Closed benjamin-lieser closed 1 month ago
I am quite happy with the code. After a review I would add the missing tests.
We might have to wait until a release of statrs
which can be configured to not depend on rand_distr
. Or we get the cdf implementations from elsewhere or have our own.
It is only a dev-dep, so the current solution (installing it from their master branch) might be fine
I got an interesting test failure on macos:
---- distr::bernoulli::test::bernoulli_distributions_can_be_compared stdout ----
thread 'distr::bernoulli::test::bernoulli_distributions_can_be_compared' panicked at src/distr/bernoulli.rs:229:9:
assertion `left == right` failed
left: Err(InvalidProbability)
right: Ok(Bernoulli { p_int: 18446744073709551615 })
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
Could be a compiler bug, maybe?
Maybe someone with a mac can reproduce this?
It passed when re-run.
CHANGELOG.md
entrySummary
As addition and alternative to pdf testing, I implemented distribution testing based on cdf
Motivation
The cdf tests should be more robust and avoid having the bin-width choice problem
Details
I have two simple example tests for normal and binomial. If we like these tests we should also add some way to test on different parameter sets and the other distributions.
This can't be easily generalized to multivariate ones.