Feature request: vectorized phi() #3383

Activity

1. ksvanhorn

   added

   feature

   on Jul 6, 2023
2. bob-carpenter commented on Jul 6, 2023

   

   bob-carpenter

   on Jul 6, 2023

   Member

   More actions

   We do have a vectorized erf(), so a standard normal is just a factor of sqrt(2) away. Did you want the same signatures as we provide for erf()? That'd be

   vector phi(vector);
   row_vector phi(row_vector);
   matrix phi(matrix);
   real[] phi(real[]);
   ...

   There are other feature requests asking us to supply vectorized versions of all of our probability functions. They're useful for things like log likelihoods for leave-one-out cross-validation.
3. rok-cesnovar commented on Jul 7, 2023

   

   rok-cesnovar

   on Jul 7, 2023

   Last edited by rok-cesnovar

   Member

   More actions

   The Phi function supports the following signatures in the latest few versions:

   Phi(int) => real
   Phi(real) => real
   Phi(vector) => vector
   Phi(row_vector) => row_vector
   Phi(matrix) => matrix
   Phi(array[] int) => array[] real
   Phi(array[] real) => array[] real
   Phi(array[] vector) => array[] vector
   Phi(array[] row_vector) => array[] row_vector
   Phi(array[] matrix) => array[] matrix
   Phi(array[,] int) => array[,] real
   Phi(array[,] real) => array[,] real
   Phi(array[,] vector) => array[,] vector
   Phi(array[,] row_vector) => array[,] row_vector
   Phi(array[,] matrix) => array[,] matrix
   Phi(array[,,] int) => array[,,] real
   Phi(array[,,] real) => array[,,] real
   Phi(array[,,] vector) => array[,,] vector
   Phi(array[,,] row_vector) => array[,,] row_vector
   Phi(array[,,] matrix) => array[,,] matrix
   Phi(array[,,,] int) => array[,,,] real
   Phi(array[,,,] real) => array[,,,] real
   Phi(array[,,,] vector) => array[,,,] vector
   Phi(array[,,,] row_vector) => array[,,,] row_vector
   Phi(array[,,,] matrix) => array[,,,] matrix
   Phi(array[,,,,] int) => array[,,,,] real
   Phi(array[,,,,] real) => array[,,,,] real
   Phi(array[,,,,] vector) => array[,,,,] vector
   Phi(array[,,,,] row_vector) => array[,,,,] row_vector
   Phi(array[,,,,] matrix) => array[,,,,] matrix
   Phi(array[,,,,,] int) => array[,,,,,] real
   Phi(array[,,,,,] real) => array[,,,,,] real
   Phi(array[,,,,,] vector) => array[,,,,,] vector
   Phi(array[,,,,,] row_vector) => array[,,,,,] row_vector
   Phi(array[,,,,,] matrix) => array[,,,,,] matrix
   Phi(array[,,,,,,] int) => array[,,,,,,] real
   Phi(array[,,,,,,] real) => array[,,,,,,] real
   Phi(array[,,,,,,] vector) => array[,,,,,,] vector
   Phi(array[,,,,,,] row_vector) => array[,,,,,,] row_vector
   Phi(array[,,,,,,] matrix) => array[,,,,,,] matrix
4. bob-carpenter commented on Jul 7, 2023

   

   bob-carpenter

   on Jul 7, 2023

   Member

   More actions

   Thanks @rok-cesnovar. This is the list you get from using the automatic vectorization of a unary function, which will be an easy way to implement this. Better to provide an analytic gradient for reverse mode. Callingstd_normal_lpdf function to implement unary standard normal is probably too heavy.
5. WardBrian

   transferred this issue fromstan-dev/stanon Sep 1, 2026
6. WardBrian

   added

   good first issue

   distributionsIssues that deal with distribution functions: pdf, pmf, cdfIssues that deal with distribution functions: pdf, pmf, cdf

   on Sep 1, 2026
7. SteveBronder commented on Sep 1, 2026

   

   SteveBronder

   on Sep 1, 2026

   Collaborator

   More actions

   Related, I have #2751 which vectorizes the normal_lpdf in a way that allows for a scalar or vector return.

   One thing we could do is, if we can agree on the form/code for the normal_lpdf in that branch then this is a good thing to have a swarm of claude's apply through the math library.