paddle2.1目前还不支持多维的tensor indexing吗？比如下面这个例子

import numpy as np
import paddle

t = paddle.to_tensor(np.arange(0, 12).reshape(3, 4))
i = paddle.to_tensor([0, 0, 1, 2])
j = paddle.to_tensor([0, 0, 3, 3])

print('t[i]:')
print(t[i])
print('t[i, j]:')
print(t[i, j])

报错信息如下：

t[i]:
W0723 14:43:26.482378  6014 device_context.cc:404] Please NOTE: device: 0, GPU Compute Capability: 6.1, Driver API Version: 10.2, Runtime API Version: 10.2
W0723 14:43:26.487318  6014 device_context.cc:422] device: 0, cuDNN Version: 7.6.
Tensor(shape=[4, 4], dtype=int64, place=CUDAPlace(0), stop_gradient=True,
       [[0 , 1 , 2 , 3 ],
        [0 , 1 , 2 , 3 ],
        [4 , 5 , 6 , 7 ],
        [8 , 9 , 10, 11]])
t[i, j]:
Traceback (most recent call last):
  File "debug_index.py", line 11, in <module>
    print(t[i, j])
  File "/home/bio/tools/paddle2.0/lib/python3.7/site-packages/paddle/fluid/dygraph/varbase_patch_methods.py", line 562, in __getitem__
    return _getitem_impl_(self, item)
  File "/home/bio/tools/paddle2.0/lib/python3.7/site-packages/paddle/fluid/variable_index.py", line 179, in _getitem_impl_
    format(len(item)))
IndexError: When index contains a Tensor, its length must be 1, but received 2