Open sunjunling666 opened 4 years ago
yes,I also meet.It's so sad.
yes,I also meet.It's so sad.
Have you solved it?
yes,I also meet.It's so sad.
Have you solved it?
hello,then I just use the bash download.sh
.it all finish.first, i set my code in Windows,it doesn't work. so i change the OS into Ubuntu,it really works. wish it be helpful.
yes,I also meet.It's so sad.
Have you solved it?
hello,then I just use the
bash download.sh
.it all finish.first, i set my code in Windows,it doesn't work. so i change the OS into Ubuntu,it really works. wish it be helpful.
I guess that the download.sh only is the source of segmentation dataset instead of modelnet40. Do you use the segmentation dataset?
yes,I also meet.It's so sad.
Have you solved it?
hello,then I just use the
bash download.sh
.it all finish.first, i set my code in Windows,it doesn't work. so i change the OS into Ubuntu,it really works. wish it be helpful.I guess that the download.sh only is the source of segmentation dataset instead of modelnet40. Do you use the segmentation dataset?
hello,sorry,I don't solve it.
download.sh contains information about shapenet dataset only. From where will we get modelnet40 dataset. I have tried manually downloading modelnet40, but when run the command got errors.
python3 train_classification.py --dataset /.../ModelNet40 --nepoch=10 --dataset_type modelnet40
Namespace(batchSize=32, num_points=2500, workers=4, nepoch=10, outf='cls', model='', dataset='home/.../ModelNet40', dataset_type='modelnet40', feature_transform=False)
Random Seed: 2348
Traceback (most recent call last):
File "/.../pointnet.pytorch/utils/train_classification.py", line 57, in
hey, having the same problem:
All the files from ModelNet40 is .off files.
Edit: It seems that we have to figure out how to preprocess the .off files into .ply.