fluid.layers.reshape BUG

[BigFishMaster]

I use reshape operator to reshape my input layer. But meet the following bug:

Traceback (most recent call last):
  File "se_resnext_metric.py", line 339, in <module>
    layers=layers)
  File "se_resnext_metric.py", line 204, in train
    cost = triplet_loss(input=out, margin = 0.1)
  File "se_resnext_metric.py", line 117, in triplet_loss
    output = fluid.layers.reshape(input, shape = [-1, 3, 1000])
  File "/home/ssd1/yancanxiang/code_paddle/code_metric/python-gcc482-paddle/lib/python2.7/site-packages/paddle/fluid/layers/nn.py", line 3415, in reshape
    outputs={"Out": reshaped})
  File "/home/ssd1/yancanxiang/code_paddle/code_metric/python-gcc482-paddle/lib/python2.7/site-packages/paddle/fluid/layer_helper.py", line 45, in append_op
    return self.main_program.current_block().append_op(*args, **kwargs)
  File "/home/ssd1/yancanxiang/code_paddle/code_metric/python-gcc482-paddle/lib/python2.7/site-packages/paddle/fluid/framework.py", line 816, in append_op
    op = Operator(block=self, desc=op_desc, *args, **kwargs)
  File "/home/ssd1/yancanxiang/code_paddle/code_metric/python-gcc482-paddle/lib/python2.7/site-packages/paddle/fluid/framework.py", line 494, in __init__
    self.desc.infer_shape(self.block.desc)
paddle.fluid.core.EnforceNotMet: enforce output_shape[unk_dim_idx] * capacity == -in_size failed, 0 != 1000
Invalid shape is given. at [/paddle/paddle/fluid/operators/reshape_op.h:94]