資源簡介
源項目
https://github.com/sfzhang15/RefineDet
基于caffe搭建RefineDet并訓練自己的模型,
實測對于小目標的檢測強于YOLOv3
注:文檔c步驟第一步要在RefineDet_root/data 下新建VOC2008文件夾。文檔里寫錯

代碼片段和文件信息
from?os?import?listdir
import?os
import?random
import?io
image_folder_name=‘processed_images‘
path=os.getcwd()+‘/‘
images=listdir(path+‘{}/‘.format(image_folder_name))
random_list=[x.split(‘.‘)[0]?for?x?in?random.sample(imageslen(images))]
half=int(len(random_list)/2)
trainval=random_list[0:half]
test=random_list[half:]
train=trainval[0:int(half/2)]
val=trainval[int(half/2):]
with?io.open(‘trainval.txt‘‘w‘encoding=‘utf-8‘)?as?f:
????for?x?in?trainval:
????????f.write(unicode(x)+u‘\n‘)
with?io.open(‘test.txt‘‘w‘encoding=‘utf-8‘)?as?f:
????for?x?in?test:
????????f.write(unicode(x)+u‘\n‘)
with?io.open(‘train.txt‘‘w‘encoding=‘utf-8‘)?as?f:
????for?x?in?train:
????????f.write(unicode(x)+u‘\n‘)
with?io.open(‘val.txt‘‘w‘encoding=‘utf-8‘)?as?f:
????for?x?in?val:
????????f.write(unicode(x)+u‘\n‘)
?屬性????????????大小?????日期????時間???名稱
-----------?---------??----------?-----??----
?????目錄???????????0??2019-04-18?09:50??RefineDet搭建\
?????文件?????????782??2019-04-18?09:36??RefineDet搭建\create_data.sh
?????文件????????1312??2019-04-18?09:36??RefineDet搭建\create_list.sh
?????文件?????????837??2019-04-18?08:40??RefineDet搭建\generator.py
?????文件?????????139??2019-04-18?09:36??RefineDet搭建\labelmap_voc.prototxt
?????文件????????4587??2019-04-18?09:37??RefineDet搭建\refinedet_demo.py
?????文件???????18471??2019-04-18?09:50??RefineDet搭建\RefineDet框架搭建.docx
?????文件???????21694??2019-04-18?09:37??RefineDet搭建\VGG16_VOC2007_512.py
- 上一篇:vrPlus地址
- 下一篇:SAP視頻教程全集-70G
評論
共有 條評論