[CVPR 2025] Mr. DETR: Instructive Multi-Route Training for Detection Transformers
active 2025-03-13 → 2026-04-20 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 404 days from 2025-03-13 to 2026-04-20. Pushes: 9 total, peak 4 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 24 total, peak 2 in a day. Comments: 68 total, peak 11 in a day. Stars: 128 total, peak 26 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| zhangchbin | 48 | 9 | 0 | 31 |
| ffxxy | 29 | 0 | 0 | 24 |
| Caoyy02 | 4 | 0 | 0 | 3 |
| t2387508130-collab | 3 | 0 | 0 | 3 |
| Han-lim | 2 | 0 | 0 | 1 |
| kasteric | 1 | 0 | 0 | 0 |
| Tt202333 | 1 | 0 | 0 | 0 |
| tanorcot | 1 | 0 | 0 | 0 |
| zss313 | 1 | 0 | 0 | 0 |
| lime-j | 1 | 0 | 0 | 1 |
| T2387508130 | 1 | 0 | 0 | 0 |
| yangrongkun | 1 | 0 | 0 | 0 |
| Gaofan666 | 1 | 0 | 0 | 0 |
| ajinkya-kulkarni | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#25tanorcot2026-01-13 01:31如何训练自己的数据集，在哪里修改路径？
- Issue#24Tt2023332025-12-25 03:16Checkpoint projects_dev/swin_large_patch4_window12_384_22kto1k.pth not found
- Issue comment#23Han-lim2025-12-05 05:07Question about train.max_iter setting in config file
- Issue#23Han-lim2025-12-05 05:07Question about train.max_iter setting in config file
- Issue comment#23zhangchbin2025-12-05 05:01Question about train.max_iter setting in config file
- Issue comment#22zhangchbin2025-11-02 07:11UnboundLocalError: local variable 'dn_query_embeds' referenced before assignment
- Issue comment#21zhangchbin2025-09-25 17:25请问为什么MOE层值添加在decoder的最后一层呢，因为每一层的预测都需要经过一对一以及一对多的匹配训练，如果只在最后一层采用MOE，其他层采用普通的FFN，这样是否影响训练效果呢
- Issue#21yangrongkun2025-09-22 11:52请问为什么MOE层值添加在decoder的最后一层呢，因为每一层的预测都需要经过一对一以及一对多的匹配训练，如果只在最后一层采用MOE，其他层采用普通的FFN，这样是否影响训练效果呢
- Issue#19zhangchbin2025-09-20 10:28内存泄漏问题projects/mr_detr_align/configs/models，这下面的swin配置文件训练，内存会不断增加，直到out of cuda
- Issue comment#20zhangchbin2025-09-20 07:53为什么gpu越多bs越大反而训练的时间越长，正常来说不应该时间变短吗？
- Issue#20T23875081302025-09-18 08:11为什么gpu越多bs越大反而训练的时间越长，正常来说不应该时间变短吗？
- Issue comment#19t2387508130-collab2025-09-15 02:24内存泄漏问题projects/mr_detr_align/configs/models，这下面的swin配置文件训练，内存会不断增加，直到out of cuda
- Issue comment#19zhangchbin2025-09-07 04:35内存泄漏问题projects/mr_detr_align/configs/models，这下面的swin配置文件训练，内存会不断增加，直到out of cuda
- Issue comment#19t2387508130-collab2025-09-06 07:45内存泄漏问题projects/mr_detr_align/configs/models，这下面的swin配置文件训练，内存会不断增加，直到out of cuda
- Issue comment#19t2387508130-collab2025-09-06 07:31内存泄漏问题projects/mr_detr_align/configs/models，这下面的swin配置文件训练，内存会不断增加，直到out of cuda
- Issue#17zhangchbin2025-08-25 14:29Are there any new operators introduced, if I want to convert to onnx format?
- Issue comment#18zhangchbin2025-08-25 14:29training a little bit slow
- Issue comment#17zhangchbin2025-08-16 15:29Are there any new operators introduced, if I want to convert to onnx format?
- Issue#17kasteric2025-08-14 07:48Are there any new operators introduced, if I want to convert to onnx format?
- Issue comment#16ffxxy2025-08-12 01:33Mr.DETR++预训练权重加载问题
- Issue comment#16ffxxy2025-08-11 06:41Mr.DETR++预训练权重加载问题
- Issue comment#16zhangchbin2025-08-11 06:39Mr.DETR++预训练权重加载问题
- Issue comment#16ffxxy2025-08-10 12:06Mr.DETR++预训练权重加载问题
- Issue comment#16zhangchbin2025-08-10 11:41Mr.DETR++预训练权重加载问题
- Issue comment#16ffxxy2025-08-10 11:37Mr.DETR++预训练权重加载问题
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 128 stars here means stars gained during the window, not the repo's star count.