shape mismatch

[garspace2]

when i finetune model with a single custom dataset, and print the batch, the error is:
---> {'observation': {'image_primary': <tf.Tensor 'strided_slice_17:0' shape=(None, 64, 64, 3) dtype=uint8>, 'image_wrist': <tf.Tensor 'Repeat/Reshape_1:0' shape=(None,) dtype=string>, 'proprio': <tf.Tensor 'strided_slice_21:0' shape=(None, 7) dtype=float32>, 'timestep': <tf.Tensor 'range_1:0' shape=(None,) dtype=int32>}, 'task': {'language_instruction': <tf.Tensor 'strided_slice_16:0' shape=(None,) dtype=string>}, 'action': <tf.Tensor 'concat_1:0' shape=(None, 7) dtype=float32>, 'dataset_name': <tf.Tensor 'Repeat_1/Reshape_1:0' shape=(None,) dtype=string>}
Traceback (most recent call last):
File "/data/RND/dengjie/code/robot/octo/scripts/finetune_mydataset.py", line 415, in
app.run(main)
File "/home/user8/anaconda3/lib/python3.9/site-packages/absl/app.py", line 308, in run
_run_main(main, args)
File "/home/user8/anaconda3/lib/python3.9/site-packages/absl/app.py", line 254, in _run_main
sys.exit(main(argv))
File "/data/RND/dengjie/code/robot/octo/scripts/finetune_mydataset.py", line 181, in main
example_batch = next(train_data_iter)
File "/home/user8/anaconda3/lib/python3.9/site-packages/tensorflow/python/data/ops/dataset_ops.py", line 4733, in next
return nest.map_structure(to_numpy, next(self._iterator))
File "/home/user8/anaconda3/lib/python3.9/site-packages/tensorflow/python/data/ops/iterator_ops.py", line 810, in next
return self._next_internal()
File "/home/user8/anaconda3/lib/python3.9/site-packages/tensorflow/python/data/ops/iterator_ops.py", line 773, in _next_internal
ret = gen_dataset_ops.iterator_get_next(
File "/home/user8/anaconda3/lib/python3.9/site-packages/tensorflow/python/ops/gen_dataset_ops.py", line 3029, in iterator_get_next
_ops.raise_from_not_ok_status(e, name)
File "/home/user8/anaconda3/lib/python3.9/site-packages/tensorflow/python/framework/ops.py", line 5883, in raise_from_not_ok_status
raise core._status_to_exception(e) from None # pylint: disable=protected-access
tensorflow.python.framework.errors_impl.InvalidArgumentError: {{function_node _wrapped__IteratorGetNext_output_types_23_device/job:localhost/replica:0/task:0/device:CPU:0}} condition [10], then [10,64,64,3], and else [10,64,64,3] must be broadcastable

when I use bridege_dataset and print the batch, the log is:
---> {'observation': {'image_primary': <tf.Tensor 'strided_slice_17:0' shape=(None,) dtype=string>, 'image_wrist': <tf.Tensor 'Repeat/Reshape_1:0' shape=(None,) dtype=string>, 'proprio': <tf.Tensor 'strided_slice_21:0' shape=(None, 7) dtype=float32>, 'timestep': <tf.Tensor 'range_1:0' shape=(None,) dtype=int32>}, 'task': {'language_instruction': <tf.Tensor 'strided_slice_16:0' shape=(None,) dtype=string>}, 'action': <tf.Tensor 'concat_1:0' shape=(None, 7) dtype=float32>, 'dataset_name': <tf.Tensor 'Repeat_1/Reshape_1:0' shape=(None,) dtype=string>}

how to solve it? thanks.