bubbliiiing/yolov4-pytorch

Issues

• yolotraining.py中参数问题

  #340 opened 2 years ago by ChenMaolong
  5

• 默认参数没有调整，loss一直是nan

  #379 opened a year ago by yandlee
  2

• 请问yolov4的预训练权重从哪下载呢

  #383 opened 7 months ago by starrysealuck
  1

• RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [4096]] is at version 3; expected version 2 instead.

  #382 opened 8 months ago by CXCxyz
  1

• File "mtrand, pyx", line 934, in numpy.random. mtrand. RandomState. choiceValueError: a cannot be empty unless no samples are taken

  #381 opened a year ago by 503455003
  0

• 计算量怎么查看？

  #380 opened a year ago by 503455003
  1

• train_loss下降，val_loss不变，map一直不变，自建数据集预测效果很差。

  #363 opened 2 years ago by Chris-zixuan
  9

• 求教predict和get_map中detect_image的区别

  #378 opened a year ago by baojunqi
  0

• AP很低

  #375 opened a year ago by wangminj
  2

• 请求

  #377 opened a year ago by intellectuall
  1

• 导，你的heatmap生成原理是什么？

  #376 opened a year ago by lty1234-deep
  0

• 请问加载yolov5_s.pt权重时为什么会报下面这个问题

  #346 opened 2 years ago by NuaaYH
  1

• 1

  #373 opened a year ago by 1qingning
  0

• please help me，why are the images of the validation set used during the training process?

  #372 opened a year ago by tianhaoliuzhu
  1

• get_map很低

  #371 opened a year ago by songjiahao-wq
  0

• loss计算太慢

  #368 opened 2 years ago by ZGJY95
  7

• 找不到dropblock，SAF等方法的人看这里

  #367 opened 2 years ago by Jamesgender
  1

• 你好，想问下更改预测框的颜色字体粗细大小 要如何更改啊

  #366 opened 2 years ago by cycnb233
  3

• 替换YOLOv4主干网络为ghost或者mobilenetv2/3，检测精度直接为0或者零点几？

  #365 opened 2 years ago by lyl-9
  14

• 验证损失低于训练损失

  #364 opened 2 years ago by huansu
  2

• 博主大佬，yolov4单小目标检测怎么找指标最好的权重，是根据F1指标还是？

  #334 opened 2 years ago by ChenMaolong
  6

• kmeans聚类问题？

  #358 opened 2 years ago by ChenMaolong
  2

• opencv加载图像进行推理

  #361 opened 2 years ago by PengPeng-JunJun
  5

• train_loss下降正常，val_loss不变，map一直很低，实际预测检测效果很差。

  #362 opened 2 years ago by Chris-zixuan
  2

• 预训练模型问题

  #359 opened 2 years ago by PengPeng-JunJun
  5

• 关于如何在MS coco测试集上验证自己模型的精度的问题

  #360 opened 2 years ago by WDQGO
  0

• 请问一下关于运行get_map.py报错的问题

  #356 opened 2 years ago by sticktn
  5

• 训练问题

  #345 opened 2 years ago by guanxuefeng
  8

• AttributeError: 'list' object has no attribute 'cuda'

  #357 opened 2 years ago by Jie-Huangi
  4

• 先验框

  #354 opened 2 years ago by zhuasd
  4

• 图像增强

  #353 opened 2 years ago by zhuasd
  4

• RuntimeError: Error(s) in loading state_dict for YoloBody:

  #355 opened 2 years ago by xuwentai
  0

• 使用yolov4，精度很低，同样的数据集用yolov5、yolox精度正常

  #336 opened 2 years ago by lunaWU628
  6

• 可以输出模型的参数量和浮点运算数吗

  #352 opened 2 years ago by zlr-rrr
  1

• RuntimeError: [enforce fail at ..\c10\core\CPUAllocator.cpp:73] data. DefaultCPUAllocator: not enough memory:问题

  #350 opened 2 years ago by WeaverCai
  1

• 大佬你是在COCO上训练多少个Epoch?

  #351 opened 2 years ago by Lemonzlj
  1

• yolo4_voc_weights.pth权重问题？

  #347 opened 2 years ago by ChenMaolong
  5

• yolo4_voc_weights.pth权重问题？

  #348 opened 2 years ago by ChenMaolong
  1

• 损失函数图

  #344 opened 2 years ago by AgSword
  9

• 怎么找最优权重？

  #342 opened 2 years ago by ChenMaolong
  7

• 对于损失计算方式的疑惑

  #343 opened 2 years ago by zhiyiYo
  2

• 单类别小目标检测在yolo.py文件中修改confidence,nms_iou指标对map,recall,precision结果没有作用？

  #338 opened 2 years ago by ChenMaolong
  4

• 检测框和真实框相差不大，但是检测率很低，是要改什么指标吗，单类别小目标检测？

  #339 opened 2 years ago by ChenMaolong
  5

• 权重文件的下载方式

  #341 opened 2 years ago by zhiyiYo
  3

• 怎么提高recall?

  #337 opened 2 years ago by ChenMaolong
  1

• PR曲线达不到（1，p）这一点

  #329 opened 2 years ago by lunaWU628
  5

• 运行predict.py后报错

  #333 opened 2 years ago by liutangming
  1

• 如何从断掉的地方继续开始训练

  #332 opened 2 years ago by songjiahao-wq
  1

• map为-1

  #331 opened 2 years ago by songjiahao-wq
  2

• 生成annotation的train.txt和val.txt问题，请帮忙

  #330 opened 2 years ago by songjiahao-wq
  0