Fascinating World of Neural Nets  
Neural Network Forums
Home PageHome Page : Neural Networks for Pattern Recognition : Finding a pattern in data
  You are currently not logged in. You can view the forums, but cannot post messages. | HOME | Log In | Register | Search | Help
Post a Reply on This Topic Post a Reply on This Topic

Author Topic: Finding a pattern in data
morghul Posted: 16-Feb-11 07:20
  Edit Edit
Email the Author Mail   View Author's Profile Profile  
Hey all,
I'm working on program that finds a pattern in data. What I want to get is ilustrated on this picture.
0's if the point is not a part of the pattern, 1's if it is.
For learning process I have generated 100 sets of 500 points - only Y values of the function (first set is from 3.0 to 7.0, 2nd from 3.05 to 7.05, 3rd 3.1, 7.1, etc.) as input data, and a vectors of 0's and 1s as target vectors (0-point is not in the pattern, 1-point is part of the pattern). My nural network is 500 neurons in input layer (1 for each data point), 35 neurons in hidden, 500 neurons in output layer.
And it doesn't work properly. Is it possible to do it that way? What am I doing wrong?

Post a Reply on This Topic Post a Reply on This Topic

Copyright © 2001-2003 Pejman Makhfi. All rights Reserved.