Animesh Gupta

> 多谢哈，我这边也解决了，但是我发现是我的数据增强没有用上才导致精确度上不去的 您能告诉我如何解决这个问题，因为我无法获得相同的数字吗？

> 我自己实现的时候没有用dataloader，datasets虽然有transform这个属性，但是并没有对图片进行操作，在用dataloader之后才会对生成的图片进行数据增强，直接使用dataloader产生的数据就可以了。 > […](#) > ------------------ 原始邮件 ------------------ 发件人: "Animesh Gupta"

> With three runs of resnet18, I got an average acc around 76.1. However, on resnet50, I got 76.0. > Does anyone have the same problem? By the way, resnet101...

> PyTorch 1.4.0 I m also trying to run this with PyTorch 1.0.0 and I am not able get an exact error on resnet101. Can you tell me if you...

Hi, PFA for the pdf version of the book. I hope the authors of the book have no issue with this.😓 Thank you @Aryaman1706 for helping me in converting to...

> ml-interviews-book.pdf Hi @chiphuyen. I tried to answer your follow-up question in the email. Hope the problem is resolved now. 😇

> Please pin this issue and add **FaceX-API** label. This is a good issue. @AdityaNikhil will it be okay if I contribute to this?

can anybody tell me which files I should look at in order to solve the issue?

> @animesh-007 > > Can you point out what version of GradMatch you are using? > > Ideally, subset selection should be faster unless something is wrong with the experimental...

@shiyf129 What is the resolution of the images you are using while training? I am using 224x224.