-
Notifications
You must be signed in to change notification settings - Fork 191
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
请问一下,训练几个Epoch可以收敛啊? #11
Comments
我遇到acc一直是0 |
我也是,老铁你是怎么解决的 |
你后面解决了么?? |
如果是微调的话acc是0可以稍微调低一些初始学习率。
| |
婷婷
|
|
[email protected]
|
签名由网易邮箱大师定制
在2020年8月25日 16:58,lmw0320<[email protected]> 写道:
我遇到acc一直是0
你后面解决了么??
我也是ACC一直都是0,数据集是3W+的,不知道是不是数据集太少了。。
之前的情况是,用训练集来检测,acc大概训练个几十上百个epoch,就达到了99%,而拿另外的数据集测试的话,效果就很差。然后我将训练集进行了分割,在训练过程中检测验证集的效果,则acc训练个大几百个epoch,acc都一直是0.。。
不知道怎么回事。。
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub, or unsubscribe.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
按照 steps_per_epoch =145750// batch_size,
The text was updated successfully, but these errors were encountered: