Keras(FIT_GENERATOR)- 检查目标时出错：预期 activation_1 具有 3 个维度，但得到的数组形状为 (32, 416, 608, 3)

Question

我已经研究分割问题很多天了，在终于找到了如何正确读取数据集之后，我遇到了这个问题：

 ValueError: Error when checking target: expected activation_1(Softmax) to have 3 dimensions, but got array with shape

(32, 416, 608, 3)

I used the functional API, since I took the FCNN architecture from [here](https://github.com/divamgupta/image-segmentation-keras/blob/master/Models/FCN32.py).

它根据我的任务（IMAGE_ORDERING =“channels_last”（TensorFlow后端））稍微修改和改编。 任何人都可以帮助我吗？ 提前致谢。 下面的架构适用于 FCNN，我尝试实现它以实现分割。 这是架构（在调用 model.summary() 之后）：

1.

2.

3. 具体错误是：

4. “导入数据集”功能：

5. “Fit_Generator 方法调用”：

    img_input = Input(shape=(input_height,input_width,3)) #Block 1 x = Convolution2D(64, (3, 3), activation='relu', padding='same', name='block1_conv1', data_format=IMAGE_ORDERING)(img_input) x = BatchNormalization()(x) x = Convolution2D(64, (3, 3), activation='relu', padding='same', name='block1_conv2', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block1_pool', data_format=IMAGE_ORDERING)(x) f1 = x # Block 2 x = Convolution2D(128, (3, 3), activation='relu', padding='same', name='block2_conv1', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = Convolution2D(128, (3, 3), activation='relu', padding='same', name='block2_conv2', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block2_pool', data_format=IMAGE_ORDERING )(x) f2 = x # Block 3 x = Convolution2D(256, (3, 3), activation='relu', padding='same', name='block3_conv1', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = Convolution2D(256, (3, 3), activation='relu', padding='same', name='block3_conv2', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = Convolution2D(256, (3, 3), activation='relu', padding='same', name='block3_conv3', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block3_pool', data_format=IMAGE_ORDERING )(x) f3 = x # Block 4 x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block4_conv1', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block4_conv2',data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block4_conv3',data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block4_pool', data_format=IMAGE_ORDERING)(x) f4 = x # Block 5 x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block5_conv1', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block5_conv2',data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block5_conv3', data_format=IMAGE_ORDERING)(x) x = BatchNormalization()(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block5_pool', data_format=IMAGE_ORDERING)(x) f5 = x x = (Convolution2D(4096,(7,7) , activation='relu' , padding='same', data_format=IMAGE_ORDERING))(x) x = Dropout(0.5)(x) x = (Convolution2D(4096,(1,1) , activation='relu' , padding='same',data_format=IMAGE_ORDERING))(x) x = Dropout(0.5)(x) #First parameter = number of classes+1 (de la background) x = (Convolution2D(20,(1,1) ,kernel_initializer='he_normal' ,data_format=IMAGE_ORDERING))(x) x = Convolution2DTranspose(20,kernel_size=(64,64), strides=(32,32),use_bias=False,data_format=IMAGE_ORDERING)(x) o_shape = Model(img_input,x).output_shape outputHeight = o_shape[1] print('Output Height is:', outputHeight) outputWidth = o_shape[2] print('Output Width is:', outputWidth) #https://keras.io/layers/core/#reshape x = (Reshape((20,outputHeight*outputWidth)))(x) #https://keras.io/layers/core/#permute x = (Permute((2, 1)))(x) print("Output shape before softmax is", o_shape) x = (Activation('softmax'))(x) print("Output shape after softmax is", o_shape) model = Model(inputs = img_input,outputs = x) model.outputWidth = outputWidth model.outputHeight = outputHeight model.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics =['accuracy'])

Answer 1

FCNN 架构示例中的原始代码适用于(416, 608)的输入维度。 而在您的代码中，输入维度是(192, 192) （忽略通道维度）。 现在如果你仔细观察，这个特殊的层

x = MaxPooling2D((2, 2), strides=(2, 2), name='block5_pool', data_format=IMAGE_ORDERING)(x)

生成维度(6, 6)的输出（您可以在model.summary()进行验证）。

下一个卷积层

o = (Convolution2D(4096,(7,7) , activation='relu' , padding='same', data_format=IMAGE_ORDERING))(o)

使用大小为(7, 7)卷积滤波器，但您的输入已经减小到小于该大小（即(6, 6) ）。 尝试先修复它。

此外，如果您查看model.summary()输出，您会注意到它不包含在block5_pool层之后定义的层。 其中有一个transposed convolution层（它基本上对您的输入进行了上采样）。 您可能想看看并尝试解决这个问题。

注意：在我所有的维度中，我都忽略了通道维度。

---

编辑下面的详细答案

首先，这是我的keras.json文件。 它使用Tensorflow后端，在channel_last设置image_ordering 。

{
    "floatx": "float32",
    "epsilon": 1e-07,
    "backend": "tensorflow",
    "image_data_format": "channels_last"
}

