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

自动test和单独test结果不一致 #145

Open
jc991206 opened this issue Jan 18, 2024 · 1 comment
Open

自动test和单独test结果不一致 #145

jc991206 opened this issue Jan 18, 2024 · 1 comment

Comments

@jc991206
Copy link

我仅仅对pointvector.py和group.py做了更改,在进行S3DIS的area5测试的时候,为什么train完之后自动test的结果和我单独拿出best_bth进行test的结果不一样, 但我在运行原始的代码的时候,train之后的自动test和单独进行的test(CUDA_VISIBLE_DEVICES=0 bash script/main_segmentation.sh cfgs/s3dis/pointvector-xl.yaml wandb.use_wandb=False mode=test --pretrained_path /home/jichao/jc/code/PointNeXt-vector-original/examples/segmentation/log/s3dis/61/checkpoint/s3dis-train-pointvector-xl-ngpus1-seed7362-20240118-201954-m7diD5EE9smkfMuTtnUYbY_ckpt_best.pth)结果一样,并且效果高于72.3(论文中公布的结果)请问这是什么原因。

@sobigrain
Copy link

我仅仅对 pointvector.py 和 group.py 做了更改,在进行S3DIS的area5测试的时候,为什么train完之后自动test的结果和我单独拿出best_bth进行test的结果不一样, 但我在运行原始的代码的时候,train之后的自动测试和单独进行的test(CUDA_VISIBLE_DEVICES=0 bash script/main_segmentation.sh cfgs/s3dis/pointvector-xl.yaml wandb.use_wandb=False mode=test --pretrained_path /home/jichao/jc/code/PointNeXt-vector-original/examples/segmentation/log/s3dis/61/checkpoint/s3dis-train-pointvector-xl-ngpus1-seed7362-20240118-201954-m7diD5EE9smkfMuTtnUYbY_ckpt_best.pth)结果一样,并且效果高于72.3(论文中公布的结果)请问这是什么原因。

Note testing is a must step since evaluation in training is performed only on subsampled point clouds not original point clouds.

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

2 participants