"SKEWED gamma like Non-negative pdf modeling= poor learner performance"
Hi
I have noticed that if a label attribute has a non negative highly skewed distribution the performance of learners such as Neural net is very poor compared to symmetrical (about zero) pdf problems.
Any way to tweak/bias a learner for such a distribution?
thx
I have noticed that if a label attribute has a non negative highly skewed distribution the performance of learners such as Neural net is very poor compared to symmetrical (about zero) pdf problems.
Any way to tweak/bias a learner for such a distribution?
thx
Find more posts tagged with
Sort by:
1 - 2 of
21
Hi Ingo
hmm
I am using nn for it and there is a parameter on the nn operator to normalize the data, it doesnt really help thoug...
hmm
I am using nn for it and there is a parameter on the nn operator to normalize the data, it doesnt really help thoug...
maybe it helps if you normalize the label before training and de-normalize the prediction after model application... I am not sure if this helps (not using NN a lot myself...)
Cheers,
Ingo