我正在尝试使用Keras和TensorFlow进行模型预测。假设使用model.predict（）函数时，我给它一个形状为[1,2160,8]的np.array，程序将抛出错误，说明：
ValueError: Input 0 of layer sequential_9 is incompatible with the layer: expected axis -1 of input shape to have value 1 but received input with shape [None, 2160, 8]
另外，我正在馈送的np.array的形状为[1,2160,8]，但错误表明它的形状为[None，2160，8]。
这是我的预测代码：
test = np.array([[[1,2,3,4,5,6,7,8]]*2160], np.uint32)
reconstructed_model = keras.models.load_model('saved_model/my_model_cnn')
reconstructed_model.summary()
res = reconstructed_model.predict(test)
model.summary如下：
Model: "sequential_9"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv1d_9 (Conv1D) (None, 2160, 8) 40
_________________________________________________________________
flatten_9 (Flatten) (None, 17280) 0
_________________________________________________________________
dropout_9 (Dropout) (None, 17280) 0
_________________________________________________________________
output (Dense) (None, 1) 17281
=================================================================
Total params: 17,321
Trainable params: 17,321
Non-trainable params: 0
_________________________________________________________________
有人知道如何解决此问题吗？