search

erikwijmans / Pointnet2_PyTorch

PyTorch implementation of Pointnet2/Pointnet++
The Unlicense
1.45k stars 333 forks source link

core dumped problem #48

Open lqzhao opened 5 years ago

lqzhao commented 5 years ago

Hi, thanks for your selfless code sharing. I encountered the following error when I run python -m pointnet2.train.train_cls. {'batch_size': 16, 'bn_momentum': 0.5, 'bnm_decay': 0.5, 'checkpoint': None, 'decay_step': 200000.0, 'epochs': 200, 'lr': 0.01, 'lr_decay': 0.7, 'num_points': 1024, 'run_name': 'cls_run_1', 'visdom': False, 'visdom_port': 8097, 'weight_decay': 1e-05} epochs: 0%| | 0/200 [00:00<?, ?it/sSegmentation fault (core dumped)

I don't know how to sovle it by myself, hope you can help me. THANKS!

erikwijmans commented 5 years ago

Segfaults can be very challenging to debug and I haven't seen this this one happen.

One thing:

gdb python
(gdb) r -m pointnet2.train.train_cls
<segfualt happens>
(gdb) bt
<call stack gets printed>
(gdb) f 0
<the code at frame 0 gets printed>

Can you copy paste the output from that scene of gdb commands?

HaozheQi commented 5 years ago

I have the same problem just like you, and I find out that Segmentation fault comes from the gcc version you use to compile the cuda code, the gcc version you use need to higher than 4.9.4. I update my gcc with the help of this blog https://blog.csdn.net/weixin_44259900/article/details/86508590