我尝试按照深度学习专业课程构建神经网络，但是使用了完全不同的损失函数和图片。现在，我收到一条奇怪的错误消息：
---------------------------------------------------------------------------
InvalidArgumentError Traceback (most recent call last)
/opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/ops.py in _create_c_op(graph, node_def, inputs, control_inputs)
1625 try:
-> 1626 c_op = c_api.TF_FinishOperation(op_desc)
1627 except errors.InvalidArgumentError as e:
InvalidArgumentError: Shape must be rank 0 but is rank 1 for 'Adam/update_W1/ApplyAdam' (op: 'ApplyAdam') with input shapes: [4,4,3,8], [4,4,3,8], [4,4,3,8], [], [], [9322], [], [], [], [4,4,3,8].
您是否知道那里发生了什么？ You can see the whole code here
谢谢！