xianglei3

I work in an AI company, and we hope that NNI can not only save JSON representing the optimal model, but also the optimal model containing weights.

2

**What would you like to be added**: I work in an AI company, and we hope that NNI can not only save JSON representing the optimal model, but also the...

I am a student of the second year of Tsinghua graduate school. I sincerely ask a question: does it support real-time verification during training? For example, do you consider adding a new function, that is, each time you complete an epoch training, you will verify on the specified verification data set and print the loss on the verification set?

2

I am a student of the second year of Tsinghua graduate school. I sincerely ask a question: does it support real-time verification during training? For example, do you consider adding...

Hello, I'm a student of Tsinghua University. I found that running unet2_ test. The values of the image generated by py file are not all 0 and 255, but also some values between 0 and 255. I want to ask why. Usually, the segmentation result is not only two values, that is, 0 or 255

2

Hello, I'm a student of Tsinghua University. I found that running unet2_ test. The values of the image generated by py file are not all 0 and 255, but also...

I'm a student of the second Graduate School of Tsinghua University. When I trained, I didn't quite understand the meaning of L1 to L5 loss printed out by the command line. Among them, do L1 and L2 loss refer to absolute value error and mean square error? In addition, I don't understand L3 to L5 loss, and I sincerely look forward to your reply

1

The printing information is as follows l0: 0.439068, l1: 0.418778, l2: 0.421991, l3: 0.425464, l4: 0.438959, l5: 0.530802, l6: 0.505693

I am a student of Tsinghua University. I want to ask whether u2net only supports single category segmentation. For example, for the picture of people riding on bicycles, can people and bicycles be separated separately? Thank you. Thank you very much

1

I am a student of Tsinghua University. I want to ask whether u2net only supports single category segmentation. For example, for the picture of people riding on bicycles, can people...

thanks

Thanks for your error report and we appreciate it a lot. **Checklist** 1. I have searched related issues but cannot get the expected help. 2. The bug has not been...

你好，我是清华的一名学生，论文提到分割部分的backbone是mobileV2吗，这部分是否可以换成其他网络，有一些细节比较困惑，真心请教，感激不尽

1

按照论文说法， modnet包含分割，细节预测和融合三个部分，我想问的是， 问题1，如果把分割部分的backbone换成其他网络，是否可行？例如换成efficintnet，是否可以无缝替换？ 问题2，分割部分和其他部分之间的交互问题，例如，其中细节预测部分用到了部分分割部分的中间层的特征信息，如果把分割的backbone换成其他网络，例如efficinet网络，那么中间层的特征交互问题该如何处理呢？ 谢谢，感激不尽

Hello, I am a student of Tsinghua University. From your paper, I know that modnet includes segmentation, detail prediction and fusion. I want to ask whether the backbone of the...

thanks , NMF