Skip to content
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

paddle_trainer.INFO里auc的计算机制? #2443

Closed
wangsouc opened this issue Jun 12, 2017 · 2 comments
Closed

paddle_trainer.INFO里auc的计算机制? #2443

wangsouc opened this issue Jun 12, 2017 · 2 comments
Assignees

Comments

@wangsouc
Copy link

MPI cluster训练,请教一下__auc_evaluator_0__的Eval和CurrentEval的计算机制。
I0612 15:12:50.297226 18323 TrainerInternal.cpp:165] Batch=2760 samples=11304960 AvgCost=0.113212 CurrentCost=0.109198 Eval: auc_evaluator_0=0.7042 classification_error_evaluator=0.02633 CurrentEval: auc_evaluator_0=0.727945 classification_error_evaluator=0.0255005

@jacquesqiao
Copy link
Member

训练逻辑主要在TrainerInternal::trainOneBatch:

1,

先调用forwardBackwardBatch,计算出当前batch的gradient,然后调用callback 更新了参数。

2,

然后调用了eval计算evaluator的值。

3,

之后才是显示结果

结论:

先更新了参数,然后进行evaluat。

@jacquesqiao jacquesqiao self-assigned this Jun 12, 2017
@lcy-seso
Copy link
Contributor

lcy-seso commented Jun 12, 2017

这个是问 Eval 和 CurrentEval 这两个指标的区别吗?如果是的话:

  • CurrentEval 是在当前Batch上的评估结果
  • Eval 是到当前batch 为止,评估结果的平均值

因此,通常 current eval 抖动更强,在第一个 pass 的时候,current eval 评估效果会更好,eval 看上去更差一些。

heavengate pushed a commit to heavengate/Paddle that referenced this issue Aug 16, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants