labixiaochao
labixiaochao
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。
我想使用flash attention运行程序 而不是# Scratched S3DIS, S3DIS rely on RPE, also an example for disable flash attention sh scripts/train.sh -g 4 -d s3dis -c semseg-pt-v3m1-0-rpe -n semseg-pt-v3m1-0-rpe 是不是使用flash attention显存占用会减少
How do you visualize the two datasets, shapenetpat and scanobjectnn, and if there is a way to visualize them, please send them to my email [email protected]:issue is:open
How do you visualize the two datasets, shapenetpat and scanobjectnn, and if there is a way to visualize them, please send them to my email [email protected]. Thank you very much.
**How to visualize the results of scanobjectnn and shapenetpart *
How to train all areas of S3Dis at once
How to generate visualization files of scanobjectnn and shapenet and visualize them Thank you very much
The avss dataset does not have folder classifications like S4 and ms3, such as all gtmask, viusual frames How do I use it, it is defined as a folder per...
请问可以分享一下CUHK-PEDES和ICFG-PEDES数据集吗，万分感谢，我的教育邮箱是这个[email protected]