microsoft/SoftTeacher

Issues

• 损失为零

  #247 opened 2 months ago by huangnana1
  0

• The model weight on google drive has expired

  #246 opened 2 months ago by WaterThunder
  0

• > When I use mmdet 2.17, the problem is solved.

  #245 opened 4 months ago by huangnana123
  1

• 训练了4000iters，验证完之后卡着了，请问这是怎么回事呢？

  #223 opened 2 years ago by mary-0830
  4

• if the batch is bigger , the training time is longer?

  #242 opened 10 months ago by 1shenhui
  1

• Environmental configuration

  #243 opened 8 months ago by Re-dot-art
  0

• Unable to train using 1 GPU

  #234 opened 2 years ago by amira-essawy
  5

• how I can use this in dota datasets?

  #241 opened 10 months ago by 1shenhui
  0

• ModuleNotFoundError: No module named 'ssod'

  #236 opened 2 years ago by Banks1y
  1

• How to calculate this parameter "epoch_length=7330"

  #231 opened 2 years ago by wzr0108
  1

• by mistake

  #240 opened a year ago by lamaMajedd
  0

• 如何处理的student多次forward 的loss

  #239 opened a year ago by Whale-ice
  2

• unable to train on custom dataset

  #200 opened 3 years ago by luisfra19
  1

• list index out of range

  #209 opened 3 years ago by xiangtaowong
  5

• sup：unsup

  #238 opened 2 years ago by thestars-maker
  0

• MMDetection v3.0

  #237 opened 2 years ago by austinmw
  0

• Welcome update to OpenMMLab 2.0

  #235 opened 2 years ago by vansin
  0

• for instance segmentation

  #233 opened 2 years ago by ahomez
  0

• sup_loss_bbox: nan and loss:nan

  #232 opened 2 years ago by Simeon340703
  1

• 在代码中看到TrimapMasks这个类，是做半监督实例分割要用的吗

  #230 opened 2 years ago by wzr0108
  0

• 在compute_pseudo_label_loss()的时候已经对teacher生成的伪标签做了坐标变换，为什么在unsup_rcnn_cls_loss()的时候还要变换一次呢？

  #229 opened 2 years ago by wzr0108
  0

• Cascade RCNN with Soft Teacher does not run: _bbox_forward() missing 1 required positional argument: 'rois'

  #228 opened 2 years ago by planaria158
  0

• assert len(indices) == len(self)

  #205 opened 3 years ago by RinggaV
  4

• colab

  #227 opened 2 years ago by alaa-shubbak
  0

• This repo is missing important files

  #213 opened 2 years ago by microsoft-github-policy-service
  2

• 关于PseudoSamples

  #226 opened 2 years ago by code-zjy
  0

• applying Yolo in place of Faster RCNN

  #221 opened 2 years ago by tahirashehzadi
  15

• Config files for evaluating the provided models

  #225 opened 2 years ago by Bai-YT
  0

• lr_config set up according to max iter?

  #224 opened 2 years ago by vefak
  0

• Inference and get the results of mAP accuracy between the resulted bounding box and the ground-truth one

  #222 opened 2 years ago by alaa-shubbak
  0

• Confusion about the $ sign in the config file

  #220 opened 2 years ago by Fengzeheng
  1

• Pretrained model weights for supervised baselines with partially labeled data?

  #219 opened 2 years ago by vuptran
  0

• Error in training

  #218 opened 2 years ago by linc4ekk
  0

• Multi-GPU training is reducing speed compared to single GPU

  #217 opened 2 years ago by tanvir-utexas
  0

• 关于baseline

  #216 opened 2 years ago by lzhhha
  0

• Training on Imbalanced Custom Data

  #215 opened 3 years ago by Hananali1
  0

• when init FasterRCNN, train_cfg and test_cfg are None

  #214 opened 3 years ago by wangyx-chn
  0

• ustom

  #211 opened 3 years ago by KobryLee
  0

• Getting mAP for all classes

  #203 opened 3 years ago by sarmientoj24
  2

• is dataset have transform_matrix？unsup_loss_bbox: 0.0000 when i try to train, I wanted to know if it was because of the data set, but my data set was fully trained and calculated on the supervised model？

  #206 opened 3 years ago by joeyslv
  7

• about test AP

  #210 opened 3 years ago by joeyslv
  0

• Plot the results of different iteration

  #208 opened 3 years ago by alaa-shubbak
  0

• unable to debug train.py on my custom dataset

  #201 opened 3 years ago by xiangtaowong
  3

• Too many boxes are overlapping as result

  #204 opened 3 years ago by sarmientoj24
  2

• mask_soft_teacher is not in the models registry'

  #207 opened 3 years ago by alaa-shubbak
  0

• training only for one classes

  #202 opened 3 years ago by alaa-shubbak
  1

• End of training Sample Ratio decay

  #197 opened 3 years ago by PeterVennerstrom
  2

• How to Visualize all the validation images?

  #199 opened 3 years ago by sarmientoj24
  0

• No validation on Custom dataset

  #198 opened 3 years ago by mjehanzaib999
  1

• How to change batch size of training?

  #196 opened 3 years ago by vefak
  1