How to run test_wider_ms.py with multiple GPUs

[nankaielsasong]

Hi~

I ran the test_wider_ms.py with one GPU but it's too slow. Even if I set os.environ["CUDA_VISIBLE_DEVICES"] = "0, 1, 2", there is still only one working GPU, the memory of the other two GPUs are used, but the Volatile GPU-Util keeps 0% when I checked with command nvidia-smi. So can I run it with multiple GPUs to accelerate testing on the WiderFace dataset?

Thanks~