xmy21sunshine的博客

from thop import profile device = 'cuda:1' model = Transformer().to(device) input = torch.randn(1, 201, 192).to(device) macs, params = profile(model, inputs=(input,)) print("macs\params:",macs,params)

34 # model = BiLSTM_Attention(128, 64, 2, 40, True) 35 model = Transformer() 36 model = nn.DataParallel(model, device_ids=[0, 1, 2, 3]).cuda() # 增加改行代码后即可运行多GPU环境 37 optimizer = torch.optim.Adam(model.parameters(), lr=opt.init_lr) ...

【深度学习】Windows安装Pycocotools（Microsoft Visual C++ 14.0 or greater is required.报错提示解决 https://www.cnblogs.com/Flat-White/p/15123308.html

arima视频：https://www.bilibili.com/video/BV1Ks411V7W4?p=2 arima视频对应的资料链接：https://blog.csdn.net/weixin_41988628/article/details/83149849 ARIMA模型实例https://blog.csdn.net/sunnyxidian/article/details/929465...