Hello! I am an engineering student currently learning bayesian stats. I want to model a time series of proportions (values between 0 and 1), using the Gaussian process. I find that there seems to be 2 (inverse) link functions that I can use: pm.math.sigmoid() pm.math.invlogit() I tried both and …

Since you do not have a large dataset, these two functions should be the same. Going forward, sigmoid() is implemented in theano.nn and invlogit() is implemented in PyMC (via theano tensor ops). They are the same function (up to machine epsilon).

Invlogit() vs. sigmoid() - which one to use?

Questions

jdolphin7 April 4, 2022, 4:04pm 3

Hey there, I’m wondering which of the two ought to be faster if you do have a large dataset?

Topic		Replies	Views
When to use `pm.math.sigmoid` and `pm.math.invlogit` for Logistic Regression? Questions modeling	3	761	December 29, 2022
Comparing Theano values Questions theano	13	2483	July 20, 2017
ADVI Is Much Slower than that Showed in the tutorial case Questions development	4	1179	October 17, 2018
Poor Accuracy of ADVI for Linear Regression Questions	12	3457	April 18, 2018
How to implement a GLM with a personalized link function different from the Identity function? Questions	7	1877	June 7, 2021

Invlogit() vs. sigmoid() - which one to use?

Related topics