wtte-rnn icon indicating copy to clipboard operation
wtte-rnn copied to clipboard
verified publisher icon ragulpr

Loss Function - Not the PCF?

Open ghost opened this issue 6 years ago • 2 comments

Hello, when looking at the likelihood function for the Weibull, I derived a different function than you do. Here is your function:

loglikelihoods = censored * (K.log(shape) + shape * K.log(x)) - K.pow(x, shape)

However when I do this, I end up with a K.log(scale) term as well. Can you explain how you arrived at your loglikelihood function? I can provide my working if necessary

ghost avatar Apr 18 '19 09:04 ghost

Totally agree, made a pull request

michael-tsel avatar May 09 '19 07:05 michael-tsel

Hi there, thanks for contributing! I added some comments on PR https://github.com/ragulpr/wtte-rnn/pull/59

Otherwise I recommend reading

Proposition 2.26 and chapter 3.1.1 and 3.2 in particular shows some alternative forms of the loss function. http://publications.lib.chalmers.se/records/fulltext/253611/253611.pdf

Is there any qualitative difference to the current form? I'd be happy to see how you arrived at your definition

ragulpr avatar Oct 13 '19 16:10 ragulpr