HikariTJU/LD

Issues

• Can we pass different set of image to teacher and student.

  #77 opened 7 months ago by anshum5024
  6

• loadcheckpoint出现问题

  #79 opened 4 months ago by 622tongtong
  2

• samples_per_gpu

  #78 opened 6 months ago by swift1988
  1

• 将DKD与LD结合的问题

  #67 opened 6 months ago by 1014llj
  8

• Reproduce LD in yolov8?

  #59 opened a year ago by wangtao2668129173
  12

• About Adaptive Layer

  #76 opened 7 months ago by minhhotboy9x
  2

• 请问是使用 ldv2/ld_r50_gflv2_r101_fpn_1x.py 这个文件来复现 Temperature τ in LD 这个实验吗?

  #73 opened 7 months ago by JinYu1998
  1

• kd_loss implementation issue

  #74 opened 9 months ago by ZaberKo
  9

• 关于数据集minicoco

  #75 opened 9 months ago by aowuaaa
  4

• 知识蒸馏

  #72 opened 9 months ago by cxq1
  10

• 参考install.md 里面配置的环境无法找到cuda

  #65 opened a year ago by JinYu1998
  13

• 请问下是否可以迁移到centernet呢

  #71 opened a year ago by nancy618618
  2

• mmpycocotools

  #70 opened 10 months ago by cxq1
  1

• 学生网络与教师网络性能差距较大

  #69 opened a year ago by TinyOD
  0

• Implement LD on VisDrone dataset

  #68 opened a year ago by melika-sce
  5

• 简易版本复现问题

  #66 opened a year ago by luckin99
  5

• If I set output_feature = False , Code don't work

  #64 opened a year ago by shaby3
  1

• 请问下代码里的教师模型是跟着学生模型一起训练的吗

  #63 opened a year ago by touchsome
  2

• 耦合头效果怎么样

  #57 opened 2 years ago by huangjun12
  6

• Where is definition of `LocalizationDistillationLoss`?

  #47 opened 2 years ago by songtianhui
  5

• Some Questions about the VLR ?

  #32 opened 2 years ago by ttjjmm
  4

• 如何对yolox进行知识蒸馏

  #38 opened 2 years ago by Zhao-Jiayi
  6

• Welcome update to OpenMMLab 2.0

  #62 opened a year ago by vansin
  2

• 请问如果使用自己的feature-imitation方法应该在哪里添加，您的feature-imitation代码写在哪里了，我没有找到

  #60 opened a year ago by yg333
  6

• I reimplement your model (Teacher: R101, Student: R50, Detection head: Foveabox) and got RuntimeError: shape '[-1, 68]' is invalid for input of size 115200

  #61 opened a year ago by letruongngochai
  5

• FCOS_GFL_R101_2x的教师权重文件可以提供一下吗？

  #58 opened a year ago by TinyTigerPan
  3

• 教师助理策略

  #56 opened 2 years ago by yuhua666
  1

• 如何蒸馏自定义模型

  #55 opened 2 years ago by yuhua666
  3

• 想在VOC上进行训练时报错

  #54 opened 2 years ago by mk-fault
  4

• 将LD的骨干网更改为shufflenetV2后的问题

  #53 opened 2 years ago by cape-zck
  4

• loss_kd_neg计算的意义

  #52 opened 2 years ago by huangjun12
  3

• 请问Main 和 VLR 这两个区域是怎么确定的？

  #49 opened 2 years ago by yangyangtiaoguo
  5

• 关于回归的变量离散化选择

  #50 opened 2 years ago by botaoye
  1

• 关于Two-Stage Detector Distillation 的实现问题

  #51 opened 2 years ago by lanqz7766
  1

• 请问是否可以使用其他预训练教师模型

  #48 opened 2 years ago by changxin111
  3

• subprocess.CalledProcessError: Command '[]' returned non-zero exit status 1.

  #46 opened 2 years ago by szwlh-c
  2

• config问题

  #44 opened 2 years ago by TinyTigerPan
  6

• AP landscape

  #45 opened 2 years ago by jwfanDL
  2

• Both of GFL and LD use a distribution to model bbox, what's the difference between LD and GFL?

  #42 opened 2 years ago by DryPilgrim
  0

• 关于分类蒸馏的问题

  #43 opened 2 years ago by shuowang666
  2

• 关于ld_head.py 中loss的问题？

  #41 opened 2 years ago by yangyangtiaoguo
  1

• TypeError: KnowledgeDistillationSingleStageDetector: __init__() got an unexpected keyword argument 'output_feature'

  #40 opened 2 years ago by zhangnan12d
  7

• 请问LD中对预测框的概率分布的回归是用的类似GFocal中一个卷积层进行学习的吗

  #39 opened 2 years ago by Zhao-Jiayi
  2

• How can I train my own VOC datasets?

  #36 opened 2 years ago by Audrey528
  16

• 请问教师模型是在哪里训练得到的？

  #37 opened 2 years ago by FlyGreyWolf
  2

• 使用LD的参数变化

  #34 opened 2 years ago by Audrey528
  1

• Whiat TOOL of Feature Visualization Do YOU USE?

  #35 opened 2 years ago by Audrey528
  1

• 关于GI(General Instance) Feature-based Distillation 实现的问题

  #33 opened 2 years ago by flazerain
  8

• Funny Code !!!

  #30 opened 2 years ago by ttjjmm
  1

• I dont understand the meaning of these codes?

  #31 opened 2 years ago by ttjjmm
  0