Comments (5)
michuanhaohao
commented on September 2, 2024
@Qidian213 Mutual learning has been removed from AlignedReID++. If you want to reproduce the mutual learning, you can refer to https://github.com/huanghoujing/AlignedReID-Re-Production-Pytorch
from alignedreid.
Qidian213
commented on September 2, 2024
Thanks ! I had did it by myself ,later may push a test result.
from alignedreid.
Qidian213
commented on September 2, 2024
你好，我按着https://github.com/huanghoujing/AlignedReID-Re-Production-Pytorch思路做了deep mutual learning .使用resnet50 ,inceptionv4 两个模型，DukeMTMCreID数据集。发现个现象：单独训练时都能收敛，但是mutual learning时 1，只使用分类做mutual loss 时resnet50能收敛但最后效果不如单独训练时（测试rank-1 83%），inceptionv4 刚开始收敛大概15epochs后开始发散，发散到一定程度后稳定， 最后对resetnet50测试 rank-1 82.1% 。2，只使用global distance做mutual loss 时两个模型都能收敛最后测试resnet50 rank-1 80.0%，inceptionv4 rank-1 77.1%. 3, 使用分类和global distance 做mutual loss时，另个模型都能收敛，最后resnnet50 rank-1 79.5%,inceptionv4 75.2%.
请教问题，只使用分类做mutual loss 时 inceptionv4先收敛后发散的原因可能是什么？
或者对上述现象是否还有别的建议？
谢谢！
from alignedreid.
michuanhaohao
commented on September 2, 2024
是不是inception的性能不行，mutual learning如果两个网络差距太大的话，网络会被带偏的。黄厚景的涨点结果也是在两个resent50上得到的吧。大概理解为一个大佬和一个猪队友一起组队，大佬也会被带崩。
祝好
罗浩 浙江大学
…-----原始邮件-----
发件人:Singularity <[email protected]>
发送时间:2019-03-05 22:11:21 (星期二)
收件人: michuanhaohao/AlignedReID <[email protected]>
抄送: "Hao Luo" <[email protected]>, Comment <[email protected]>
主题: Re: [michuanhaohao/AlignedReID] About mutual learning (#18)
你好，我按着https://github.com/huanghoujing/AlignedReID-Re-Production-Pytorch思路做了deep mutual learning .使用resnet50 ,inceptionv4 两个模型，DukeMTMCreID数据集。发现个现象：单独训练时都能收敛，但是mutual learning时 1，只使用分类做mutual loss 时resnet50能收敛但最后效果不如单独训练时（测试rank-1 83%），inceptionv4 刚开始收敛大概15epochs后开始发散，发散到一定程度后稳定， 最后对resetnet50测试 rank-1 82.1% 。2，只使用global distance做mutual loss 时两个模型都能收敛最后测试resnet50 rank-1 80.0%，inceptionv4 rank-1 77.1%. 3, 使用分类和global distance 做mutual loss时，另个模型都能收敛，最后resnnet50 rank-1 79.5%,inceptionv4 75.2%.
请教问题，只使用分类做mutual loss 时 inceptionv4先收敛后发散的原因可能是什么？
或者对上述现象是否还有别的建议？
谢谢！
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub, or mute the thread.
from alignedreid.
Qidian213
commented on September 2, 2024
谢谢，我再研究一下。
from alignedreid.
Related Issues (20)
- assert num_valid_q > 0, "Error: all query identities do not appear in gallery" HOT 2
- Threshold to say that it's the same person HOT 5
- what globalpooling did you use HOT 2
- how to use the test data to draw a missrate picture? HOT 1
- how to use center loss to train?
- when i train use market1501 i have face some problems HOT 1
- max identities, max images per identity required for training and min identities, min images per identity required for training
- multi-gpu training
- Training on custom dataset tutorials
- Some wrong with the code? HOT 2
- Questions about "mutual learning" in the process of reproducing the paper
- UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe3 in position 0: HOT 3
- How to calculate the aligned distance and original distance from the distance map?
- poor metrics when training on market1501 HOT 2
- Is there any performance (speed) results that I can refer to?
- Reranking Consume way too much memory space especially for MSMT17 dataset
- about run demo issue HOT 4
- The actual performance is far from the report HOT 3
- If the person in the detection bbox is incomplete, can the features be aligned?
- Why is local feature (lf) different in training phase and test phase?
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 alignedreid.