Computes the log-predictive probability for positive exponential noise
[allLog] = logp_exp(likpar,y,mX,sigX2);
[allLog] = logp_exp(likpar,y,mX,sigX2)
- returns the sum of
log-predictive probabilities of the inputs -- assuming additive positive
exponential noise. The samples are treated independently.
Parameters:
likpar
- the likelihood parameters.
y
- desired output.
mx
- mean of the GP marginal at the new input.
allLog
- the average error.
ogp
, ogptrain
, demogp_reg
, err_abs
Copyright (c) Lehel Csató (2001-2004)