multidimensional array and deep learning

2018 5 10
1 个回答
回答已采纳
更新时间：2018 5 11
5 次查看（30 天）

0 个投票

Hi,
I am trying to create a simple deep learning network but it 'appears' that they are not compatible with multidimensional arrays. I have a data-set of (30,30,97000) of positive example and (30,30,200000) negative examples. the names and pos/neg tags are in separate arrays. Need some direction on how to get the data into the prop format for deep learning.
Do i need to write a loop and save all of the files individually so i can build a data store afterwards? seems time inefficient...
quick mock-up example
good=rand(30,30,100); bad=rand(30,30,200);
g_label=(1:1:100); g_label(2,:)=1;
b_label=(101:1:300); b_label(2,:)=1;

请先登录，再回答此问题。

 采纳的回答

Jason Dortch
Jason Dortch 2018-5-11

2 个投票

I obtained help from tech support. The Deep learning tools will take a 4-dimensional array. using the permute function, I added one dimension so my data is arranged like a grayscale image - (width, height, channel, number of images).
Good=rand(30,30,100);
Good=permute(Good, [1 2 4 3]);
Size(Good)
30 30 1 100
Hope that helps someone, Jason

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