YOLOv4_tensorflow

yolov4的tensorflow实现.
Implement yolov4 with tensorflow
持续更新
continuous update the code

中科院计算所二战失败考生，跪求好老师调剂收留

[email protected]
[email protected]

使用说明

introductions

直接执行命令, 程序就能执行
python train.py
just run the following command
python train.py

训练自己的数据集

train with own dataset

./data/JPEGImages 文件夹中存放用labelme标注json文件的jpg图片和对应的json文件, 参考我给的文件夹
The jpg image and the corresponding json file which marked with 'labelme' are stored in the folder "./data/JPEGImages", just like what I do

然后在 ./data 文件夹下执行 python 命令, 会自动产生 label 文件和 train.txt 文件
python generate_labels.py
and then, go to the folder "./data", execute the following python command, it automatically generates label files and train.txt
python generate_labels.py

继续执行命令,得到 anchor box
python k_means.py
excute the python command, to get anchor box
python k_means.py

打开 config.py, 将得到的 anchor box 写入到第六行，就像这样
anchors = 12,19, 19,27, 18,37, 21,38, 23,38, 26,39, 31,38, 39,44, 67,96
open config.py, write the anchor box to line 6, just like this

所有的配置参数都在 config.py 中，你可以按照自己的实际情况来修改
all configuration parameters are in the config.py, you can modify them according to your actual situation

配置完成,执行命令
python train.py
ok, that's all, execute the command
python train.py

有关 config.py 和训练的提示

some tips with config.py and train the model

config.py 中的 width 和 height 应该是 608，显存不够才调整为 416 的
the parameters of width and height in config.py should be 608, but i have not a powerful GPU, that is why i set them as 416
学习率不宜设置太高
learning rate do not set too high

自己的设备

my device

GPU : 1660ti (华硕猛禽)
CPU : i5 9400f
mem : 16GB
os : ubuntu 18.04
cuda: 10.2
cudnn : 7