接下来，我复制粘贴我的确切型号代码。 请特别注意下面代码中的内联注释。

from keras.models import *
from keras.layers import *

IMAGE_ORDERING = 'channels_last' # In consistency with the json file

def getFCN32(nb_classes = 20, input_height = 416, input_width = 608):

    img_input = Input(shape=(input_height,input_width, 3)) # Expected input will have channel in the last dimension

    #Block 1
    x = Convolution2D(64, (3, 3), activation='relu', padding='same', name='block1_conv1', data_format=IMAGE_ORDERING)(img_input) 
    x = BatchNormalization()(x)
    x = Convolution2D(64, (3, 3), activation='relu', padding='same', name='block1_conv2', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = MaxPooling2D((2, 2), strides=(2, 2), name='block1_pool', data_format=IMAGE_ORDERING)(x)
    f1 = x
    # Block 2
    x = Convolution2D(128, (3, 3), activation='relu', padding='same', name='block2_conv1', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = Convolution2D(128, (3, 3), activation='relu', padding='same', name='block2_conv2', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = MaxPooling2D((2, 2), strides=(2, 2), name='block2_pool', data_format=IMAGE_ORDERING )(x)
    f2 = x

    # Block 3
    x = Convolution2D(256, (3, 3), activation='relu', padding='same', name='block3_conv1', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = Convolution2D(256, (3, 3), activation='relu', padding='same', name='block3_conv2', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = Convolution2D(256, (3, 3), activation='relu', padding='same', name='block3_conv3', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = MaxPooling2D((2, 2), strides=(2, 2), name='block3_pool', data_format=IMAGE_ORDERING )(x)
    f3 = x

    # Block 4
    x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block4_conv1', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block4_conv2',data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block4_conv3',data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = MaxPooling2D((2, 2), strides=(2, 2), name='block4_pool', data_format=IMAGE_ORDERING)(x)
    f4 = x

    # Block 5
    x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block5_conv1', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block5_conv2',data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = Convolution2D(512, (3, 3), activation='relu', padding='same', name='block5_conv3', data_format=IMAGE_ORDERING)(x)
    x = BatchNormalization()(x)
    x = MaxPooling2D((2, 2), strides=(2, 2), name='block5_pool', data_format=IMAGE_ORDERING)(x)
    f5 = x

    x = (Convolution2D(4096,(7,7) , activation='relu' , padding='same', data_format=IMAGE_ORDERING))(x)
    x = Dropout(0.5)(x)
    x = (Convolution2D(4096,(1,1) , activation='relu' , padding='same',data_format=IMAGE_ORDERING))(x)
    x = Dropout(0.5)(x)

    x = (Convolution2D(20,(1,1) ,kernel_initializer='he_normal' ,data_format=IMAGE_ORDERING))(x)
    x = Convolution2DTranspose(20,kernel_size=(64,64), strides=(32,32),use_bias=False,data_format=IMAGE_ORDERING)(x)
    o_shape = Model(img_input, x).output_shape

    # NOTE: Since this is channel last dimension ordering, the height and width dimensions are along [1] and [2], not [2] and [3]
    outputHeight = o_shape[1]
    outputWidth = o_shape[2]

    x = (Reshape((outputHeight*outputWidth, 20)))(x) # Channel should be along the last dimenion of reshape
    # No need of permute layer anymore

    print("Output shape before softmax is", o_shape)
    x = (Activation('softmax'))(x)
    print("Output shape after softmax is", o_shape)
    model = Model(inputs = img_input,outputs = x)
    model.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics =['accuracy'])

    return model

model = getFCN32(20)
print(model.summary())

接下来，我将提供我的model.summary()外观的片段。 如果你看一下最后几层，它是这样的：

因此，这意味着，该conv2d_transpose层产生尺寸的输出(448, 640, 20)并在其上施加前SOFTMAX变平出来。 所以输出的维度是(286720, 20) 。 同样，您的target_generator （在您的情况下为mask_generator ）也应该生成类似维度的目标。 同样，您的input_generator也应该生成大小为[batch size, input_height,input_width, 3]输入批次，如函数的img_input中所述。

希望这将帮助您找到问题的根源并找出合适的解决方案。 请查看代码中的细微变化（以及内嵌注释）以及如何创建输入和目标批次。

Answer 2

我再次尝试使用 SegNet 架构，但我得到了完全相同的错误。 看来这不是架构问题，而是来自 fit_generator && 来自使用掩码的问题。

更新：通过向神经网络提供正确形式的输入掩码解决了该问题。

Answer 3

您可能在flow_from_directory()调用中缺少color_mode='grayscale'掩码。 RGB 是color_mode的默认值。

flow_args = dict(
    batch_size=batch_size,
    target_size=target_size,
    class_mode=None,
    seed=seed)

image_generator = image_datagen.flow_from_directory(
    image_dir, subset='training', **flow_args)

mask_generator = mask_datagen.flow_from_directory(
    mask_dir, subset='training', color_mode='grayscale', **flow_args)