wayne8088的博客

https://blog.csdn.net/winter2121/article/details/108799776

resnet = resnet50(pretrained=False)resnet.load_state_dict(torch.load(args.predir)) res_conv31 = Bottleneck_dilated(1024, 256,dilated_rate = 2)print("---------------------",res_conv31)print("---------------------",resnet.layer3[1]) res_conv31.load_

# 先安装!pip install tensorboardX%load_ext tensorboard%tensorboard --logdir logs/sample_data

参照网址，https://www.pythonheidong.com/blog/article/147611/

pip install torchsummaryimport torchimport torchvision as tvfrom torch.utils.data import DataLoaderimport torchvision.transforms as transforms # 必须要有，把PIL.Image.Image变成tensor才能计算from torchsummary import summaryvgg = tv.models.vgg16(pretrained=True

# x,y为训练输入，和标准输出 train_dataset = torch.utils.data.TensorDataset(x, y)train_dataLoader = Data.DataLoader(train_dataset, batch_size=64, shuffle=True, num_workers=0)for step, (images, labels) in enume...

加载模型的时候加上：map_locationmodel = torch.load(‘xxx.pkl’, map_location=lambda storage, loc: storage)model = model.module

版本不匹配的问题。完美解决方案如下：目前torchvision最新版本是0.5，torch最新版本是1.5，使用命令。pip install --no-deps torchvision==0.5.0