QuantEcon/QuantEcon.jl

Use `StableRNGs` in tests

Open

#275 opened on Jul 5, 2022

View on GitHub
 (0 comments) (1 reaction) (0 assignees)Julia (495 stars) (300 forks)batch import
enhancementhelp wantedtest

Description

@oyamad asked me to expand my comment in #274 into an issue. Here it is.

A lot of tests currently depend on outcomes of simulations. In order to get reproducible results, the tests set seeds. However the random number streams are not guaranteed to be the same across Julia versions. This led to the problem visible in #274: a test suddenly broke on Julia 1.7 and above.

In order to avoid this one can use the package https://github.com/JuliaRandom/StableRNGs.jl, which guarantees reproducible streams of random numbers across Julia versions.

I am afraid that this would require rewriting some of the functionality of this package to accept an RNG as in

rng = StableRNG(seed::Integer) # random number generator object
rand(rng, X)

Contributor guide

Tech stack
None
Domain
testing
Issue type
test
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
Julia basicsFamiliarity with RNG in JuliaUnderstanding of test setup
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
65
Research direction
First, review the test files in the repository (likely under a `test/` directory) to find instances where `rand` is called with a seed. Then, import the `StableRNGs` package and modify each test to create a `StableRNG` object and pass it to `rand`. Ensure that the changes do not alter the expected test outcomes. Check the linked issue #274 for context on the Julia version compatibility issue.