Yan Roo
@payal6792 Which version of mmdetection are you using?
@Vericoware Hello, I think protocol B in object detection & instance segmentation didn't use Dropblock. Therefore, this repo only implement protocol B (w/o Dropblock, [0.5, 2.0] scale, 350 epochs) 
@vishaldhull09 Have you tried to use single GPU training?
https://github.com/open-mmlab/mmdetection/issues/3423 https://github.com/open-mmlab/mmdetection/issues/1293 https://github.com/open-mmlab/mmdetection/issues/878 Maybe you can check your GPU usage by nvidia-smi? Is there any process working in the same time?
@farshid-mi https://github.com/yan-roo/SpineNet-Pytorch
Which config version are you using when inference? The config file "spinenet_49_B" got error when your size_divisor = 64 (Line 81) **Try to set size_divisor from 64 to 128 **...
@jinfagang I tried to set different test input resolution. However, when the divisor lower than 128 the _RuntimeError_ will occur as you mentioned above.