How I can fuzzify two dimension matrix, then train it using back propagation neural network, and defuzzify the output ?

1 次查看（过去 30 天）
nada fady
nada fady 2016-8-11
Hi, I am working on speech processing, I used MFCC to extract features. now I have a matrix of size 70*5 for each sample this represent the input data.I reshaped the matrix to get vector of size 350, as I have 50 sample so the the input matrix became 350*50.
and I have matrix of size 70*11 , as a target or desired output. I also reshaped it to be vector of size 770, then the desired matrix became 770*50.
that mean I have an input matrix of size 350*50, and an output or desired matrix of size 770*50. I saved these matrices in two excel files.
Now I wont to fuzzify these values and use neural network to train the input data using back propagation algorithm then defuzzify the output of the neural network.
I read about ANFIS function , but I still confused about the large size of input and target matrix that I have. And I do not now how to use this instruction.
can any one tell an example code to use the instruction with large data?
I would like to mention that I read the help of anfis instruction but still can not use it with my large size data.
Thank you in advance.

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Fuzzy Logic 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