-
如题，我样本 最小的类和 最大类 数量 倍数达到了 50倍，tensorflow 里用 focal loss 起到了作用，请问 API方式啥时候提供？目前在用你们 API方式
-
Nice work！Thanks for your released code.
I have a question regarding conducting experiments on the `Barn` scene of the `Tanks` dataset using the default.yaml file, I found that the results were wor…
-
Hello all,
I've been trying to play around with modifying the loss function for solving class imbalance. I tried applying weights to the cross entropy loss and also replacing the cross entropy loss…
-
Hi!
Thank you for your work on this project!
I'm training the s-model on a custom dataset, and I’ve encountered an issue after several successful epochs. Up until the 12th epoch, training seems to…
-
### Reminder
- [X] I have read the README and searched the existing issues.
### System Info
- llamafactory version: 0.8.4 dev
- python 3.10
- ubuntu 20.04 Focal
- torch: 2.4
- accelerate: 0.32.…
-
## 前置准备https://github.com/feigechuanshu/ncnn-android-yolov8/issues/8
ultralytics\nn\modules\block.py
ultralytics\nn\modules\head.py
1.需要修改，ultralytics\nn\modules\block.py class C2f(nn.Module): …
-
Hi, I want to use PSPNet Model to segmentation 9 classes: Road, Line, Left, Right, NoLeft, NoRight, Straight, Stop.
This problem is my model can't segment correctly all signs when my self-driving ca…
-
I am wondering why loss_masks and in particular focal_loss only works with segmentation and not bbox prediction?
This is quite unexpected as focal loss can help class imbalance for bbox prediction …
-
你好, 感谢你的实现,有些细节处想和你请教下:
官方版本的label是只把center这一个点当作true, 剩下的都是负样本来计算focal_loss, focal_loss里面会找哪些位置的点的值是1, 但由于是在center附近打的一个高斯核, center处并不一定能保证是1? 或者说由于打得是个高斯核, center处有可能无限趋近于1但不是1, 在focal_loss里会去寻…
-
Thanks for your error report and we appreciate it a lot.
**Checklist**
1.Binary split out_ There is a conflict with Focal loss when channel=1!
**Describe the bug**
When performing binary s…