数据挖掘 - 构建顺序模型时出现 ResourceExhaustedError - 吾爱随笔录

我在尝试构建模型时遇到了一个大问题，

input shape: (1447, 224, 224, 3)
output shape: (1447, 154457)

model = Sequential([
        Conv2D(filters=64, kernel_size=(3, 3), activation='relu', padding='valid', input_shape=(224,224,3)),
        BatchNormalization(axis=-1),
        Conv2D(filters=64, kernel_size=(3, 3), activation='relu', padding='valid'),
        BatchNormalization(axis=-1),
        Conv2D(filters=128, kernel_size=(3, 3), activation='relu', padding='valid'),
        BatchNormalization(axis=-1),
        MaxPool2D(pool_size=(2, 2), strides=(2, 1)),
        Conv2D(filters=128, kernel_size=(3, 3), activation='relu', padding='valid'),
        BatchNormalization(axis=-1),
        MaxPool2D(pool_size=(2, 2), strides=(2, 1)),
        Conv2D(filters=128, kernel_size=(3, 3), activation='relu', padding='valid'),
        BatchNormalization(axis=-1),
        MaxPool2D(pool_size=(2, 2), strides=(2, 1)),
        Conv2D(filters=256, kernel_size=(3, 3), activation='relu', padding='valid'),
        BatchNormalization(axis=-1),
        MaxPool2D(pool_size=(2, 2), strides=(2, 1)),
        Conv2D(filters=512, kernel_size=(3, 3), activation='relu', padding='valid'),
        BatchNormalization(axis=-1),
        Conv2D(filters=512, kernel_size=(3, 3), activation='relu', padding='valid', strides=3),
        BatchNormalization(axis=-1),
        Conv2D(filters=512, kernel_size=(3, 3), activation='relu', padding='valid', strides=3),
        AveragePooling2D(pool_size=(6, 1), strides=1, padding='same'),
        ReLU(),
        BatchNormalization(axis=-1),
        Flatten(),
        Dense(154457, activation='relu'),
        Dense(154457)
    ])

我得到以下错误，

ResourceExhaustedError: OOM when allocating tensor with shape[11264,154457] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc [Op:RandomUniform]

这里有什么问题，因为我是神经网络的新手。