Skip to content

qwerasdf887/TF2_YOLOv4

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

TF2_YOLOv4

Tensorflow 2.2版本的YOLOv4,幾乎個function都有相對應的註解。
若有任何問題歡迎討論,Model layer加入Drop Block layer。
Loss function : CIOU

環境 environment

  1. Tensorflow 2.2
  2. tensorflow_addons (moving average opt)
  3. Python 3.5~3.7
  4. OpenCV 3~4

Tiny version

Tiny-YOLOv4

Dropblock

train_default下的dropblock: [False, True, True, True, True]分別對應每個大的residual block後面增加。

Weights

ALexeyAB : download weight

YOLOv4架構如下

Predict Img:

python predict.py

Result

Training

需修改train.py。 使用labelImg 生成的xml檔,並且放入標籤位置即可。

About

YOLOv4 Tensorflow2-keras version

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages