yjh0410/yolov2-yolov3_PyTorch

Issues

• 请问我在运行demo的时候出现这个问题怎么解决呢

  #60 opened 3 months ago by Haoyue-code
  0

• Why does box_w only divide by strides [-1]

  #25 opened 4 years ago by fengxiaoming520
  6

• Please add license

  #59 opened a year ago by Denny-kef
  1

• Can someone help me? Why txty_loss_function is BCEWithLogitsLoss?

  #58 opened 2 years ago by WHYwyh
  4

• About training yolov3 gradient explode problem

  #57 opened 2 years ago by kill2013110
  5

• class YOLO里的nms函数是否有错

  #55 opened 2 years ago by heibaidianying
  1

• The infer speed is too slow

  #56 opened 2 years ago by kill2013110
  0

• the loss is nan

  #53 opened 2 years ago by lostlose
  4

• [WARN] Cannot find rule for <class 'torch.nn.modules.container.Sequential'>. Treat it as zero Macs and zero Params.

  #54 opened 2 years ago by LSun0717
  0

• Unable to donwnload weight from the link?

  #52 opened 3 years ago by AI-Hunter
  0

• how to apply your code on other dataset?

  #51 opened 3 years ago by chenslcool
  0

• Cannot find rule for <class 'torch.nn.modules.container.Sequential'>. Treat it as zero Macs and zero Params.

  #48 opened 3 years ago by Albert-Gao-1
  1

• Why using sigmoid when decode_xy?

  #49 opened 3 years ago by chenslcool
  2

• how to get anchor box's size?

  #50 opened 3 years ago by chenslcool
  2

• Can this code run on GeForce GTX 2080 Ti？

  #47 opened 3 years ago by fanxxxxyi
  2

• Multiple processes of training

  #46 opened 3 years ago by beyza-yildirim
  3

• Google drive link of models

  #45 opened 3 years ago by rlrahulkanojia
  1

• about loss function

  #40 opened 4 years ago by Broad-sky
  10

• 下载了预训练集，yolo2 和yolo3的AP都达不到，大概只有30~40那个样子？

  #44 opened 3 years ago by zkbutt
  0

• A mistake in yolov2_d19.py

  #43 opened 3 years ago by jiangfeng94
  0

• A question abount num_classes

  #42 opened 4 years ago by AndSonder
  1

• Normalization

  #41 opened 4 years ago by developer0hye
  4

• 训练模型

  #39 opened 4 years ago by yjhdhr
  1

• ground truth width and height policy

  #7 opened 5 years ago by developer0hye
  12

• mAP of yolov3-tiny on VOC

  #38 opened 4 years ago by developer0hye
  8

• mis match erros in training Yolo_v2

  #37 opened 4 years ago by bruce1408
  1

• Why is training so slow

  #34 opened 4 years ago by lilslsls
  8

• pretrained darknet-light model

  #36 opened 4 years ago by developer0hye
  3

• utils.modules.Conv2d seems wrong

  #35 opened 4 years ago by TheLuminary
  5

• Normalized WH Prediction seems to get better performance

  #27 opened 4 years ago by developer0hye
  28

• Tiny Backbone Structure

  #26 opened 4 years ago by developer0hye
  2

• Training with different IGNORE_THRESH

  #24 opened 4 years ago by developer0hye
  6

• Hyper parameters for Training

  #19 opened 4 years ago by developer0hye
  2

• Random Crash during Training

  #18 opened 4 years ago by developer0hye
  2

• Swap Axis Augmentation

  #17 opened 4 years ago by developer0hye
  4

• May be an error in yolov3

  #15 opened 4 years ago by fengxiaoming520
  1

• Class Prediction Softmax vs Sigmoid

  #13 opened 4 years ago by developer0hye
  2

• The Input Image Normalization Policy used in your pretrained backbone network

  #4 opened 5 years ago by developer0hye
  7

• Share Pre-trained weight with Google Drive

  #2 opened 5 years ago by developer0hye
  2