形状为 (32, 3) 的目标数组被传递为形状 (None, 15, 15, 3) 的输出，同时用作损失`categorical_crossentropy`

Question

import tensorflow as tf
from keras.preprocessing.image import ImageDataGenerator

train_datagen = ImageDataGenerator(
        rescale=1./255,
        shear_range=0.2,
        zoom_range=0.2,
        horizontal_flip=True)
training_set = train_datagen.flow_from_directory(
        'animals/training_set',
        target_size=(64, 64),
        batch_size=32,
        class_mode='categorical')

test_datagen = ImageDataGenerator(rescale=1./255)
test_set = test_datagen.flow_from_directory(
        'animals/test_set',
        target_size=(64, 64),
        batch_size=32,
        class_mode='categorical')

cnn = tf.keras.models.Sequential()

cnn.add(tf.keras.layers.Conv2D(filters = 32, kernel_size = 2, activation = 'relu', input_shape = [64, 
64, 3]))

cnn.add(tf.keras.layers.MaxPool2D(pool_size = 2, strides = 2))

cnn.add(tf.keras.layers.Conv2D(filters = 32, kernel_size = 2, activation = 'relu', input_shape = [64, 
64, 3]))
cnn.add(tf.keras.layers.MaxPool2D(pool_size = 2, strides = 2))

cnn.add(tf.keras.layers.Dense(units = 128, activation = 'relu'))

cnn.add(tf.keras.layers.Dense(units = 3, activation = 'softmax'))

cnn.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics = ['accuracy'])

cnn.fit(x = training_set, validatian_data = test_set, epochs = 15)

弹出如下错误：

ValueError: 形状为 (32, 3) 的目标数组被传递为形状 (None, 15, 15, 3) 的输出，同时用作损失 categorical_crossentropy。这种损失期望目标具有与输出相同的形状。

Answer 1

您必须在最后一个 Maxpool2D 之后添加一个 tf.keras.layers.Flatten 层，以便在一维数据上使用 Dense 层。否则，密集层适用于导致不匹配的 2D 数据。

Answer 2

在第二个 cnn.add(tf.keras.layers.Conv2D()) 函数中，您不能传递输入形状。输入形状仅传递给第一层。