Itsanewday

I download the papers from arxiv: https://arxiv.org/abs/1605.07146v1. The main difference is in the table 5. Thanks a lot! @szagoruyko

Actually,I reproduced the wide residual networks, but i got slightly worse results. eg. wrn28-10 with top1err 4.20% . And i do exactly the same augmentation and preprocess. What should i...

我觉得这里的causal_embed 应该对应为论文的  你再看看论文

> 是需要自己复现的 > […](#) > 2023年1月5日 上午10:05，Itsanewday ***@***.***> 写道： 作者您好，首先感谢公开代码。我想问下论文里与其他公开的论文结果做了一些比较，这些其他论文的结果是您自己复现的吗？比如SMORE，它是针对每个样本单独训练一个超分辨模型，请问这个该怎么实现？谢谢 — Reply to this email directly, view it on GitHub , or unsubscribe . You are receiving this because you...