-
I have tried to train RegNet variants with strong augmentations, such as AutoAugment or CutMix.
But the performance can not be improved with them.
For example, I have reproduced the paper's resu…
-
I was training on my own dataset. The backbone was yolov3-tiny.
I added cutmix=1 on the [net] layer, but after I run the command
`./darknet detector train ...`
The training process just stopped …
-
In addition to TokenMix before the first Transformer block, have you considered or tried TokenMix in the middle of the model?
-
Hello, so I had taken the task of adding Bounding Boxes in FourierMix and CutMix from #1992, but was having trouble in implementing it.
I looked at the references for it as well, but am still stuck …
-
Hello,
I've been trying to run train the model on 4 GPUs but I am having some problems with the rank of the GPUs. In fact it cannot see the arguement. More specifically:
```shell
bash train.sh …
-
### 🚀 The feature
Sample different random parameters of Mixup/Cutmix for different elements of the batch to avoid loss instability in large batch setups.
### Motivation, pitch
Hello!
My understa…
-
Hello!
Thank you for sharing the code and the impressive results.
Will it be possible for you to please share the exact command involving main_finetune.py that has been used to finetune ViT-L on t…
-
Hi, Dose detectron2 achieve CutMix? I do not find CutMix in detectron2.
-
作者你好，我在我的数据集上运行了你的代码，实验结果很不错。但我有一个问题，就是在cut_mix_label_adaptive函数中，为什么要用到两次cutmix？这个本质不就是将一个更小的labeled区域转移到unlabeled吗？感觉用一次cutmix就行了。有点不太理解，希望作者解答一下，感谢！
-
hi. i'm interested in training custom dataset with YOLO.
i'm trying to train custom dataset using data augmentation like cutmix, mosaic, mixup.
but with cutmix, i got the error like the this. we…