search

ifzhang / FairMOT

[IJCV-2021] FairMOT: On the Fairness of Detection and Re-Identification in Multi-Object Tracking
MIT License
4.01k stars 934 forks source link

同时训练四个类别，进行tracking时报错 #353

Open HappyPeanuts opened 3 years ago

HappyPeanuts commented 3 years ago

Traceback (most recent call last): File "demo.py", line 44, in demo(opt) File "demo.py", line 33, in demo use_cuda=opt.gpus!=[-1]) File "FairMOT-master/src/track.py", line 90, in eval_seq online_targets = tracker.update(blob, img0) File "FairMOT-master/src/lib/tracker/multitracker.py", line 264, in update id_feature = id_feature[remain_inds] IndexError: boolean index did not match indexed array along dimension 0; dimension is 500 but corresponding boolean dimension is 420

HappyPeanuts commented 3 years ago

请问这是什么情况呢？期待有人可以提供一点帮助，万分谢谢

JH95-ai commented 3 years ago

看报错，应该是维度没对上

JAYCHOU2020 commented 3 years ago

Traceback (most recent call last): File "demo.py", line 44, in demo(opt) File "demo.py", line 33, in demo use_cuda=opt.gpus!=[-1]) File "FairMOT-master/src/track.py", line 90, in eval_seq online_targets = tracker.update(blob, img0) File "FairMOT-master/src/lib/tracker/multitracker.py", line 264, in update id_feature = id_feature[remain_inds] IndexError: boolean index did not match indexed array along dimension 0; dimension is 500 but corresponding boolean dimension is 420

你把代码改成了多类别的了吗？能请教一下如何改的吗？谢谢啦

HappyPeanuts commented 3 years ago

非常感谢你能回复，我代码里修改了类别，还是会报错，我看网上说FairMOT只支持单类别的多目标追踪，现在有一个任务需要多类别多目标，我还在研究怎么修改

------------------ 原始邮件 ------------------ 发件人: @.***>; 发送时间: 2021年6月16日(星期三) 下午2:22 收件人: "ifzhang/FairMOT"; 抄送: "Author"; 主题: Re: [ifzhang/FairMOT] 同时训练四个类别，进行tracking时报错 (#353)

看报错，应该是维度没对上

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

CaptainEven commented 3 years ago

@HappyPeanuts 请参考我的repo: https://github.com/CaptainEven/MCMOT

HappyPeanuts commented 3 years ago

好的好的，万分感谢

------------------ 原始邮件 ------------------ 发件人: "ifzhang/FairMOT" @.>; 发送时间: 2021年7月1日(星期四) 下午5:35 @.>; @.**@.>; 主题: Re: [ifzhang/FairMOT] 同时训练四个类别，进行tracking时报错 (#353)

@HappyPeanuts 请参考我的repo: https://github.com/CaptainEven/MCMOT

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

CaptainEven commented 3 years ago

@HappyPeanuts 觉得有帮助的话，欢迎star和fork

HappyPeanuts commented 3 years ago

好的好的，我还有一个疑惑，我在训练FairMOT的时候，我的损失值竟然得到了负值，这是什么情况？是因为本来的损失值设置就存在这种可能吗？

------------------ 原始邮件 ------------------ 发件人: "ifzhang/FairMOT" @.>; 发送时间: 2021年7月1日(星期四) 下午5:39 @.>; @.**@.>; 主题: Re: [ifzhang/FairMOT] 同时训练四个类别，进行tracking时报错 (#353)

@HappyPeanuts 觉得有帮助的话，欢迎star和fork

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

CaptainEven commented 3 years ago

loss < 0是正常的