Closed zhangbeile closed 4 years ago
Hi Wang I have a question how to test my own datasets like I have two ply I want to register one ply Thanks
发自我的iPhone
------------------ Original ------------------ From: Wang, Yue <notifications@github.com> Date: Thu,Nov 7,2019 10:46 PM To: WangYueFt/dcp <dcp@noreply.github.com> Cc: zhangbeile <a610725238@qq.com>, Author <author@noreply.github.com> Subject: Re: [WangYueFt/dcp] how to run my own dataset (#14)
You probably need to rewrite or modify the dataloader, then you're ready to train on your own data.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.
You could load the two ply and convert them to pytorch tensor. But I think there is work to do.
thanks
so I can use your code to register two ply file to one ply file right?I just change the two ply file convert them to pytorch tensor and register to one file why do i need label?if I need use v2 to test
You could try but I am afraid you still have to some coding working.
Best, Yue On Thu, Nov 7, 2019 at 9:07 PM zhangbeile notifications@github.com wrote:
so I can use your code to register two ply file to one ply file right?I just change the two ply file convert them to pytorch tensor and register to one file why do i need label?if I need use v2 to test
— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/WangYueFt/dcp/issues/14?email_source=notifications&email_token=ACMGCNX6MSD7C5SSNQDPRL3QSTCXZA5CNFSM4JKA43GKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEDOPMDI#issuecomment-551351821, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACMGCNQAO2EXRTWI6UI5OPTQSTCXZANCNFSM4JKA43GA .
Hello WangYue,I am an engineer form China. I am very excited about your work and i want to use it in my robotic grasp project. Do you have demo code to test two ply data ? or can you give me some detail advice about how to modify your code to test two ply data? thank you very much!
You probably need to rewrite or modify the dataloader, then you're ready to train on your own data.