Add missing distribution cdf and ppf functions to scipy.stats - Githubissues

jax-ml / jax

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

http://jax.readthedocs.io/

Apache License 2.0

30.06k stars 2.75k forks source link

Add missing distribution cdf and ppf functions to scipy.stats #13291

Open carlosgmartin opened 1 year ago

carlosgmartin commented 1 year ago

Add the following functions to jax.scipy.stats (SciPy counterparts here):

[ ] bernoulli.cdf
[ ] bernoulli.ppf
[ ] beta.cdf
[ ] beta.ppf
[ ] betabinom.cdf
[ ] betabinom.ppf
[ ] cauchy.cdf
[ ] cauchy.ppf
[ ] chi2.cdf
[ ] chi2.ppf
[ ] expon.cdf
[ ] expon.ppf
[ ] gamma.cdf
[ ] gamma.ppf
[ ] gennorm.ppf
[ ] geom.cdf
[ ] geom.ppf
[ ] laplace.ppf
[ ] pareto.cdf
[ ] pareto.ppf
[ ] poisson.cdf
[ ] poisson.ppf
[ ] t.cdf
[ ] t.ppf
[ ] truncnorm.ppf
[ ] uniform.cdf
[ ] uniform.ppf

harryjulian commented 1 year ago

I'll submit a PR for the bernoulli functions this week, they're nearly done.

MatrixRanger98 commented 11 months ago

Please! This is useful. Is there any workaround right now?

IlayMenahem commented 8 months ago

@MatrixRanger98, for many of them there are analytical expressions