我正在創(chuàng)建一個(gè)分割模型，我正在添加轉(zhuǎn)換輸出，如下所示。 conv38_1 = Conv2D(filters=NUM_CLASSES, kernel_size=3, dilation_rate=(6,6), padding='same', name='fc1_voc12_c0')(relu38_1)    conv38_2 = Conv2D(filters=NUM_CLASSES, kernel_size=3, dilation_rate=(12,12), padding='same', name='fc1_voc12_c1')(relu38_1)    conv38_3 = Conv2D(filters=NUM_CLASSES, kernel_size=3, dilation_rate=(18,18), padding='same', name='fc1_voc12_c2')(relu38_1)    conv38_4 = Conv2D(filters=NUM_CLASSES, kernel_size=3, dilation_rate=(24,24), padding='same', name='fc1_voc12_c3')(relu38_1)    output   = Add(name='fc1_voc12')([conv38_1, conv38_2, conv38_3, conv38_4])    output   = Lambda(lambda image: tf.image.resize_images(image, (H,W)))(output)但是我收到一個(gè)奇怪的錯(cuò)誤如下，ValueError：檢查目標(biāo)時(shí)出錯(cuò)：預(yù)期 lambda_1 具有形狀 (321, 321, 1) 但得到形狀為 (321, 321, 2) 的數(shù)組為什么第三個(gè)形狀元素是 2？有什么辦法可以解決嗎？