I get a "Performance function replaced with squared error performance" warning when trying to set 'crossentropy' as the performance function.
15 ビュー (過去 30 日間)
古いコメントを表示
If I run the following code:
[x,t] = house_dataset;
net = fitnet(10);
net.performFcn = 'crossentropy';
[net,tr] = train(net,x,t);
I get this warning:
Warning: Performance function replaced with squared error performance.
> In trainlm>formatNet (line 155)
In trainlm (line 65)
In nntraining.setup (line 14)
In network/train (line 335)
How can I use 'crossentropy' as the performance function then?
Regards
1 件のコメント
Greg Heath
2017 年 8 月 4 日
1. Where did you find house_dataset?
It does not come with MATLAB17a
2. For classification
help patternnet
doc patternnet
Hope this helps
Greg
回答 (4 件)
Greg Heath
2016 年 4 月 19 日
Crossentropy is, theoretically, not appropriate for regression.
Classically, it is only used for classification and pattern-recognition. It's definition involves probability distribution functions and their logarithms.
That is not to say that it will not yield good answers for regression problems. Obviously I have never tried it and, one day when I get bored, I might tinker around with it.
Hope this helps.
If you think this answer is worth accepting, THANKS!
Greg
0 件のコメント
Greg Heath
2017 年 8 月 5 日
編集済み: Greg Heath
2019 年 6 月 18 日
If you insist on using CROSSENTROPY, try PATTERNNET.
Hope this helps.
Thank you for formally accepting my answer
Greg
0 件のコメント
Alex
2016 年 4 月 7 日
It worked for me after adding
net.performParam.regularization = 0.1;
net.performParam.normalization = 'none';
Seems to be necessary when using cross entropy.
0 件のコメント
Chinmay Maheshwari
2017 年 8 月 2 日
Any updates on it as i am also facing the same trouble. I am not to customize my performance function because of this Thanks in advance
0 件のコメント
参考
カテゴリ
Help Center および File Exchange で Sequence and Numeric Feature Data Workflows についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!