Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

loss_bbox和loss_dfl都为0 #581

Open
sosgu opened this issue Feb 25, 2025 · 0 comments
Open

loss_bbox和loss_dfl都为0 #581

sosgu opened this issue Feb 25, 2025 · 0 comments

Comments

@sosgu
Copy link

sosgu commented Feb 25, 2025

我在终端执行
CUDA_VISIBLE_DEVICES=1 python tools/train.py configs/finetune_coco/yolo_world_v2_l_vlpan_bn_sgd_1e-3_40e_8gpus_finetune_coco.py --work-dir log --amp --resume
命令后,训练之后的显示
style="text-decoration-style:single">INFO</u></font> - Epoch(train) [25][50/90] base_lr: 1.0000e-03 lr: 4.3075e-04 eta: 0:09:11 time: 0.3959 data_time: 0.0126 memory: 5473 grad_norm: 0.0283 loss: 0.0016 loss_cls: 0.0016 loss_bbox: 0.0000 loss_dfl: 0.0000
我在网上搜索相关的网页说要保持json文件里面标签要一直,但是我将其修改后还是这个效果,所以我想在这里咨询一下大家,这个问题应该怎么解决;
我在训练集中的类别为
"categories": [
{
"id": 0,
"name": "head",
"supercategory": "head"
}
]
在测试集中的类别为
"categories": [
{
"id": 0,
"name": "head",
"supercategory": "head"
}
]
coco_class_texts.json里面的内容为
[["head"]]

最后是yolo_world_v2_l_vlpan_bn_sgd_1e-3_40e_8gpus_finetune_coco.py文件中的为
metainfo = dict(classes=("head"))
coco_train_dataset = dict(delete=True,
type='MultiModalDataset',
dataset=dict(
type='YOLOv5CocoDataset',
metainfo=metainfo,
只截取了部分代码,还请各位大佬能够指导一下

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant