负维度和模块keras.backend没有属性常见错误答案

【问题标题】：Negative dimension and module keras.backend has no attribute common errors负维度和模块keras.backend没有属性常见错误
【发布时间】：2021-01-16 16:17:52
【问题描述】：

我在使用 Keras 训练 MNIST 模型（CNN）时出现以下错误。顺便说一句，机器学习的完整初学者，如果有明显的问题，请道歉。任何帮助表示赞赏

from keras.layers import Flatten
from keras.layers.convolutional import Convolution2D, MaxPooling2D
from keras import backend as K
K.common.set_image_dim_ordering('th')

导致错误模块 'keras.backend' 没有属性 'common'。当用 image_data_format 替换时，它仍然没有修复任何错误

# fix random seed for reproducibility
seed = 3
np.random.seed(seed)

# load data
(X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(X_train.shape[0], 1, 28, 28).astype('float32')
X_test = X_test.reshape(X_test.shape[0], 1, 28, 28).astype('float32')

print (X_train.shape, y_train.shape)
print (X_test.shape, y_test.shape)

# normalize inputs
X_train = X_train / 255
X_test = X_test / 255

# one hot encoding
y_train = np_utils.to_categorical(y_train)
y_test = np_utils.to_categorical(y_test)
num_op = y_test.shape[1]

print (num_op)

def cnn_model():
    # create model
    model = Sequential()
    model.add(Convolution2D(32, 5, 5, padding='valid', input_shape=(1, 28, 28), activation='relu'))
    model.add(MaxPooling2D(pool_size=(2, 2)))
    model.add(Dropout(0.2))
    model.add(Flatten())
    model.add(Dense(128, activation='relu'))
    model.add(Dense(num_op, activation='softmax'))
    # Compile Model
    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
    return model

# Build the model
model = cnn_model()

# Fit the model
model.fit(X_train, y_train, validation_data=(X_test, y_test), epochs=10, batch_size=200, verbose=2)

这会导致错误 ValueError: Negative dimension size 由于从 1 中减去 5 导致 '{{node conv2d_9/Conv2D}} = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1 , 1], explicit_paddings=[], padding="VALID", strides=[1, 5, 5, 1], use_cudnn_on_gpu=true](conv2d_9_input, conv2d_9/Conv2D/ReadVariableOp)' 输入形状：[?,1, 28,28]，[5,5,28,32]。

每次我尝试修复某些东西时，我都会忽略什么总是出错，我可能会错过一些明显的东西，或者在尝试修复它时可能犯了一两个错误

【问题讨论】：

标签： python tensorflow machine-learning keras deep-learning

【解决方案1】：

你必须把通道放在之后，而不是放在高度和宽度之前。

X_train = X_train.reshape(X_train.shape[0], 28, 28, 1).astype('float32')

更好：

X_train = np.expand_dims(X_train, axis=-1)

【讨论】：