e6bbf54046cf4567e88cb130300b6b78ec88cb27,test_imagenet.py,,validate,#,95

Before Change

            // measure accuracy and record loss
            prec1, prec5 = accuracy(output.data, target, topk=(1, 5))
            losses.update(loss.data[0], input.size(0))
            top1.update(prec1[0], input.size(0))
            top5.update(prec5[0], input.size(0))

            // measure elapsed time
            batch_time.update(time.time() - end)

After Change

            // measure accuracy and record loss
            prec1, prec5 = accuracy_sum(output.data, target, topk=(1, 5))

            loss *= target.shape[0]
            count = target.new_tensor([target.shape[0]],dtype=torch.long)
            if dist.is_initialized():
              dist.all_reduce(count, dist.reduce_op.SUM)
            for meter,val in (losses,loss), (top1,prec1), (top5,prec5):
              if dist.is_initialized():
                dist.all_reduce(val, dist.reduce_op.SUM)
              val /= count.item()
              meter.update(val.item(), count.item())
            
            // measure elapsed time
            batch_time.update(time.time() - end)
            end = time.time()

            if i % args.print_freq == 0:

Instances