lyuwenyu/RT-DETR

Issues

• about train_dataloader and val_dataloader

  #323 opened 15 days ago by hzy-del
  1

• 如何保存最好的pth文件

  #322 opened 15 days ago by hzy-del
  0

• 更换优化器

  #314 opened 19 days ago by todesti2
  4

• 后处理问题

  #317 opened 17 days ago by WuShaogui
  9

• Error: Type parameter (T) of Optype (Where) bound to different types (tensor(float) and tensor(double) in node (/model/decoder/Where).

  #307 opened 24 days ago by ramonhollands
  7

• 关于uncertainty-minimal Query Selection

  #321 opened 16 days ago by Nicole-1996
  0

• 将BCE改成CE无法收敛

  #315 opened 18 days ago by YoungsunPan
  5

• loss值无法下降

  #304 opened 16 days ago by todesti2
  3

• 固定随机种子后每次跑的结果不同

  #311 opened 22 days ago by BigworldNebula
  4

• 关于验证集loss

  #318 opened 17 days ago by ZZQ565697818
  1

• 请教有关batchsize的问题

  #310 opened 23 days ago by xuzefangfangtu
  2

• raise Valu_HybridEncoder already registeredeError('{} already registered'.format(cls.__name__))

  #316 opened 18 days ago by chhchhchh
  1

• 主干加载预训练权重复现问题

  #313 opened 19 days ago by ainawen
  5

• categories of pretrained weight problem

  #305 opened a month ago by jeonjj1
  5

• 论文中的预训练模型问题

  #309 opened 23 days ago by Godk02
  1

• No normalization for validation data during model evaluation

  #312 opened 20 days ago by Geo99pro
  2

• 关于GPU使用的问题

  #308 opened 24 days ago by TokaiTeio3
  1

• Why there are only 80 classes in `pred_logits`?

  #306 opened 25 days ago by pdt0609
  1

• what are the components in the pred_logits of the decoder?

  #298 opened 25 days ago by pdt0609
  1

• 如何可视化每个类别的AP值？

  #302 opened a month ago by Godk02
  7

• 单卡3090ti，源码pytorch版本训练问题

  #301 opened a month ago by Acc2386
  2

• 现在跑pytorch版本的模型，会爆现存，3080ti，是为什么呢

  #303 opened a month ago by Steven-MR
  1

• 关于trt模型推理

  #279 opened a month ago by NeKoooo233
  1

• 使用3090训练自己数据集时GPU利用率经常为0

  #300 opened a month ago by BigworldNebula
  1

• 关于训练时长的问题

  #278 opened 2 months ago by lkxxxxx
  4

• FPS相关问题

  #299 opened a month ago by txw-github
  4

• FPS-related issues

  #297 opened a month ago by txw-github
  0

• 模型训练时，ap一直为-1

  #295 opened a month ago by Steven-MR
  2

• It's urgent, why is it that my training and reasoning directly using the source code is so slow?

  #296 opened a month ago by txw-github
  0

• 论文急用！！——绘制训练曲线

  #275 opened 2 months ago by todesti2
  2

• The threshold used during verification

  #294 opened a month ago by chenhao324
  1

• 一元机场为什么这么便宜?

  #292 opened a month ago by jc1211d
  0

• hyperparameters

  #293 opened a month ago by FanglinLiu1
  0

• 关于调参

  #291 opened a month ago by wzhwantmoney
  2

• why is pytorch image normalization not applied?

  #289 opened a month ago by hweejuni
  1

• batch_size 和 num_workers的问题

  #290 opened a month ago by DaCheng1823
  1

• GPU训练

  #287 opened a month ago by DaCheng1823
  1

• changing backbone

  #274 opened a month ago by hweejuni
  3

• 关于不同层网络使用不同学习率的问题

  #288 opened a month ago by xiaoming-master
  1

• 关于deformable_attention_core_func

  #285 opened a month ago by ocrhei
  1

• Question: Frozen layers?

  #284 opened a month ago by aaronrmm
  1

• On Uncertainty Minimum Query Selection

  #286 opened a month ago by FanglinLiu1
  1

• IoU of model inferencing

  #283 opened 2 months ago by vishalvijay99
  1

• pytorch权重

  #280 opened 2 months ago by leoxxxxxD
  2

• 关于pytorch版推理代码的问题

  #282 opened 2 months ago by super-song-sir
  3

• Questions about real-time target detection model based on DETR

  #281 opened 2 months ago by FanglinLiu1
  1

• 关于训练不同batch size配置的问题

  #270 opened 2 months ago by Peterande
  4

• The validation loss was large and almost unchanged, and the training loss has been decreasing

  #277 opened 2 months ago by 20231211
  0

• 模型转换

  #273 opened 2 months ago by lxyjyy
  4

• pytorch版backbone

  #269 opened 2 months ago by changzipeng
  1