Why do Predict and Probability functions use different operators?

[wagslane]

I understand why the Naive Bayes "Predict" function uses a math.Log() to avoid an underflow. I don't understand why on lines 288 and 293 the operator is += instead of *=... Could you provide an explanation? Maybe an update to the docs?

[cdipaolo]

Hi Lane,

Thanks for the important question:

In the model we would like to compute the product p1 * p2 * ... * pn of a bunch of probabilities. As you know, this has serious underflow problems, so instead we compute log(p1 * p2 * ... * pn) to get around this. But based on the basic log-property that log(a*b) = log(a) + log(b), this says that we can just compute log(p1 * p2 * ... * pn) = log(p1) + log(p2) + ... * log(pn), which is exactly what we do.

Does that help?

[wagslane]

Yes! Thanks

…

On Fri, Apr 12, 2019 at 12:01 PM Conner DiPaolo ***@***.***> wrote: Hi Lane, Thanks for the important question: In the model we would like to compute the product p1 * p2 * ... * pn of a bunch of probabilities. As you know, this has serious underflow problems, so instead we compute log(p1 * p2 * ... * pn) to get around this. But based on the basic log-property that log(a*b) = log(a) + log(b), this says that we can just compute log(p1 * p2 * ... * pn) = log(p1) + log(p2) + ... * log(pn), which is exactly what we do. Does that help? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#19 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AU-VVWsVAd4xnH0NKVHp_728CcnKdff-ks5vgMn4gaJpZM4csaqi> .