faster-rcnn出错:confidence = np.array([float could not convert stringto loat:

Chinapeople1 2018-12-11 09:42:48
faster-rcnn训练出错:

saving cached annotations to /home/...../annots.pkl
Traceback(most recent call last):
File "./tools/test_net.py",line 90, in <module>
...
...
...
File "/home/zy/py-faster-rcnn/tools/../lib/datasets/voc_eval.py", line 142, in voc_eval
confidence = np.array([float(x[1]) for x in splitlines])
ValueError: could not convert string to float: (2)

[
...全文
173 1 打赏 收藏 转发到动态 举报
写回复
用AI写文章
1 条回复
切换为时间正序
请发表友善的回复…
发表回复
liliujun2009 2019-07-03
  • 打赏
  • 举报
回复
Hi, did you fix it by the following? basically edit tf-faster-rcnn/lib/datasets/voc_eval.py line121, change 'b' to 'wb' so it will become with open(cachefile,'wb') as f After you do that, it may throw you a new EOF ran out of input error.
You then need to edit same file line 105. Change it to cachefile = os.path.join(cachedir, '%s_annots.pkl' % imagesetfile.split("/")[-1].split(".")[0]) delete test.txt_annots.pkl

4,436

社区成员

发帖
与我相关
我的任务
社区描述
Linux/Unix社区 内核源代码研究区
社区管理员
  • 内核源代码研究区社区
加入社区
  • 近7日
  • 近30日
  • 至今
社区公告
暂无公告

试试用AI创作助手写篇文章吧