BicScore Question?

[mannou08]

Hey :)

Does the BicScore implemented in pgmpy minimize the negative log Likelihood function or maximize the
log Likelihood function? In other words: if I have two models with the scores -100 and -150, respectively.
Which model fits the data better?

Thanks in advance for your help :D

[ankurankan]

@mannou08 If I remember correctly it should be minimizing the negative log likelihood.