我正在尝试准备一个模型,该模型采用56x56像素和3个通道的输入图像:(56,56,3) . 输出应该是216个数字的数组 . 我重用了数字识别器中的代码并对其进行了一些修改:
model = Sequential()
model.add(Conv2D(filters = 32, kernel_size = (5,5),padding = 'Same',
activation ='relu', input_shape = (56,56,3)))
model.add(Conv2D(filters = 32, kernel_size = (5,5),padding = 'Same',
activation ='relu'))
model.add(MaxPool2D(pool_size=(2,2)))
model.add(Dropout(0.25))
model.add(Conv2D(filters = 64, kernel_size = (3,3),padding = 'Same',
activation ='relu'))
model.add(Conv2D(filters = 64, kernel_size = (3,3),padding = 'Same',
activation ='relu'))
model.add(MaxPool2D(pool_size=(2,2), strides=(2,2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(256, activation = "relu"))
model.add(Dropout(0.5))
model.add(Dense(216, activation = "linear"))
from tensorflow.python.keras.losses import categorical_crossentropy
model.compile(loss = categorical_crossentropy,
optimizer = "adam",
metrics = ['accuracy'])
这给了我一个错误:
ValueError: Error when checking target: expected dense_1 to have shape (216,) but got array with shape (72,)
我知道如何编码分类器模型,但不知道如何获取数组作为输出,所以可能我没有在最后的Dense层设置正确的形状 . 我不知道它应该是1还是216 .
我在this post中读到问题可能是损失函数,但我不确定应该使用哪种其他损失函数 .
提前致谢!
1 回答
最终图层应与目标类具有相同的形状
更改
至