missbook520
missbook520
I encountered the same problem, did you solve it？
> > > > 我直接运行昨天刚下载的您的仓库的代码就 可以跑通， > > > > > > > > > 你说的跑的通是SwinTransformer可以跑通吗，我总是在第一轮训练计算map时出现错误 > > > > > > 我下载作者仓库的代码就可以跑通 > > 你用的是自己的数据集吗？ 我用的是自己的数据集，也是这个问题，请问您解决了吗
> model.module.module_list[Drop_layer[i]].reset(block_size[i], keep_p[i]) > torch.save(model.module.state_dict..) > all meet this problem,I don't know how to solve this. may be you can replace it with model.state_dict..
> model.module.module_list[Drop_layer[i]].reset(block_size[i], keep_p[i]) > torch.save(model.module.state_dict..) > all meet this problem,I don't know how to solve this. are you meet the error? RuntimeError: Function 'MulBackward0' returned nan values in its 1th...
Thank you very much for your reply. I followed your steps when compiling this project on jeston nano, but the following error was reported when using unet -s. What is...
请问如果使用您默认使用bilinear的代码，然后在python里面使用bilinear的话factor为2，也就是1024/2=512，这个代码是否可以跑通呢？
> i want to know the version of opencv 3.4.0