Cannot generate all 10 pruned models for cifar10

[Kenan-Jiang]

When running: python3 group_selection.py --arch=resnet110 --resume=resnet110-bottleneck-cifar10.pth.tar --dataset=cifar10 --ngroups=10 -gpu_num=10 We have this error: RuntimeError: cuda runtime error (711) : peer mapping resources exhausted at /opt/conda/conda-bld/pytorch_1573049304260/work/aten/src/THC/THCGeneral.cpp:157 It seems that the code has error when we choose (>= 10) gpus and 10 groups for any models on cifar10 dataset.