-
Thanks for the great work!
I use 4 gpus to run script/main_segmentation.sh cfgs/s3dis/pointnext-xl.yaml. Only got val_oa 89.21, val_macc 76.35, val_miou 69.08 (Best ckpt @E61) which lower than the …
-
I have some errors when I run train.py based on your guideline on s3dis dataset. I also keep configuration files and install libraries like requirements.txt. Because I use Cuda 11.3, I install torch 1…
-
Hello,
I have used the previous versions of Softgroup and I see that the test_s3dis.py is removed.
How can I check performance on a validation set now for the s3dis dataset?
Thanks.
-
Hi,
This paper is a really interesting work!
I wonder when will you realize the code for ModelNet 10 and s3dis experiments?
Thank you!
-
Hi, thank you for your open codes.
Why do we need an offset when searching the k-nearest neighbours of a given central point?
-
Hi,
I seem to be running in to the same error as in #513 although as far as I can see the transforms are being set.
But I'm not sure if they are being applied.
I am running just the standard con…
-
### Checklist
- [X] I have searched for [similar issues](https://github.com/isl-org/Open3D/issues).
- [X] For Python issues, I have tested with the [latest development wheel](http://www.open3d.org…
-
Hi @ma-xu
Have you tested PointMLP on semantic segmentation datasets? For example, S3DIS dataset.
Thanks.
-
File "/home/liy0r/cvpr/contrastBoundary/tensorflow/models/../datasets/tf_s3dis_dataset.py", line 166, in
object_data = np.array([[float(x) for x in line.split()] for line in f])
ValueError: c…
-
py configs/pointnet2/pointnet2_ssg_16x2_cosine_50e_s3dis_seg-3d-13class.py --result work_dirs/pointnet2_ssg_16x2_cosine_50e_s3dis_seg-3d-13class/latest.pth --show-dir results
Traceback (most recent …