ehion

I trained Imagenet using 32 GPUs via horovod (8V100*4) but got acc 77.1% which was much less than 78.6% reported in your paper by running: python train.py -c confs/resnet50_imagenet_b4096.yaml --aug...

BN within ASPP included by deeplabv3 as it illustrated in paper(deeplabv3) while deeplabv2 seems not.