-
Notifications
You must be signed in to change notification settings - Fork 14
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
测试效果很差 #1
Comments
我测了下mjsynth的valid数据,大概准确率在0.9左右。这个代码是eager模式写的,训练和测试必须在同种设备上运行,另外记着改一下 os.environ["CUDA_VISIBLE_DEVICES"] = "1" 为自己的显卡号 |
主要是长序列问题:7,8个单词的那种,如果图中单词数目小于5个效果还行,在0.9左右,长一点的就不行,在训练时效果acc效果都很差。。。 |
不过长序列确实没用了, 单词数量大于2就基本全错了,只能用来检测单个单词,另外有下划线也会导致全错 |
你可以试一下把attention换成 self attention ,那样对长序列效果就会好很多,而且也不用用rnn单元了,模型也会小很多 |
ok, 🙏多谢。 |
先用水平投影做单词分割再识别呢?看看效果如何 |
建议这个加中文识别哦 |
File "E:\anaconda3\Anaconda3\envs\py_36\lib\site-packages\tensorflow\python\keras\engine\network.py", line 854, in _run_internal_graph 你好,我运行测试的时候报错,这个怎么改 |
eager保存模型不通用有点坑,CPU不是要跑炸 |
目前跑完2epochs,但是测试时的效果非常差,测了13张,模型输出全是错的。尤其是长序列,单词数目较多(大于2时),模型的输出非常差。
The text was updated successfully, but these errors were encountered: