How to create Neural Network classifier for pattern Recognition ?

2 views (last 30 days)
I want to create a Neural network with 50 thousand nodes for EEG recognition . Can anybody help me to do this?

Accepted Answer

Greg Heath
Greg Heath on 2 Oct 2013
How many classes/categories? c = ?
What is the dimensionality of your input vectors? I= ?
How many input/target examples do you have? N = ?
Choose the output target vectors to be O = c dimensional unit column vectors with a 1 in the row corresponding to the class index.
[ I N ] = size(input) % = ?
[ O N ] = size(target) % = ?
Using defaults,
Ntrn = N - 2*round(0.15*N) % No. of training examples
Ntrneq = Ntrn*O % No. of training equations
H = 10 % No. of hidden layer nodes
Nw = (I+1)*H+(H+1)*O % No of unknown weights
Is Ntrneq > Nw ?
Is Ntrneq >> Nw ?
Use the help and doc commands and try one or more of the following practice examples
simpleclass_dataset - Simple pattern recognition dataset.
cancer_dataset - Breast cancer dataset.
crab_dataset - Crab gender dataset.
glass_dataset - Glass chemical dataset.
iris_dataset - Iris flower dataset.
ovarian_dataset - Ovarian cancer dataset.
thyroid_dataset - Thyroid function dataset.
wine_dataset - Italian wines dataset.
Search for more info and examples in the NEWSGROUP and ANSWERS using
greg patternnet
Hope this helps.
Thank you for formally accepting my answer
Greg
  1 Comment
moahaimen talib
moahaimen talib on 6 May 2017
dear sir i need your help i need some explaining of using nn for character recognition i have 20 images to read and i must recognize 20 character of them could please help me

Sign in to comment.

More Answers (1)

Shashank Prasanna
Shashank Prasanna on 30 Sep 2013
Here is a walk through example of how to do pattern classification using the Neural Network Toolbox:
Feel free to go through the other examples in the documentation as well.

Community Treasure Hunt

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

Start Hunting!