Thread Subject: trainlm need an RMS Log error function for training

Subject: trainlm need an RMS Log error function for training

From: Dr. W. Kurt Dobson

Date: 24 Jul, 2011 23:10:09

Message: 1 of 1

Reply to this message
Add author to My Watch List
View original format
Flag as spam

I have a nnet problem I want to run where I need RMSLE as an error function, i.e.

e = sqrt(1/n * sum[ log(pi+1) - log(ai+1) ]^2)

Is there any way to define my own error function and have any of the training functions (trainlm, etc) to call this instead of mse(), sse(), etc.?

Kurt
WorldHeart

Tags for this Thread

Everyone's Tags:

trainlm, mse

Add a New Tag:

Separated by commas
Ex.: root locus, bode

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.

Tag Activity for This Thread
Tag Applied By Date/Time
trainlm Dr. W. Kurt Dobson 24 Jul, 2011 19:14:10
mse Dr. W. Kurt Dobson 24 Jul, 2011 19:14:10
rssFeed for this Thread

Contact us at files@mathworks.com