本文整理匯總了Python中data.create_dataset方法的典型用法代碼示例。如果您正苦於以下問題：Python data.create_dataset方法的具體用法？Python data.create_dataset怎麽用？Python data.create_dataset使用的例子？那麽, 這裏精選的方法代碼示例或許可以為您提供幫助。您也可以進一步了解該方法所在類data的用法示例。
在下文中一共展示了data.create_dataset方法的1個代碼示例，這些例子默認根據受歡迎程度排序。您可以為喜歡或者感覺有用的代碼點讚，您的評價將有助於係統推薦出更棒的Python代碼示例。
示例1: test
# 需要導入模塊: import data [as 別名]
# 或者: from data import create_dataset [as 別名]
def test(cfg, writer, logger):
torch.manual_seed(cfg.get('seed', 1337))
torch.cuda.manual_seed(cfg.get('seed', 1337))
np.random.seed(cfg.get('seed', 1337))
random.seed(cfg.get('seed', 1337))
## create dataset
default_gpu = cfg['model']['default_gpu']
device = torch.device("cuda:{}".format(default_gpu) if torch.cuda.is_available() else 'cpu')
datasets = create_dataset(cfg, writer, logger) #source_train\ target_train\ source_valid\ target_valid + _loader
model = CustomModel(cfg, writer, logger)
running_metrics_val = runningScore(cfg['data']['target']['n_class'])
source_running_metrics_val = runningScore(cfg['data']['target']['n_class'])
val_loss_meter = averageMeter()
source_val_loss_meter = averageMeter()
time_meter = averageMeter()
loss_fn = get_loss_function(cfg)
path = cfg['test']['path']
checkpoint = torch.load(path)
model.adaptive_load_nets(model.BaseNet, checkpoint['DeepLab']['model_state'])
validation(
model, logger, writer, datasets, device, running_metrics_val, val_loss_meter, loss_fn,\
source_val_loss_meter, source_running_metrics_val, iters = model.iter
)