Comments (3)
Oh I got the reason. However, which layer should be written in the parameter of --confusion?
from confusion.
emmm... And where do I modify the regularization parameter λ?
from confusion.
Hi @abcdvzz, you should use the confusion typically on the last (logit) layer of the network, but it should provide moderate improvements (with very small λ of around 10e-4
to 10e-5
) for intermediate layers as well.
The formula explored in the ECCV paper can be replicated with the command:
./train.py .... --confusion '{"final_layer_name": 0.01}'
The parameters to the --confusion
argument are basically the layer name and the weight (λ), which in this case are final_layer_name
and 0.01
. The README already covers this usage.
from confusion.
Related Issues (12)
- Implementation issue for bilinear VGG HOT 1
- Logits or softmax probabilities HOT 3
- training encounter with NAN HOT 9
- [HELP] : The loss_train is always 5.3833 HOT 7
- is RPN_PRE_NMS_TOP_N per image?
- Wish HOT 1
- pytorch loss HOT 1
- Ask for help, Compile error! HOT 1
- How to reproduce the result of densenet161 HOT 6
- about train problems HOT 2
- How to avoid log zero in EntropicConfusion ?
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 confusion.