InvalidArgumentError

I am getting an error in the decoder inference part.

InvalidArgumentError Traceback (most recent call last)
in <cell line: 2>()
1 # Setup the loop variables.
----> 2 next_token, done, state = decoder.get_initial_state(ex_context)
3 tokens = []
4
5 for n in range(10):

6 frames
/usr/local/lib/python3.10/dist-packages/tensorflow/python/framework/ops.py in raise_from_not_ok_status(e, name)
5981 def raise_from_not_ok_status(e, name) -> NoReturn:
5982 e.message += (" name: " + str(name if name is not None else ""))
-> 5983 raise core._status_to_exception(e) from None # pylint: disable=protected-access
5984
5985

InvalidArgumentError: {{function_node _wrapped__Pack_N_2_device/job:localhost/replica:0/task:0/device:GPU:0}} Shapes of all inputs must match: values[0].shape = [64,1,256] != values[1].shape = [] [Op:Pack] name:

Help me with this(Important!!!)

Start by reproducing the notebook call to decoder.get_initial_state(ex_context) and inspect the shapes passed into the decoder. Compare the traceback's [64,1,256] value with the scalar input reported by TensorFlow's ops.py Pack operation; done means the inference call runs without the InvalidArgumentError.