add python version evaluate function #1036

dayhaha · 2016-12-29T09:16:23Z

fix #1035
As evaluate.sh don't work in Mac, I add a python version evaluate function. The output is the same as evaluete.sh

➜  src git:(develop) ✗ ./evaluate.py mlp_train.log
Best pass is 00085, error is 0.164746, which means predict get error as 0.202945
The classification accuracy is 97.44%
evaluating from pass output/pass-00085

wangkuiyi · 2016-12-29T13:03:55Z

demo/recommendation/evaluate.py

+import math
+
+
+def get_best_pass(filename):


filname ==> log_filename 这样更具体一些，让人知道这个函数的输入是什么

https://google.github.io/styleguide/pyguide.html?showone=Naming#Naming

done，谢谢指出!

…t_preview (PaddlePaddle#1036) Done

dayhaha added 2 commits December 29, 2016 09:12

add python version evaluate function

7e769ac

change LOG to log

381892b

wangkuiyi requested changes Dec 29, 2016

View reviewed changes

modify follow comment

3f3a9ff

wangkuiyi approved these changes Jan 1, 2017

View reviewed changes

dayhaha merged commit 495649a into PaddlePaddle:develop Jan 2, 2017

zhhsplendid pushed a commit to zhhsplendid/Paddle that referenced this pull request Sep 25, 2019

fix the CUDA support description in indtall_windows DOC, test=documen…

56914e0

…t_preview (PaddlePaddle#1036) Done

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

add python version evaluate function #1036

add python version evaluate function #1036

dayhaha commented Dec 29, 2016

wangkuiyi Dec 29, 2016

dayhaha Dec 29, 2016

		import math


		def get_best_pass(filename):

add python version evaluate function #1036

add python version evaluate function #1036

Conversation

dayhaha commented Dec 29, 2016

wangkuiyi Dec 29, 2016

Choose a reason for hiding this comment

dayhaha Dec 29, 2016

Choose a reason for hiding this comment