Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

Thread Subject:
Test the trained network

Subject: Test the trained network

From: Livia Tengor

Date: 29 May, 2012 08:20:07

Message: 1 of 2

I want to make a stand alone application that contains two main features, they are Training and Recognizing. When the first input data have trained, I save the network. Then, I trained the network again with the second input data, and I save the network too. When I test the network with test data which is similiar to the first input data, the result is not same with the target of first data, it's like the target of the second data. help me please :'(

Subject: Test the trained network

From: Peter Bone

Date: 29 May, 2012 14:37:07

Message: 2 of 2

"Livia Tengor" wrote in message <jq20rn$h02$1@newscl01ah.mathworks.com>...
> I want to make a stand alone application that contains two main features, they are Training and Recognizing. When the first input data have trained, I save the network. Then, I trained the network again with the second input data, and I save the network too. When I test the network with test data which is similiar to the first input data, the result is not same with the target of first data, it's like the target of the second data. help me please :'(

Not sure you have explained your problem very well but I will try to help anyway. Are you talking about a neural network. What type? The weights of the network will update to the most recent data that it's been trained on. You must make sure that during training, your data is mixed up so that it's in no particular order. So interleave your 2 data sets into a single data set and it should get trained equally on both.

Tags for this Thread

No tags are associated with this thread.

What are tags?

A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.

Anyone can tag a thread. Tags are public and visible to everyone.

Contact us