Distributions.jl
Distributions.jl copied to clipboard
JuliaStats
Native RNGs
Targeted at #294. Removes Rmath dependence of RNGs for Noncentral distributions.
Codecov Report
Patch coverage: 100.00% and project coverage change: -0.01 :warning:
Comparison is base (
630e1c9) 85.82% compared to head (8b816f2) 85.81%.
:exclamation: Current head 8b816f2 differs from pull request most recent head a5fa5e2. Consider uploading reports for the commit a5fa5e2 to get more accurate results
Additional details and impacted files
@@ Coverage Diff @@
## master #1698 +/- ##
==========================================
- Coverage 85.82% 85.81% -0.01%
==========================================
Files 137 137
Lines 8315 8311 -4
==========================================
- Hits 7136 7132 -4
Misses 1179 1179
| Impacted Files | Coverage Δ | |
|---|---|---|
| src/univariate/continuous/noncentralbeta.jl | 100.00% <ø> (ø) |
|
| src/univariate/continuous/noncentralf.jl | 90.00% <ø> (-1.67%) |
:arrow_down: |
| src/univariate/continuous/noncentralchisq.jl | 90.00% <100.00%> (+2.00%) |
:arrow_up: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.
Thank you for the PR! I made some suggestions below. We should also make sure that everything is covered by the tests. In principle, the standard tests for univariate distributions include tests for sampling but maybe the distributions with Rmath-based sampling were explicitly excluded.
Not really sure on how to test the RNGs 😓
What's left in the PR @ArunSanganal?
It should be good to merge if the tests added are satisfactory.
It should be good to merge if the tests added are satisfactory.
I improved the tests. They were a bit redundant and even a bit worse than some existing tests, it seems. I increased the number of samples used in the standard tests as well.
It says, counts are out of Confidence Interval. Where are we getting the values of Distribution parameters from? I am not able to reproduce the same for custom values.
It says,
counts are out of Confidence Interval. Where are we getting the values of Distribution parameters from? I am not able to reproduce the same for custom values.
The RNG isn't working, in that case. You should check your implementation carefully for any bugs, e.g. using the wrong parameters or parameter name.
In addition, you can try increasing the number of samples and changing the seed for the RNG to see if it might just be bad luck.
