sean

Hi guys, I'm also curious why we set the flip probabilities to say 0.5? Does it get better performance than doing it with 100% flip?

> > > 查看数据格式？查看文件不是七维矩阵 > > > […](#) > > > ------------------ 原始邮件 ------------------ 发件人: _**@**_._**>; 发送时间: 2022年6月18日(星期六) 中午12:15 收件人: "yanx27/Pointnet_Pointnet2_pytorch"; 抄送: "Mention"; 主题: Re: [yanx27/Pointnet_Pointnet2_pytorch] 跑自己的数据做语义分割，跑不通 (Issue #162) 请问我制作自己的数据集，遇到这个问题该怎么解决，我只放了少量数据进行尝试...

@mmiku1 make sure you change those parameters regarding classes number with both files. line 24, 92 in train_semseg.py line 24, 30 in /data_utils/S3DISDataLoader.py

> > @mmiku1 make sure you change those parameters regarding classes number with both files. line 24, 92 in train_semseg.py line 24, 30 in /data_utils/S3DISDataLoader.py > > Hello, If my...

> > @mmiku1 make sure you change those parameters regarding classes number with both files. line 24, 92 in train_semseg.py line 24, 30 in /data_utils/S3DISDataLoader.py > > Hello, If my...

@mmiku1 Hi, did you manage to train it successfully?

@bhagatdas https://github.com/yanx27/Pointnet_Pointnet2_pytorch/blob/eb64fe0b4c24055559cea26299cb485dcb43d8dd/data_utils/S3DISDataLoader.py#L59 change this number to a smaller one (say 120), this may resolve your problem.

facing similar problem

@rabbit-tree Hi, I also encountered similar problem. It runs the double "for loop" for a long time, I dont know if it's normal? Have you resolved this probelm? Thank you.

@MageeChan 我發現我卡在0%是因為它真的會跑很久，所以我在兩個for loop都讓他print一些東西，確保是有在執行的。 如果你的test沒有error的話，而是一直卡著，也可以等久一些看看。我印象中這大概跑了1天左右吧，依數據集不同可能會有差異。