Chaospy

==========================

ADDED:
* `include_axis_dim` flag added to `Distribution.sample` to force the
inclusion of extra dimension. (Currently first dimension is omitted is
`len(dist) == 1`.)
* Code of conduct and contribution descriptions in repo root.
* Tutorial for doing sequential polynomial chaos kriging.
CHANGED:
* `chaospy.E_cond` changed to accept simple polynomials as second argument,
allowing for e.g. `chaospy.E_cond(q0*q1, q0, dist)` which can be
interpreted as "expectation of `q0*q1` given `q0` with respect to `dist`".
* Full refactorization of the documentation.
* Updates `numpoly` to version 1.1.0. (some small breaking changes).
FIXED:
* Bugfixes to `chaospy.Spearman`
REMOVED:
* Deprecated `report_on_exception`. Caused recursion problems, and only a
semi-useful diagnostic tool to begin with.
* No more support for Python 3.5. This allows the poetry install to use
newer version of `numpy` and `scipy`. (This relates to poetry install, so
working in py35 might still be possible in practice.)

==========================

ADDED:
* `include_axis_dim` flag added to `Distribution.sample` to force the
inclusion of extra dimension. (Currently first dimension is omitted is
`len(dist) == 1`.)
CHANGED:
* `chaospy.E_cond` changed to accept simple polynomials as second argument,
allowing for e.g. `chaospy.E_cond(q0*q1, q0, dist)` which can be
interpreted as "expectation of `q0*q1` given `q0` with respect to `dist`".
* Bugfixes to `chaospy.Spearman`
* Updates to the documentation.
REMOVED:
* Deprecated `report_on_exception`. Caused recursion problems, and only a
semi-useful diagnostic tool to begin with.
* No more support for Python 3.5. This allows the poetry install to use
newer version of `numpy` and `scipy`.

==========================

Refactored `chaospy.quadrature.recurrence` -> `chaospy.recurrence`.

CHANGED:
* `chaospy.constructor` removed in favor for `chaospy.UserDistribution`.
* Bugfix: `chaospy.InverseGamma` moments needed to be reciprocal.
* Increased range on distributions: `StudentT`.
* Moved submodule `chaospy{.orthogonal->}.recurrence`.
* Stieltjes method get common interface `chaospy.stieltjes` which uses
analytical TTR if present, and approximation if not.
* Refactor `discretized_stieltjes` to be an iterative method with
tolerance criteria instead of brute forced. Also added max iterations and
scaling.
* Flag: Default `recurrence_algorithm` default changed to `stieltjes` (as
it covers both `analtical` and discretized Stieltjes).
* Discretization default in Lanczos and Stieltjes changed from `fejer` to
`clenshaw_curtis` as edge evaluation is better handled these days, and the
latter is better for when edges are finite.
REMOVED:
* `chaospy.basis` and `chaospy.prange` (which was superseded by
`chaospy.monomial` in June).
* Removal of "analytical" TTR where it is approximated: `Triangle`.
* `chaospy.chol` modules and the Cholesky functions: `bastos_ohagen`,
`gill_murry_wright` and `schnabel_eskow`. `gill_king` moved to
`chaospy.orthogonal.cholesky` as it is used by `orth_chol`.
* Flag: `accuracy` deprecated in favor for `tolerance`.

==========================

CHANGED:
* `lower > upper` illegal for all `LowerUpperDistribution` and `Trunc`.
* `scale <= 0` illegal for all `ShiftScaleDistribution`.
* Add epsilon buffer to all quadrature rules that evaluate at the edges.
* `numpoly` update to version 1.0.8.

==========================

Release!

ADDED:
* Gaussian Mixture Model: `GaussianMixture`.
* Tutorial for how to use `scikit-learn` mixture models to fit a model, and
`chaospy` to generate quasi-random samples and orthogonal polynomials.
CHANGED:
* `chaospy.Trunc` updated to take both `lower` and `upper` at the same time.
REMOVED:
* `chaospy.SampleDist` removed in favor of `chaospy.GaussianKDE`.

==============================

Additive recursion sampler.

ADDED:
* Support for additive recursive sampling scheme.
* Tutorial for Monte-Carlo now includes compare of difference sampling
schemes.
CHANGED:
* Bugfix to antithetic variate.