Sense-X/Co-DETR

Issues

• 我用swin-tiny做backbone代替swin-L为什么推理速度相差不大？进行推理时，模型自动把辅助头去除了吗？还是要手工去除？

  #142 opened 3 months ago
  4

• 多尺度测试

  #141 opened 5 months ago
  1

• The model and loaded state dict do not match exactly

  #140 opened 6 months ago
  2

• ModuleNotFoundError: No module named 'mmdet.registry'

  #139 opened 7 months ago
  1

• encoder co deformable detr

  #138 opened 4 months ago
  1

• summary of model

  #137 opened 7 months ago
  4

• 可视化问题

  #136 opened 4 months ago
  5

• ModuleNotFoundError: No module named 'projects'

  #135 opened 7 months ago
  2

• 如何获得像论文中一样的可视化注意力图

  #134 opened 7 months ago
  1

• start training

  #133 opened 7 months ago
  1

• Can't find torch

  #132 opened 8 months ago
  1

• Flops计算

  #131 opened 7 months ago
  2

• Running on CPU

  #130 opened 7 months ago
  3

• OpenAI api.Connection Error 国内无法连接

  #129 opened 8 months ago
  2

• AssertionError: scope builder exists in model registry

  #128 opened 8 months ago
  1

• KeyError: 'MotionDiffusion is not in the models registry'

  #127 opened 8 months ago
  1

• frozen ViT-L backbone

  #126 opened 8 months ago
  3

• [Solved] Why doesn't best.pth File Automatically Generate?

  #125 opened 8 months ago
  3

• Collaborative Hybrid Assignments Training 问题

  #124 opened 7 months ago
  1

• Which one in config is co-detr?

  #122 opened 7 months ago
  1

• Training memory requirement

  #121 opened 8 months ago
  7

• co_dino_5scale_r50_1x_coco这个版本下的训练日志可以提供下嘛

  #120 opened 7 months ago
  1

• test result unnormal

  #119 opened 8 months ago
  1

• 使用ViT-L训练时出错！

  #117 opened 9 months ago
  7

• vit.py是不是有问题？

  #116 opened 9 months ago
  2

• IndexError: list index out of range

  #115 opened 2 months ago
  1

• Co-DETR/mmdet/datasets/pipelines/transforms.py", line 2825, in __call__ assert 'mix_results' in results

  #114 opened 9 months ago
  0

• fatal: not a git repository (or any of the parent directories): .git

  #113 opened 9 months ago
  2

• Co-DETR w/ ViT-Large checkpoint

  #112 opened 9 months ago
  1

• Data Augmentation

  #111 opened 2 months ago
  1

• The SOTA model about Co-DETR w/ ViT-Large

  #110 opened 9 months ago
  1

• freeze most of the layers and train head

  #109 opened 2 months ago
  1

• mmcv error: libGL.so.1 Not Found; No module named 'fairscale'

  #108 opened 10 months ago
  2

• Which EVA-02 checkpoint did you use for SOTA LVIS?

  #107 opened 8 months ago
  12

• 为什么辅助head的loss权重这么大？

  #106 opened 10 months ago
  1

• The reasoning results obtained from the coco-64.1-map model are questionable.

  #105 opened a year ago
  1

• How to use the pre-trained model to make predictions, and where are the prediction files

  #104 opened 2 months ago
  1

• How much GPU memory is needed?

  #103 opened a year ago
  2

• Please help: Co-Detr can't work with single class object detection!

  #101 opened a year ago
  1

• "CoDETR: 'CoATSSHead is not in the models registry'

  #100 opened a year ago
  2

• where is Deformable-DETR

  #99 opened 2 months ago
  1

• 为什么我的显存开销非常大，这正常吗？

  #98 opened a year ago
  13

• How large GPU memory will it cost to train a CO-DETR model (Swin-L)?

  #97 opened a year ago
  3

• how to freeze the backbone

  #96 opened a year ago
  7

• May I ask how the visualization feature map of encoder discrimination score in the paper was obtained

  #95 opened 7 months ago
  2

• For instance segmentation

  #94 opened a year ago
  3

• KeyError: 'CoDETR is not in the models registry'

  #93 opened a year ago
  3

• TypeError: init_process_group() got an unexpected keyword argument 'port'

  #92 opened a year ago
  2

• it doesnt works!

  #91 opened a year ago
  0

• Transformer.py中，CoDinoTransformer的Decoder的key为什么设置为None，而不是memory呢

  #90 opened a year ago
  1