Error When Training Retinaface with mobilenet.25

I got this error while training retinaface using mobilenet.25 as backbone.

> Traceback (most recent call last):
File "train.py", line 503, in
main()
File "train.py", line 499, in main
lr_step=args.lr_step)
File "train.py", line 391, in train_net
num_epoch=end_epoch)
File "/usr/local/lib/python3.6/site-packages/mxnet/module/base_module.py", line 498, in fit
for_training=True, force_rebind=force_rebind)
File "/usr/local/lib/python3.6/site-packages/mxnet/module/module.py", line 429, in bind
state_names=self._state_names)
File "/usr/local/lib/python3.6/site-packages/mxnet/module/executor_group.py", line 280, in __init__
self.bind_exec(data_shapes, label_shapes, shared_group)
File "/usr/local/lib/python3.6/site-packages/mxnet/module/executor_group.py", line 383, in bind_exec
self._collect_arrays()
File "/usr/local/lib/python3.6/site-packages/mxnet/module/executor_group.py", line 312, in _collect_arrays
for name, _ in self.data_shapes]
File "/usr/local/lib/python3.6/site-packages/mxnet/module/executor_group.py", line 312, in
for name, _ in self.data_shapes]
File "/usr/local/lib/python3.6/site-packages/mxnet/module/executor_group.py", line 311, in
self.data_arrays = [[(self.slices[i], e.arg_dict[name]) for i, e in enumerate(self.execs)]
File "/usr/local/lib/python3.6/site-packages/mxnet/executor.py", line 277, in arg_dict
self._symbol.list_arguments(), self.arg_arrays)
File "/usr/local/lib/python3.6/site-packages/mxnet/executor.py", line 95, in _get_dict
raise ValueError('Duplicate names detected, %s' % str(names))

It warns me that the variable name 'face_rpn_landmark_pred_stride32_weight' is used more than once. What adjustments should I make to solve this problem?