pytorch模型转换

model = UNet(3, 1)
modelname = 'ckpt_e_50.pth'
ckpt = torch.load(opt.pretrain + modelname)
model.load_state_dict(ckpt['state_dict'], strict=False)
model.eval()
example = torch.rand(1, 3, 240, 320)
example1 = torch.rand(1, 1, 30, 40)
traced_script_module = torch.jit.trace(model, (example, example1))
traced_script_module.save("model.pt")

文章来源: blog.csdn.net，作者：网奇，版权归原作者所有，如需转载，请联系作者。

原文链接：blog.csdn.net/jacke121/article/details/102729126