Neural network - target dataset

2020 3 7
1 个回答
回答已采纳
更新时间：2020 3 10
5 次查看（30 天）

0 个投票

I am doing project on sleep apnes (face recognition). I have to use artificial neural network for training and testing the dataset. For target dataset the values must be in 0s and 1s. how can i change my own target dataset into 0 and 1. Is there any code or something else from which i can get my correct target dataset to be trained. Pls help me with it.
I have attached my train and test data.
Thank you in advance

请先登录，再回答此问题。

 采纳的回答

michael scheinfeild
编辑：michael scheinfeild 2020-3-7

0 个投票

i suppose first you can normilize the train set and use it to normalize test set
meant=mean(train)
mint=min(train)
maxt=max(train)
data=(data-mint)/(mint-maxt)
this for features , same for outputs, i think you need regression network
https://ch.mathworks.com/help/matlab/ref/double.normalize.html
https://stackoverflow.com/questions/44147203/linear-regression-with-feature-normalization-matlab-code
https://www.researchgate.net/post/how_to_normalize_data_using_matlab_How_do_you_now_use_the_normalized_data_set_to_forecast_using_ann_on_matlab

1 个评论
显示 -1更早的评论 隐藏 -1更早的评论

vidhya v
vidhya v 2020-3-10
I apologize for the late reply sir. Thank you so much for your response sir. I will definitely try it.
Thank you again.

请先登录，再进行评论。

更多回答（0 个）

类别

在 帮助中心 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

Community Treasure Hunt

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

Start Hunting!

Translated by