ShiqiYu/libfacedetection.train

Issues

• Applying the model to other tasks

  #97 opened 2 months ago by GrumpyChubbyCat
  1

• is the yunet_n_640_640.onnx crushed?

  #96 opened 6 months ago by mdyuan926
  0

• 请问怎么解决KeyError: 'YuNet is not in the models registry'的报错呢

  #95 opened 9 months ago by YousterGitHub
  0

• 如何训练

  #94 opened 9 months ago by TheTurtle2020
  1

• 训练无关键点模型

  #93 opened 9 months ago by liujianisme
  0

• 使用 compare_inference.py 中的 YUNET.forward 與 cv.FaceDetectorYN.detect 輸出結果不同

  #92 opened 10 months ago by Tsao666
  0

• 带出onnx出错

  #90 opened a year ago by zaidao2023
  1

• Not able to setup the training environment

  #88 opened a year ago by Gopal518
  9

• 为啥数据不做归一化

  #89 opened a year ago by suke27
  1

• Batch inference problem!

  #87 opened a year ago by hopeux
  1

• Train on rotated images to improve landmarks.

  #83 opened a year ago by mcourteaux
  1

• pytrch2onnx convervsion issue

  #84 opened a year ago by ZubairKhan001
  2

• Problems installing the repo

  #86 opened a year ago by LuisGF93
  1

• 训练完成导出C++部署不成功

  #85 opened a year ago by nianfd
  0

• The problem has been solved

  #82 opened a year ago by leiqinhong
  0

• Did you try Focal loss, how about its performance ?

  #81 opened a year ago by YunYang1994
  1

• Why can't I find the yunet_final.pth weight file

  #80 opened a year ago by leiqinhong
  2

• nonsquare input size training and exporting

  #73 opened a year ago by jimkuo2
  1

• How to train with custom dataset by using the pretrained model?

  #74 opened a year ago by SwEngine
  2

• 老师，如果在Win上配置了MMDET环境后，如何修改指令执行train.py呀？

  #79 opened a year ago by BandyKenny
  5

• 训练脚本模型性能复现

  #39 opened 3 years ago by xiaoxi91
  3

• 模型是否支持关键点输出

  #65 opened 2 years ago by yonglianglan
  8

• 请问余老师，我看源码，图片在送入网络前会转成27通道，我需要训练单通道的图片，应该怎么修改呢

  #37 opened 2 years ago by kevinkinking
  4

• 最小像素点修改

  #76 opened 2 years ago by alizwl
  1

• 图像预处理时的均值和方差为什么是[0,0,0]和[1,1,1]

  #77 opened 2 years ago by zaidao2023
  1

• where can we get datasets with key points to train Yunet?

  #75 opened 2 years ago by yjump
  0

• Cannot download labelsv2

  #69 opened 2 years ago by AlessandroMondin
  3

• How to generate json files for NVIDIA dali?

  #54 opened 2 years ago by ithmz
  8

• 当转onnx的时候，验证pytorch模型推理和onnx推理时候出现问题

  #66 opened 2 years ago by yonglianglan
  2

• can not export to onnx

  #63 opened 2 years ago by yonglianglan
  2

• 推理结果好像有问题

  #61 opened 2 years ago by yonglianglan
  8

• 关于Onnx 模型无法在前端的Onnxruntime-web上正常使用的问题

  #60 opened 2 years ago by dayzorro
  1

• 支持戴口罩人脸检测吗

  #59 opened 2 years ago by wuhongsheng
  1

• About the version update in mmdet framework

  #58 opened 2 years ago by cubicimage
  3

• Apply for priors box parameters of 'onnx' model

  #55 opened 2 years ago by enemy1205
  5

• 模型为640x640,转onnx输入尺寸不同，没有报错是什么原因？

  #50 opened 2 years ago by xingjunhong
  9

• May I ask do you still provide the training code for the model published at opencv github repo?

  #53 opened 2 years ago by ithmz
  1

• [Docs] Visualization of network architecture points to older onnx file link

  #51 opened 2 years ago by kishore-s-15
  4

• yunet_yunet_final_320_320_simplify.onnx not working.

  #49 opened 2 years ago by yukyon
  4

• 请问有拉流摄像头实时检测的代码吗？

  #48 opened 2 years ago by xingjunhong
  6

• can't find from .bbox import

  #47 opened 2 years ago by xingjunhong
  4

• Hello, can you write this code for training in C ++?

  #42 opened 2 years ago by problemSolvingProgramming
  3

• only 1024 is supported?我想在100x100的图片上训练，应该怎么办呢？

  #36 opened 3 years ago by kevinkinking
  1

• cannot import name '_DaliBaseIterator' from 'nvidia.dali'

  #34 opened 3 years ago by bucky999
  1

• training loss fluctuates

  #29 opened 3 years ago by nhatanhphung
  7

• 请问余老师如何理解网络结构中ConvDPUnit

  #28 opened 3 years ago by XuhongLuo
  2

• GPU利用率低

  #27 opened 3 years ago by sunny-sjj
  7

• new annotation format

  #26 opened 3 years ago by nhatanhphung
  1

• Will you release train codes for v3?

  #25 opened 4 years ago by on-your-way
  3

• 训练完一个epoch后没反应

  #24 opened 4 years ago by dengly
  2