Comments (3)
DRSAD
commented on July 29, 2024
仔细看下原论文，会发现用于计算损失的label就是[q_1,q_2,...,q_{s-1},o_s,o_{s+1},....,o_s]，其中q_1,q_2,...,q_{s-1}就是旧模型对于当前训练图片的输出，o_{s+1},....,o_s就是新类别的label，对于新类别图片，o_{s+1},....,o_s会有一个为1，对于旧类别图片，o_s,o_{s+1},....,o_s全为0
from icarl.
lufengjie051523
commented on July 29, 2024
我的qq. 534843250.能否加下，非常感谢！
…---原始邮件---
发件人: "zhuoyunli"<[email protected]>
发送时间: 2019年12月31日(周二) 上午10:52
收件人: "zhuoyunli/iCaRL"<[email protected]>;
抄送: "lufengjie051523"<[email protected]>;"Author"<[email protected]>;
主题: Re: [zhuoyunli/iCaRL] 关于这里的损失 (#2)
仔细看下原论文，会发现用于计算损失的label就是[q_1,q_2,...,q_{s-1},o_s,o_{s+1},....,o_s]，其中q_1,q_2,...,q_{s-1}就是旧模型对于当前训练图片的输出，o_{s+1},....,o_s就是新类别的label，对于新类别图片，o_{s+1},....,o_s会有一个为1，对于旧类别图片，o_s,o_{s+1},....,o_s全为0
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
from icarl.
njulhy
commented on July 29, 2024
您好，阅读了您的代码，写得很好，有个疑问，这里的损失您好像只计算了蒸馏的损失，对于论文中的第一项损失似乎没有计算？
新类损失在计算交叉熵的时候关于旧类的列标签都是0，所以直接置换旧类的列计算蒸馏损失。
from icarl.
Related Issues (11)
- 请问保存旧样本的列表太大了总是报错有什么办法解决吗，或者说您用的GPU的内存是多少G的呢？ HOT 7
- 一个小问题 HOT 7
- 问题请教
- 保存模型有误
- How to use my own dataset train my model? HOT 1
- 请教一下，为啥performence这么高？ HOT 2
- Hey i see in the requirements, you have Pytorch1.3.0 linux. I run this on Windows and get the following:
- the size of test set
- 想问一下，为什么我复现您的实验结果时，正确率总是会偏低一些
- test module
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from icarl.