不懂的问题

Traceback (most recent call last):
File "/home/user/Disk1.8T/unicoe/pytorch-ssd-2/experments/3_242_seg_merge_base512/train_vgg640_480_seg_merge_seg.py", line 269, in <module>
train_seg()
File "/home/user/Disk1.8T/unicoe/pytorch-ssd-2/experments/3_242_seg_merge_base512/train_vgg640_480_seg_merge_seg.py", line 226, in train_seg
loss_l, loss_c, loss_seg = criterion(out, bbox_targets, seg_targets)
File "/home/user/anaconda3/envs/pytorch_ssd/lib/python3.6/site-packages/torch/nn/modules/module.py", line 357, in __call__
result = self.forward(*input, **kwargs)
File "/home/user/Disk1.8T/unicoe/pytorch-ssd-2/layers/modules/multibox_seg_loss.py", line 155, in forward
conf_p = conf_data[(pos_idx+neg_idx).gt(0)].view(-1, self.num_classes)
File "/home/user/anaconda3/envs/pytorch_ssd/lib/python3.6/site-packages/torch/autograd/variable.py", line 74, in __getitem__
return MaskedSelect.apply(self, key)
File "/home/user/anaconda3/envs/pytorch_ssd/lib/python3.6/site-packages/torch/autograd/_functions/tensor.py", line 465, in forward
return tensor.masked_select(mask)
RuntimeError: cuda runtime error (77) : an illegal memory access was encountered at /opt/conda/conda-bld/pytorch_1518243271935/work/torch/lib/THC/generated/../THCReduceAll.cuh:339