Comments (10)
- I think one possible problem is that you may try to call iou_target.detach() for weighting the regression loss. It is easy to conduct the experiment by referring to the codes in this repo.
- In fact, the performances for "predicted score" and "calculated score" have no obvious differences in final performance. You can try either as you like.
from gfocal.
Hi,
Thank you for your quick response.
- I'm not sure if i understand your answer correctly. So in the FCOS experiment, we shouldn't normalize the regression loss by the iou target. The reason not to do this is because it may cause the initial training unstable ? please correct me if i'm wrong.
- Okay, thank you so much. I will give it a try.
from gfocal.
- We should normalize, but use a form like ``iou_target.detach()'' for normalizing as it contains predicted bounding box part which may introduce gradient propagation.
from gfocal.
Hi,
Yes, i already weighted the regression loss by iou target as FCOS did. In my FCOS IOU-branch experiment, it doesn't work. As training started, the iou target values are small(close to zero), probably because the initial prediction in regression branch is nothing. And then the iou branch tend to predict all zero iou value and then the training just failed. Did you use any trick like increase regression loss scale at the beginning of training ?
from gfocal.
It is a bit strange. In our experiments, we do nothing like increase regression loss scale at the beginning of training, and we do not observe training failures as you describe.
from gfocal.
Okay, thank you, so in your case, simply replace centerness branch to iou branch work, I will check my code if there is any problem.
from gfocal.
All right~
from gfocal.
I have the same question. When I replace the centerness target with the iou target, the regression loss and iou loss will all be zero. What's more, the giou loss cannot be calculated.
from gfocal.
You can simply run the code in this repo and carefully go through more details to check what is missing in your implementation.
from gfocal.
This is my label question. Thank you for your reply.
from gfocal.
Related Issues (20)
- How to draw correlation curve about IoU score and classification score? HOT 3
- 讨论下,GFL对比FL在误报率上有什么变化 HOT 3
- How to use model by C++?
- How to draw the distribution, like Fig.3 HOT 1
- TypeError: conv2d(): argument 'input' (position 1) must be Tensor, not DataContainer HOT 2
- About the alpha in focal loss.
- How to get the ground truth of classification-IoU score?
- test.py
- No such file or directory: 'data/coco/annotations/instances_val2017.json'
- centernet中Focal loss换GFocal HOT 1
- what the shapes are in pred and label of distribution_focal_loss
- Model training time
- fds
- How to get the quality scores
- 你好,有模型下载的国内链接吗?
- TypeError: ATSS: FPN: __init__() got an unexpected keyword argument 'extra_convs_on_inputs' HOT 2
- What means pos and pos_inds ?
- Questions about QFL
- 可以在anchor free上使用么
- Welcome update to OpenMMLab 2.0
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 gfocal.