u010879745的博客

课程页fast.ai releases new deep learning course, four libraries, and 600-page bookhttps://www.fast.ai/2020/08/21/fastai2-launch/fastai applications - quick starthttps://docs.fast.ai/quick_start.html教程页https://docs.fast.ai/tutorial.vision程序页https://col

Pytorch to(device)用法，如下所示：device = torch.device(“cuda:0” if torch.cuda.is_available() else “cpu”)model.to(device)非常重要的提示：这两行代码放在读取数据之前。mytensor = my_tensor.to(device)这行代码的意思是将所有最开始读取数据时的tensor变量copy一份到device所指定的GPU上去，之后的运算都在GPU上进行。https://www.jb51.n

pytorch数据加载、模型保存及加载https://blog.csdn.net/FPGATOM/article/details/85337469https://www.jb51.net/article/167892.htmtorch.train() torch.eval()https://blog.csdn.net/weixin_43593330/article/details/103365671以下是三种方式，作者亲自试过# torch.save(net.state_dict(), 'net