本文整理汇总了Python中utils.logger.Logger.append方法的典型用法代码示例。如果您正苦于以下问题：Python Logger.append方法的具体用法？Python Logger.append怎么用？Python Logger.append使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类utils.logger.Logger的用法示例。
在下文中一共展示了Logger.append方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: main
# 需要导入模块: from utils.logger import Logger [as 别名]
# 或者: from utils.logger.Logger import append [as 别名]
def main(args):
global best_acc
global best_auc
if not os.path.exists(args.checkpoint):
os.makedirs(args.checkpoint)
print("==> Creating model '{}-{}', stacks={}, blocks={}, feats={}".format(
args.netType, args.pointType, args.nStacks, args.nModules, args.nFeats))
print("=> Models will be saved at: {}".format(args.checkpoint))
model = models.__dict__[args.netType](
num_stacks=args.nStacks,
num_blocks=args.nModules,
num_feats=args.nFeats,
use_se=args.use_se,
use_attention=args.use_attention,
num_classes=68)
model = torch.nn.DataParallel(model).cuda()
criterion = torch.nn.MSELoss(size_average=True).cuda()
optimizer = torch.optim.RMSprop(
model.parameters(), lr=args.lr, momentum=args.momentum, weight_decay=args.weight_decay)
title = args.checkpoint.split('/')[-1] + ' on ' + args.data.split('/')[-1]
Loader = get_loader(args.data)
val_loader = torch.utils.data.DataLoader(
Loader(args, 'A'),
batch_size=args.val_batch,
shuffle=False,
num_workers=args.workers,
pin_memory=True)
if args.resume:
if os.path.isfile(args.resume):
print("=> Loading checkpoint '{}'".format(args.resume))
checkpoint = torch.load(args.resume)
args.start_epoch = checkpoint['epoch']
best_acc = checkpoint['best_acc']
model.load_state_dict(checkpoint['state_dict'])
optimizer.load_state_dict(checkpoint['optimizer'])
print("=> Loaded checkpoint '{}' (epoch {})".format(args.resume, checkpoint['epoch']))
logger = Logger(os.path.join(args.checkpoint, 'log.txt'), title=title, resume=True)
else:
print("=> no checkpoint found at '{}'".format(args.resume))
else:
logger = Logger(os.path.join(args.checkpoint, 'log.txt'), title=title)
logger.set_names(['Epoch', 'LR', 'Train Loss', 'Valid Loss', 'Train Acc', 'Val Acc', 'AUC'])
cudnn.benchmark = True
print('=> Total params: %.2fM' % (sum(p.numel() for p in model.parameters()) / (1024. * 1024)))
if args.evaluation:
print('=> Evaluation only')
D = args.data.split('/')[-1]
save_dir = os.path.join(args.checkpoint, D)
if not os.path.exists(save_dir):
os.makedirs(save_dir)
loss, acc, predictions, auc = validate(val_loader, model, criterion, args.netType,
args.debug, args.flip)
save_pred(predictions, checkpoint=save_dir)
return
train_loader = torch.utils.data.DataLoader(
Loader(args, 'train'),
batch_size=args.train_batch,
shuffle=True,
num_workers=args.workers,
pin_memory=True)
lr = args.lr
for epoch in range(args.start_epoch, args.epochs):
lr = adjust_learning_rate(optimizer, epoch, lr, args.schedule, args.gamma)
print('=> Epoch: %d | LR %.8f' % (epoch + 1, lr))
train_loss, train_acc = train(train_loader, model, criterion, optimizer, args.netType,
args.debug, args.flip)
# do not save predictions in model file
valid_loss, valid_acc, predictions, valid_auc = validate(val_loader, model, criterion, args.netType,
args.debug, args.flip)
logger.append([int(epoch + 1), lr, train_loss, valid_loss, train_acc, valid_acc, valid_auc])
is_best = valid_auc >= best_auc
best_auc = max(valid_auc, best_auc)
save_checkpoint(
{
'epoch': epoch + 1,
'netType': args.netType,
'state_dict': model.state_dict(),
'best_acc': best_auc,
'optimizer': optimizer.state_dict(),
},
is_best,
predictions,
checkpoint=args.checkpoint)
#.........这里部分代码省略.........