检查目标时出错：期望conv1d_7的形状为(None, 32, 1)，但得到的数组形状为(80000, 64, 1)
以下是autoencoder.summary()的输出
Layer (type) Output Shape Param # =================================================================input_1 (InputLayer) (None, 64, 1) 0 _________________________________________________________________conv1d_1 (Conv1D) (None, 64, 32) 320 _________________________________________________________________max_pooling1d_1 (MaxPooling1 (None, 16, 32) 0 _________________________________________________________________conv1d_2 (Conv1D) (None, 16, 16) 4624 _________________________________________________________________max_pooling1d_2 (MaxPooling1 (None, 4, 16) 0 _________________________________________________________________conv1d_3 (Conv1D) (None, 4, 8) 1160 _________________________________________________________________max_pooling1d_3 (MaxPooling1 (None, 1, 8) 0 _________________________________________________________________conv1d_4 (Conv1D) (None, 1, 8) 584 _________________________________________________________________up_sampling1d_1 (UpSampling1 (None, 4, 8) 0 _________________________________________________________________conv1d_5 (Conv1D) (None, 4, 16) 1168 _________________________________________________________________up_sampling1d_2 (UpSampling1 (None, 16, 16) 0 _________________________________________________________________conv1d_6 (Conv1D) (None, 8, 32) 4640 _________________________________________________________________up_sampling1d_3 (UpSampling1 (None, 32, 32) 0 _________________________________________________________________conv1d_7 (Conv1D) (None, 32, 1) 289 =================================================================Total params: 12,785Trainable params: 12,785Non-trainable params: 0_________________________________________________________________回答：
