如何将预期的density_2固定为2维

Question

我收到一个错误：“检查目标时出错：预期density_2具有2个维，但是数组的形状为（32，256，455，3）”。输入数据是具有三个通道的40,000个RGB图像（256像素x 455像素）。每批中有32张图像。输出是方向盘的角度，因此不是类别。如何解决此错误？

enter code here
datagen = ImageDataGenerator(rescale=1./255,validation_split=0.2)

train_generator = datagen.flow_from_dataframe(dataframe=data, 
class_mode = 
"input",directory="../input/driving_dataset/driving_dataset/", 
x_col="files", y_col="results", batch_size=32,color_mode = 
'rgb',target_size = (256,455),subset = 'training')

validation_generator = datagen.flow_from_dataframe(dataframe=data, 
class_mode = 
"input",directory="../input/driving_dataset/driving_dataset/", 
x_col="files", y_col="results", batch_size=32,color_mode = 
'rgb',target_size = (256,455),subset = 'validation')

model = Sequential()
model.add(Conv2D(24,(5,5),padding ="valid",input_shape = (256,455,3), 
activation = 'relu'))
model.add(Conv2D(36,(5,5),padding = "valid", activation = "relu"))
model.add(Conv2D(48,(5,5),padding = "valid", activation = "relu"))
model.add(Flatten())
model.add(Dense(1164, activation = "relu"))
model.add(Dense(100, activation = "relu"))
model.add(Dense(10, activation = "relu")
model.add(Dense(1))
model.compile(optimizer = "Adam", loss="mse")

model.fit_generator(
train_generator,
steps_per_epoch = train_generator.samples // 32,
validation_data = validation_generator, 
validation_steps = validation_generator.samples // 32,
epochs = 2)

Answer 1

问题是您正在设置class_mode="input"，这意味着生成器也将输入图像用作标签。您应该设置class_mode="other"，以便将y_col中的值用作标签。