不支持的Feed类型错误

Question

不确定feed_dict到底出了什么问题，但是我得到了这个错误：

Feed字典返回错误，并且所有占位符都正确

c:\program files\python36\lib\site-packages\tensorflow\python\client\session.py in _do_call(self, fn, *args)
   1333         except KeyError:
   1334           pass
-> 1335       raise type(e)(node_def, op, message)
   1336 
   1337   def _extend_graph(self):

InternalError: Unsupported feed type
Here are the placeholders:

Tensor("batch:0", dtype=int32)
Tensor("batch_edge_type_idx:0", shape=(), dtype=int32)
Tensor("batch_row_edge_type:0", shape=(), dtype=int32)
Tensor("batch_col_edge_type:0", shape=(), dtype=int32)
SparseTensor(indices=Tensor("Placeholder_3:0", shape=(?, ?), dtype=int64), values=Tensor("Placeholder_2:0", shape=(?,), dtype=float32), dense_shape=Tensor("Placeholder_1:0", shape=(?,), dtype=int64))
SparseTensor(indices=Tensor("Placeholder_6:0", shape=(?, ?), dtype=int64), values=Tensor("Placeholder_5:0", shape=(?,), dtype=float32), dense_shape=Tensor("Placeholder_4:0", shape=(?,), dtype=int64))
SparseTensor(indices=Tensor("Placeholder_9:0", shape=(?, ?), dtype=int64), values=Tensor("Placeholder_8:0", shape=(?,), dtype=float32), dense_shape=Tensor("Placeholder_7:0", shape=(?,), dtype=int64))
SparseTensor(indices=Tensor("Placeholder_12:0", shape=(?, ?), dtype=int64), values=Tensor("Placeholder_11:0", shape=(?,), dtype=float32), dense_shape=Tensor("Placeholder_10:0", shape=(?,), dtype=int64))
SparseTensor(indices=Tensor("Placeholder_15:0", shape=(?, ?), dtype=int64), values=Tensor("Placeholder_14:0", shape=(?,), dtype=float32), dense_shape=Tensor("Placeholder_13:0", shape=(?,), dtype=int64))
SparseTensor(indices=Tensor("Placeholder_18:0", shape=(?, ?), dtype=int64), values=Tensor("Placeholder_17:0", shape=(?,), dtype=float32), dense_shape=Tensor("Placeholder_16:0", shape=(?,), dtype=int64))
SparseTensor(indices=Tensor("Placeholder_24:0", shape=(?, ?), dtype=int64), values=Tensor("Placeholder_23:0", shape=(?,), dtype=float32), dense_shape=Tensor("Placeholder_22:0", shape=(?,), dtype=int64))
SparseTensor(indices=Tensor("Placeholder_30:0", shape=(?, ?), dtype=int64), values=Tensor("Placeholder_29:0", shape=(?,), dtype=float32), dense_shape=Tensor("Placeholder_28:0", shape=(?,), dtype=int64))
Tensor("PlaceholderWithDefault:0", shape=(), dtype=float32)