-
Notifications
You must be signed in to change notification settings - Fork 2.2k
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
验证码的accuracy似乎是单字符的accuracy吧 #1
Comments
而且我这边到了10000次以后仍然只有0.2不到的成功率 |
我跑了11个小时,得到了0.98的成功率
问题是:CPU 48线程只占用了30%不到。GPU也始终没有满载。可能程序有问题吧。
…On 3/22/2017 9:48 AM, Jcmels wrote:
@kcauchy <https://github.com/kcauchy>
其实。。我只想验证luyishisi是不是我认识的那个人。。。因为这代码跟我认识的人一毛一样,还有他博客里的tensorboard的截图。。。是我的。。。。成功率问题,你在每一层加一个Batch
Normalization,注意区分训练时跟测试时的代码。
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABT2_7n7ei9LIHzSM1GFOmF9UwlV3r1Vks5roH3fgaJpZM4Me35s>.
|
@kcauchy 单线程生成验证码图片,train的过程调用都是看TensorFlow自己去调用的,你加BN之后可以很快得到想要的结果。可能你CPU太强,我用imac的i5去train的,几乎满载 |
嗯。确实可能是生成验证码太慢。比较可以的做法是我试试用24线程去生成验证码,看看能不能喂满GPU。
2017/03/22 11:13、Jcmels <[email protected]> のメッセージ:
… @kcauchy 单线程生成验证码图片,train的过程调用都是看TensorFlow自己去调用的,你加BN之后可以很快得到想要的结果。可能你CPU太强,我用imac的i5去train的,几乎满载
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or mute the thread.
|
我跑的结果貌似很差,验证码生成为5个字符,34000 step 准确率能到0.96,但这个仅仅是训练准确率,然而进行测试时,准确率很低,感觉严重过拟合, |
Repository owner
deleted a comment from
JcmeLs
Mar 29, 2018
Repository owner
deleted a comment from
JcmeLs
Mar 29, 2018
Repository owner
deleted a comment from
BruceDone
Mar 29, 2018
leng-yue
pushed a commit
that referenced
this issue
Oct 1, 2018
这个程序有大量的问题 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
这里accuracy是单字符的话,整体的accuracy其实是会低于这个值的
The text was updated successfully, but these errors were encountered: