v0.25.119
·
2 commits
to master
since this release
Distributions v0.25.119
Merged pull requests:
- Implement std(::MultivariateDistribution) (#1352) (@bicycle1885)
- BREAKING Fix loss of structure in cov() and invcov() of AbstractMvNormal subtypes (#1373) (@st--)
- add partype method to lognormal and semicircle (#1773) (@bgctw)
- Fix underflow of
logccdf
ofPGeneralizedGaussian
(#1932) (@devmotion) - Allow ForwardDiff v1 (#1957) (@mcabbott)
- Require Julia >= 1.10 (#1959) (@devmotion)
- Fix warnings in tests (#1960) (@devmotion)
- Remove docstring of non-existent method (#1962) (@abhro)
- Reduce allocations of
cov(::MultivariateMixture)
(#1967) (@devmotion) - Use generic
axpy!
instead of callingBLAS.axpy!
(#1968) (@jishnub)
Closed issues:
- why do cov and invcov call full? (#572)
- Underflow in logccdf for PGeneralizedGaussian (#1916)
- Matrix normal distribution support (#1946)
- May I suggest adding the sprand(::Distributions.Distribution, ...) method, like rand() does? (#1955)
cov(d::MultivariateMixture)
assumes overly specific typing on component covariance (#1969)- Random sample from LogNormal does not return values of the same type as its parameters (#1970)