Image Dataframe.zip仅作为单个输入

我试图建立一个有三个输入的暹罗网络，并移交三个图像数据集的tf.data.Dataset.zip文件，但我的模型拒绝了它，声称它只是一个输入

input_anchor = layers.Input(target_shape + (3,))
input_positive = layers.Input(target_shape + (3,))
input_negative = layers.Input(target_shape + (3,))

embedding_anchor = embedding(input_anchor)
embedding_positive = embedding(input_positive)
embedding_negative = embedding(input_negative)

model = Model(inputs = [input_anchor, input_positive, input_negative], outputs= output)

当我将输入作为三个TensorFlow图像数据集的TensorFlow Dataset.zip文件移交时，我得到一个ValueError：

Layer model expects 3 input(s), but it received 1 input tensors. Inputs received: [<tf.Tensor 'IteratorGetNext:0' shape=(None, 1, 224, 224, 3) dtype=float32>]

However, my input file has the form: <MapDataset shapes: ((224, 224, 3), (224, 224, 3), (224, 224, 3)), types: (tf.float32, tf.float32, tf.float32)>

the input layer is the following: (<KerasTensor: shape=(None, 224, 224, 3) dtype=float32 (created by layer 'input_2')>, 
          <KerasTensor: shape=(None, 224, 224, 3) dtype=float32 (created by layer 'input_3')>, 
          <KerasTensor: shape=(None, 224, 224, 3) dtype=float32 (created by layer 'input_4')>)

层模型需要3个输入，但收到1个输入张量。收到的投入：[]
但是，我的输入文件的格式为：
输入层如下所示：（
          , 
          )

有人知道它为什么失败吗