RuntimeError: all tensors must be on devices[0]

VDIGPKU / M2Det

M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network

MIT License

1.45k stars 318 forks source link

RuntimeError: all tensors must be on devices[0] #118

Open 2429581027 opened 4 years ago

2429581027 commented 4 years ago

When I run the code CUDA_VISIBLE_DEVICES=0,1,2,3 python train.py -c=configs/m2det512_vgg.py --ngpu 4 -t True There is a error: RuntimeError: all tensors must be on devices[0] Does anyone know how to solve it? I tried a lot of methods but they couldn't do it.eg: link Thank you very much!

RoyLeeDongHee commented 4 years ago

same error CUDA_VISIBLE_DEVICES=0,1,2,3,4 python train.py -c=configs/m2det512_vgg.py --ngpu 5 -t True

how to train on multi-gpus ?