Hello,
I was trying to produce some skeletons from your test example, but it seems like I receive no information through skel_xyz. All 100 points of the skeleton are similar to each other and the program crashes when attempting to compute the mesh.
I was using -
python test.py --pc_list_file ../data/data-split/all-test.txt --data_root ../data/pointclouds/ --point_num 2000 --skelpoint_num 100 --gpu 0 --load_skelnet_path ../weights/weights-skelpoint.pth --load_gae_path ../weights/weights-gae.pth --save_result_path ../results/
While having the dataset, the splits, and the pre-trained weights as instructed.
Also, when I try to train from scratch (following your instructions) I get:
NaN or Inf found in input tensor.
Any idea what could cause these issues?
Hello, I was trying to produce some skeletons from your test example, but it seems like I receive no information through skel_xyz. All 100 points of the skeleton are similar to each other and the program crashes when attempting to compute the mesh. I was using -
python test.py --pc_list_file ../data/data-split/all-test.txt --data_root ../data/pointclouds/ --point_num 2000 --skelpoint_num 100 --gpu 0 --load_skelnet_path ../weights/weights-skelpoint.pth --load_gae_path ../weights/weights-gae.pth --save_result_path ../results/
While having the dataset, the splits, and the pre-trained weights as instructed.
Also, when I try to train from scratch (following your instructions) I get:
NaN or Inf found in input tensor.
Any idea what could cause these issues?Thanks!