Hi all! I’m working on a model with an ordered categorical predictor (E below). So, the associated parameter is incrementally added to the linear model: when E==0, delta_e = 0; when E==1, delta_e = 0 + delta_e1; when E==2, delta_e = 0 + delta_e1 + delta_e2, etc. In code: with pm.Model() as m: …

Would this work? delta = pm.Dirichlet("delta", np.repeat(2., 7), shape=7) delta_j = tt.concatenate([tt.zeros(1), delta]) delta_j_cumulative = tt.cumsum(delta_j) phi = bE * delta_j_cumulative[E+1] + bA * A

How to vectorize indexing of variable for pm.math.sum?

Questions

AlexAndorra April 13, 2020, 2:01pm 5

Oh, you’re right, that’s the meaning of Junpeng’s answer
That works, thanks Nicholas

Topic		Replies	Views
Pymc3 vector concatenation Questions	6	4365	August 18, 2019
What is the best way to vectorize an operation on tensor in PyMC? Questions	3	911	February 16, 2022
Making a vector of variables from a list Questions	5	1258	December 28, 2017
Creating Theano Op to work on vectors element-wise within Model()? Questions	5	2468	August 30, 2017
Problem with categorical index variable in v5 v5	1	306	January 3, 2024

How to vectorize indexing of variable for pm.math.sum?

Related topics