Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

help plzzz matlab about nearul network

Status
Not open for further replies.

ozzzz

Newbie level 6
Joined
Aug 27, 2010
Messages
14
Helped
4
Reputation
8
Reaction score
4
Trophy points
1,283
Activity points
1,358
i m doing my project about face recording using neural network . i found 20 feature from face and i have 20 value .but i dont do neural network. i use newff but always give me different performans value. why????? and performans value is very very low:(:(:( between 25 and 30 why
i did training data , target and testing data.
training data= 20x352, target=1x352 testing data=20x80
help me plzzzz
 

any idea??? why dose it give me different output value ??? may be bias and weight . i dont know:(:(:( how does i do bias ane weight??? plzzz help mee
 

if i remember this is all gray scale node map nural stuff

so each node will arrive different each time you load it

this is normal as it must do many calculations too see events move...

bias is light level on face
weight is face node value

start with a blank 100% gray level bmp

and work in gray levels from a known card

are you using a jpg bmp or a camera etc ..???

your test setup info is sparce...
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top