After test the input in Neural Network with sim command, how to get the original scale without using real outputs?

1 次查看（过去 30 天）
Sertan Kaya
Sertan Kaya 2011-4-21
Training has been done already, and "net" exists for testing of Neural Network. Given an inputs_test matrix, I scale the inputs_test matrix such a way that "sim" accepts it for testing process as follows.
[transf_inputs,transf_coff] = mapstd(inputs_test); Y = sim(net,transf_inputs);
And now, I want to scale Y back to original format what I have at the beginning. I can do it as follows.
[transf_targets,transf_coff2]= mapstd(outputs); y=mapstd('reverse',Y,transform_coff2);
However, this process requires me to have real outputs matrix instead Y (NN predicted values). I tried transf_coff reverse format process variable (which was obtained before the "sim" command), but matrix dimensions are different for inputs and outputs. In old functions (prestd/poststd and premnmx/postmnmx), it let you do this thing. Is there any way in which I can scale Y back to original format without having real outputs matrix ? Thanks for answers

请先登录，再回答此问题。

回答（1 个）

Sertan Kaya
Sertan Kaya 2011-4-25
Thanks for not responding though !
Not scaling input and target vectors solved my problem.
  2 个评论
Walter Roberson
Walter Roberson 2011-4-25
Relatively few of the people who volunteer to answer questions here have experience with Neural Nets. You also happened to post on a holiday weekend in many parts of the world.
I know that I do not have NN experience, and I do not have the appropriate toolbox to test things out with. I do not have the knowledge such that the question might have been meaningful enough for me such that I might have hunted through the documentation.
Sometimes lack of a response just means that none of the volunteers who happens to know the answer happened to read the question. If you have urgent questions, then it is better to call Mathworks Technical Support.

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by