search

traveller59 / second.pytorch

SECOND for KITTI/NuScenes object detection
MIT License
1.72k stars 721 forks source link

Filtering algorithm, delete all label in my custom LiDAR dataset #444

Open curiousboy20 opened 3 years ago

curiousboy20 commented 3 years ago

Hi, I am working with my own custom LiDAR data. However, the filter algorithm delete all label data that I have:

middle_class_name PointPillarsScatter num_trainable parameters: 66 {'Pedestrian': 5} [-1] load 0 Car database infos load 879 Pedestrian database infos load 0 Cyclist database infos load 0 Van database infos load 0 Person_sitting database infos load 0 Truck database infos load 0 Tram database infos load 0 Misc database infos After filter database: load 0 Car database infos load 0 Pedestrian database infos load 0 Cyclist database infos load 0 Van database infos load 0 Person_sitting database infos load 0 Truck database infos load 0 Tram database infos load 0 Misc database infos remain number of infos: 150 remain number of infos: 150 WORKER 0 seed: 1614466470 WORKER 1 seed: 1614466471 reset Pedestrian reset Pedestrian reset Pedestrian reset Pedestrian reset Pedestrian

and at the end it is error as espected with output below: raise ValueError('need at least one array to stack') ValueError: need at least one array to stack

How to fix this? Any help appreciated.

Thanks

clare19997 commented 3 years ago

hi,did you solove it? Can I talk with you about train own data

curiousboy20 commented 3 years ago

hi,did you solove it? Can I talk with you about train own data

Unfortunately, it has not solved yet. I still figure it out. Do you know how to solve it? Or do you have a similar problem?

clare19997 commented 3 years ago

hi,did you solove it? Can I talk with you about train own data

Unfortunately, it has not solved yet. I still figure it out. Do you know how to solve it? Or do you have a similar problem?

hi,did you solove it? Can I talk with you about train own data

Unfortunately, it has not solved yet. I still figure it out. Do you know how to solve it? Or do you have a similar problem?

May we use email to communicate details, I have some questions consultation ？email address 940568073@qq.com

clare19997 commented 3 years ago

hi, what's your data fomat and data context?and How do you make your data set

------------------ 原始邮件 ------------------ 发件人: "traveller59/second.pytorch" @.>; 发送时间: 2021年3月17日(星期三) 凌晨3:46 @.>; @.**@.>; 主题: Re: [traveller59/second.pytorch] Filtering algorithm, delete all label in my custom LiDAR dataset (#444)

hi,did you solove it? Can I talk with you about train own data

Unfortunately, it has not solved yet. I still figure it out. Do you know how to solve it? Or do you have a similar problem?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

zijianh4 commented 2 years ago

Hi, I am working with my own custom LiDAR data. However, the filter algorithm delete all label data that I have:

middle_class_name PointPillarsScatter num_trainable parameters: 66 {'Pedestrian': 5} [-1] load 0 Car database infos load 879 Pedestrian database infos load 0 Cyclist database infos load 0 Van database infos load 0 Person_sitting database infos load 0 Truck database infos load 0 Tram database infos load 0 Misc database infos After filter database: load 0 Car database infos load 0 Pedestrian database infos load 0 Cyclist database infos load 0 Van database infos load 0 Person_sitting database infos load 0 Truck database infos load 0 Tram database infos load 0 Misc database infos remain number of infos: 150 remain number of infos: 150 WORKER 0 seed: 1614466470 WORKER 1 seed: 1614466471 reset Pedestrian reset Pedestrian reset Pedestrian reset Pedestrian reset Pedestrian

and at the end it is error as espected with output below: raise ValueError('need at least one array to stack') ValueError: need at least one array to stack

How to fix this? Any help appreciated.

Thanks

Hi, I also met this problem. Have you solved it? If yes, could you please tell me the cause? Thx!