一、训练环境
Windows10下编译好的darknet
编译过程:https://blog.youkuaiyun.com/weixin_54603153/article/details/119980266?spm=1001.2014.3001.5501)
源码地址:https://github.com/AlexeyAB/darknet
二、制作自己的数据集
1.首先在darknet-master(我们下载好的源码)文件夹下创建放数据集的文件夹
Annotations放标签xml文件,
JPEGImages放照片,
Main什么都不放创就行了,一会脚本会自动给里面生成
labels不用创,一会脚本会自动生成
2.运行脚本make1把图片分为训练集,测试卷,验证集
import os
import random
trainval_percent = 1 #可以自己修改
train_percent = 0.9 #可以自己修改
xmlfilepath = 'VOCdevkit/VOC2007/Annotations'
txtsavepath = 'VOCdevkit/VOC2007/ImageSets/Main'
total_xml = os.listdir(xmlfilepath)
num=len(total_xml)
list=range(num)
tv=int(num*trainval_percent)
tr=int(tv*train_percent)
trainval= random.sample(list,tv)
train=random.sample(trainval,tr)
ftrainval = open('VOCdevkit/VOC2007/ImageSets/Main/trainval.txt', 'w')
ftest = open('VOCdevkit/VOC2007/ImageSets/Main/test.txt', 'w')
ftrain = open('VOCdevkit/VOC2007/ImageSets/Main/train.txt', 'w')
fval = open('VOCdevkit/VOC2007/ImageSets/Main/val.txt', 'w')
for i in list:
name=total_xml[i][:-4]+'\n'
if i in trainval:
ftrainval.write(name)
if i in tr