I’m relatively new to PyMC3, trying to setup a synthetic IRT model. The below code does sample, decent rhat, very low effective samples: k = np.array([1,1,1,1,1, 1,1,1,1,0, 1,1,1,0,0, 1,1,0,0,0, 1,0,0,0,0]).reshape(5,5) students = 5 questions…

As for the first question, the shape of theta is 5 because both z_student and z_question have shape of (5,). To get a value for each combination, you need to either instantiate variables with a shape of (25,) (which I think is probably not what you want) or you need to instantiate them with shapes t…

IRT w/ PyMC3, unusual number of parameters modeled?

Questions

jbuddy_13 November 19, 2020, 2:30pm 5

Oh, awesome- thank you! Btw, since you seem to be quite familiar, you may have thoughts on an extension of this topic:

Topic		Replies	Views
Is this an appropriate way to model IRT in PyMC3 Questions theano , doc	2	709	May 3, 2021
Bayesian IRT Pymc3 - Parameter inference Questions	1	1392	April 27, 2020
Trying to use pymc3 for predictive MC-Simulation [SOLVED] Questions	1	493	July 18, 2018
Trouble putting together a regression model in pymc3 Questions	4	678	June 10, 2019
Must be real number, not FreeRV Questions	2	493	October 22, 2021

IRT w/ PyMC3, unusual number of parameters modeled?

Related topics