lzx1413 / PytorchSSD

pytorch version of SSD and it's enhanced methods such as RFBSSD,FSSD and RefineDet
MIT License
709 stars 237 forks source link

每次到测试阶段都报错 #99

Open ZephyrNg opened 4 years ago

ZephyrNg commented 4 years ago
Traceback (most recent call last):
  File "refinedet_train_test.py", line 438, in <module>
    train()
  File "refinedet_train_test.py", line 264, in train
    top_k, thresh=0.01)
  File "refinedet_train_test.py", line 377, in test_net
    out = net(x=x, test=True)  # forward pass
  File "/home/wuzhefu/anaconda3/envs/ssd/lib/python3.6/site-packages/torch/nn/modules/module.py   ", line 224, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/wuzhefu/anaconda3/envs/ssd/lib/python3.6/site-packages/torch/nn/parallel/data_par   allel.py", line 59, in forward
    replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
  File "/home/wuzhefu/anaconda3/envs/ssd/lib/python3.6/site-packages/torch/nn/parallel/data_par   allel.py", line 64, in replicate
    return replicate(module, device_ids)
  File "/home/wuzhefu/anaconda3/envs/ssd/lib/python3.6/site-packages/torch/nn/parallel/replicat   e.py", line 12, in replicate
    param_copies = Broadcast(devices)(*params)
  File "/home/wuzhefu/anaconda3/envs/ssd/lib/python3.6/site-packages/torch/nn/parallel/_functio   ns.py", line 19, in forward
    outputs = comm.broadcast_coalesced(inputs, self.target_gpus)
  File "/home/wuzhefu/anaconda3/envs/ssd/lib/python3.6/site-packages/torch/cuda/comm.py", line    48, in broadcast_coalesced
    if tensor.get_device() != devices[0]:
IndexError: tuple index out of range

请问这是什么问题?