Prepare release 2.3 (#500)

* Add contributor

* Fix SGP doc

* Bump 2.3.0

AUTHORS.md

@@ -11,6 +11,7 @@ SMT is developed by:
 SMT has been developed thanks to contributions from:
 * Alexandre Thouvenot
 * Andres Lopez Lopera
+* Antoine Averland
 * Emile Roux
 * Ewout ter Hoeven
 * Florent Vergnes

doc/_src_docs/surrogate_models/sgp.rstx

@@ -20,7 +20,7 @@ Implementation
 In SMT the methods: Fully Independent Training Conditional (FITC) method and the Variational Free Energy (VFE) approximation
 are implemented inspired from inference methods developed in the GPy project [2]_
 
-In practice, the implementation rely on the expression of their respective negative marginal log
+In practice, the implementation relies on the expression of their respective negative marginal log
 likelihood (NMLL), which is minimised to train the methods. We have the following expressions:
 
 For FITC
@@ -47,7 +47,7 @@ and :math:`\eta^2` is the variance of the gaussian noise assumed on training dat
 Limitations
 -----------
 
-* Inducing points location can not be optimized.
+* Inducing points location can not be optimized (a workaround is to provide inducing points as the centroids of k-means clusters over the training data).
 * Trend function is assumed to be zero.
 
 

setup.py

@@ -22,7 +22,7 @@
 Programming Language :: C++
 Programming Language :: Python
 Programming Language :: Python :: 3
-Programming Language :: Python :: 3.7
+Programming Language :: Python :: 3.8
 Programming Language :: Python :: Implementation :: CPython
 Topic :: Software Development
 Topic :: Scientific/Engineering
@@ -43,6 +43,9 @@
 with respect to the training data. It also includes new surrogate models 
 that are not available elsewhere: kriging by partial-least squares reduction 
 and energy-minimizing spline interpolation.
+
+SMT 2.0 adds the capability to handle mixed-variable surrogate models 
+and hierarchical variables.
 """
 
 extra_compile_args = []

smt/__init__.py

@@ -1 +1 @@
-__version__ = "2.2.1"
+__version__ = "2.3.0"