I think there are some errors in file 'test_net.py'.

I think there are some errors in file 'test_net.py'.
First I run the official code, but it says:

`-- Process 1 terminated with the following error:
Traceback (most recent call last):
File "/home/dinghan/anaconda3/envs/slowfast/lib/python3.8/site-packages/torch/multiprocessing/spawn.py", line 59, in _wrap
fn(i, *args)
File "/home/dinghan/SlowFast/slowfast/utils/multiprocessing.py", line 60, in run
ret = func(cfg)
File "/home/dinghan/SlowFast/tools/test_net.py", line 243, in test
test_meter = perform_test(test_loader, model, test_meter, cfg, writer)
File "/home/dinghan/anaconda3/envs/slowfast/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "/home/dinghan/SlowFast/tools/test_net.py", line 124, in perform_test
preds, labels, video_idx = du.all_gather(
File "/home/dinghan/SlowFast/slowfast/utils/distributed.py", line 39, in all_gather
dist.all_gather(tensor_placeholder, tensor, async_op=False)
File "/home/dinghan/anaconda3/envs/slowfast/lib/python3.8/site-packages/torch/distributed/distributed_c10d.py", line 1863, in all_gather
work = default_pg.allgather([tensor_list], [tensor])
RuntimeError: Tensors must be CUDA and dense
`
I think in file 'test_net.py', line 124-133 is duplicate with line 77-90. Or maybe the indentation?