f:end of record的n种原因

  1. character没有指明长度
    program main
    implicit none
    integer::it,i,j,tmax=20,ny=100,nx=100
    real::u(100,100),v(100,100),theta(100,100),A=-1/3,B=1/3
    character::filename
    do it=1,tmax
    do j=2,ny-1
    do i=2,nx-1
    u(i,j)=Au(i,j)+B((u(i-1,j)+u(i+1,j))+u(i,j-1)u(i,j+1))
    enddo
    enddo
    do j=ny-1,2,-1
    do i=nx-1,2,-1
    v(i,j)=A
    v(i,j)+B*(v(i-1,j)+v(i+1,j)+v(i,j-1)+v(i,j+1))
    enddo
    enddo
    theta=(v+u)/2
    write(filename,"(a,i2,a)") ‘theta’,it,’.txt’
    open(unit=11,file=filename)
    do i=1,nx
    do j=1,ny
    write(11,"(f8.4)") theta(i,j)
    end do
    enddo
    close(11)
    enddo
    end
  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
def train(epoch, tloaders, tasks, net, args, optimizer, list_criterion=None): print('\nEpoch: %d' % epoch) # print('...................',tasks) net.train() batch_time = AverageMeter() data_time = AverageMeter() losses = [AverageMeter() for i in tasks] top1 = [AverageMeter() for i in tasks] end = time.time() loaders = [tloaders[i] for i in tasks] min_len_loader = np.min([len(i) for i in loaders]) train_iter = [iter(i) for i in loaders] for batch_idx in range(min_len_loader*len(tasks)): config_task.first_batch = (batch_idx == 0) # Round robin process of the tasks 任务的轮循进程 current_task_index = batch_idx % len(tasks) inputs, targets = (train_iter[current_task_index]).next() config_task.task = tasks[current_task_index] # measure data loading time data_time.update(time.time() - end) if args.use_cuda: inputs, targets = inputs.cuda(), targets.cuda() optimizer.zero_grad() inputs, targets = Variable(inputs), Variable(targets) outputs = net(inputs) # net_graph = make_dot(outputs) # net_graph.render(filename='net.dot') loss = args.criterion(outputs, targets) # measure accuracy and record loss (losses[current_task_index]).update(loss.data, targets.size(0)) _, predicted = torch.max(outputs.data, 1) correct = predicted.eq(targets.data).cpu().sum() correct = correct.numpy() (top1[current_task_index]).update(correct*100./targets.size(0), targets.size(0)) # apply gradients loss.backward() optimizer.step() # measure elapsed time测量运行时间 batch_time.update(time.time() - end) end = time.time() if batch_idx % 5 == 0: print('Epoch: [{0}][{1}/{2}]\t' 'Time {batch_time.val:.3f} ({batch_time.avg:.3f})\t' 'Data {data_time.val:.3f} ({data_time.avg:.3f})\t'.format( epoch, batch_idx, min_len_loader*len(tasks), batch_time=batch_time, data_time=data_time)) for i in range(len(tasks)): print('Task {0} : Loss {loss.val:.4f} ({loss.avg:.4f})\t' 'Acc {top1.val:.3f} ({top1.avg:.3f})'.format(tasks[i], loss=losses[i], top1=top1[i])) return [top1[i].avg for i in range(len(tasks))], [losses[i].avg for i in range(len(tasks))]
07-17

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值