wmrenr

> Yes, since I didn't find the code neither the `untargeted_DPATCH.npy`. I train the model again, `python train.py` Have you solved it? I can't also find 'untargetd'_ DPATCH.npy'

> > > > Hello, thanks for the current release. I have finished the demo part and got a perfect pairwise registration result. But the current release only supports training...

But this version seems too old, and the compilation process reported an error ： nvcc fatal：Unsupported gpu architecture 'compute_30'. my gpu is RTX3080Ti

> @heiwang1997 I tried to train on you datasets. All of them are OK. I guess maybe because the data I generated cannot be learned. Could I ask a question?...

@lixinghe1999 大佬可否发我一份声呐图像数据集，来晚一步前面的都失效了，[email protected] ，感激不尽

> I am also trying this problem, have you solved it?

I print torch.from_numpy(labels_cmap).shape :[512,512,3] and the 3D transpose should be performed, so I cant understand '.permute([0, 3, 1, 2])' in your codes.

The number of my datasets : trainset:968 valset:323 visual_test:322 The training parameter setting: location: data_root_dir: /home/jddx/wxp/lama/my_dataset/ out_root_dir: /home/jddx/wxp/lama/experiments/ tb_dir: /home/jddx/wxp/lama/tb_logs/ data: batch_size: 2 val_batch_size: 2 num_workers: 2 train: indir: my_dataset/train...

> indir: my_dataset/val/random_medium_512 I solve this problem by changing the path behind indir to an absolute path，such as ：home/jddx/wxp/lama/my_dataset/val/random_medium_512

> > > indir: my_dataset/val/random_medium_512 > > > > > > I solve this problem by changing the path behind indir to an absolute path，such as ：home/jddx/wxp/lama/my_dataset/val/random_medium_512 > > 求问大佬，我将abl-04-256-mh-dist.yaml中的val、train以及visual_test的indir都改成了绝对路径，在打印的config中显示例如indir为：...