Merge pull request #362 from lmfit/prep095

Prep095

doc/index.rst

@@ -6,16 +6,6 @@ Non-Linear Least-Square Minimization and Curve-Fitting for Python
 .. _Levenberg-Marquardt:     http://en.wikipedia.org/wiki/Levenberg-Marquardt_algorithm
 .. _MINPACK-1:               http://en.wikipedia.org/wiki/MINPACK
 
-
-.. warning::
-
-  Upgrading scripts from version 0.8.3 to 0.9.0?  See  :ref:`whatsnew_090_label`
-
-.. warning::
-
-  Support for Python 2.6 and scipy 0.13 will be dropped with version 0.9.5.
-
-
 Lmfit provides a high-level interface to non-linear optimization and curve
 fitting problems for Python. Lmfit builds on and extends many of the
 optimization algorithm of :mod:`scipy.optimize`, especially the
@@ -51,8 +41,8 @@ fitting problems, including:
 
 .. _lmfit github repository:   http://github.com/lmfit/lmfit-py
 
-The lmfit package is Free software, using an MIT license.  The software and
-this document are works in progress.  If you are interested in
+The lmfit package is Free software, using an Open Source license.  The
+software and this document are works in progress.  If you are interested in
 participating in this effort please use the `lmfit github repository`_.
 
 
@@ -70,3 +60,4 @@ participating in this effort please use the `lmfit github repository`_.
    confidence
    bounds
    constraints
+   whatsnew

doc/installation.rst

@@ -12,12 +12,10 @@ Prerequisites
 
 The lmfit package requires Python, Numpy, and Scipy.
 
-Lmfit works with Python 2.7, 3.3, 3.4, and 3.5. Lmfit version 0.9.4 works
-with Python 2.6, but support for it will dropped in version 0.9.5.  Scipy
-version 0.13 or higher is required, with 0.17 or higher recommended to be
-able to use the latest optimization features from scipy.  Support for scipy
-0.13 will be dropped in version 0.9.5.  Numpy version 1.5 or higher is
-required.
+Lmfit works with Python 2.7, 3.3, 3.4, and 3.5. Support for Python 2.6
+ended with Lmfit version 0.9.4.  Scipy version 0.14 or higher is required,
+with 0.17 or higher recommended to be able to use the latest optimization
+features from scipy.  Numpy version 1.5 or higher is required.
 
 In order to run the test suite, the `nose`_ framework is required.  Some
 parts of lmfit will be able to make use of IPython (version 4 or higher),

doc/whatsnew.rst

@@ -11,6 +11,45 @@ changes to the use and behavior of the library.  This is not meant to be a
 comprehensive list of changes.  For such a complete record, consult the
 `lmfit github repository`_.
 
+.. _whatsnew_095_label:
+
+Version 0.9.5 Release Notes
+==========================================
+
+Support for Python 2.6 and scipy 0.13 has been dropped.
+
+.. _whatsnew_094_label:
+
+Version 0.9.4 Release Notes
+==========================================
+
+Some support for the new `least_squares` routine from scipy 0.17 has been
+added.
+
+
+Parameters can now be used directly in floating point or array expressions,
+so that the Parameter value does not need `sigma = params['sigma'].value`.
+The older, explicit usage still works, but the docs, samples, and tests
+have been updated to use the simpler usage.
+
+Support for Python 2.6 and scipy 0.13 is now explicitly deprecated and wil
+be dropped in version 0.9.5.
+
+.. _whatsnew_093_label:
+
+Version 0.9.3 Release Notes
+==========================================
+
+Models involving complex numbers have been improved.
+
+The `emcee` module can now be used for uncertainty estimation.
+
+Many bug fixes, and an important fix for performance slowdown on getting
+parameter values.
+
+ASV benchmarking code added.
+
+
 .. _whatsnew_090_label:
 
 Version 0.9.0 Release Notes

lmfit/__init__.py

@@ -25,9 +25,9 @@
 
   * Many pre-built models for common lineshapes are included and ready to use.
 
-   version: 0.9.4
-   last update: 2016-Jul-1
-   License: MIT
+   version: 0.9.5
+   last update: 2016-Jul-26
+   License: BSD
    Authors:  Matthew Newville, The University of Chicago
              Till Stensitzki, Freie Universitat Berlin
              Daniel B. Allen, Johns Hopkins University
@@ -57,11 +57,11 @@
 
 # PY26 Depreciation Warning
 if sys.version_info[:2] == (2, 6):
-    warnings.warn('Support for Python 2.6.x  will be dropped in lmfit 0.9.5')
+    warnings.warn('Support for Python 2.6.x was dropped with lmfit 0.9.5')
 
 # SCIPY 0.13 Depreciation Warning
 import scipy
 scipy_major, scipy_minor, scipy_other = scipy.__version__.split('.', 2)
 
-if int(scipy_major) == 0 and int(scipy_minor) < 14:
-    warnings.warn('Support for Scipy 0.13 will be dropped in lmfit 0.9.5')
+if int(scipy_major) == 0 and int(scipy_minor) < 15:
+    warnings.warn('Support for Scipy 0.14 was dropped with lmfit 0.9.5')

lmfit/minimizer.py

@@ -14,20 +14,39 @@
 from copy import deepcopy
 import numpy as np
 from numpy import (dot, eye, ndarray, ones_like,
-                   sqrt, take, transpose, triu, deprecate)
+                   sqrt, take, transpose, triu)
 from numpy.dual import inv
 from numpy.linalg import LinAlgError
 import multiprocessing
 import numbers
 
+##
+##  scipy version notes:
+##  currently scipy 0.14 is required.
+##  feature           scipy version added
+##    minimize              0.11
+##    OptimizeResult        0.13
+##    diff_evolution        0.15
+##    least_squares         0.17
+##
+
 from scipy.optimize import leastsq as scipy_leastsq
+from scipy.optimize import minimize as scipy_minimize
 
 # differential_evolution is only present in scipy >= 0.15
 try:
     from scipy.optimize import differential_evolution as scipy_diffev
 except ImportError:
     from ._differentialevolution import differential_evolution as scipy_diffev
 
+# check for scipy.opitimize.least_squares
+HAS_LEAST_SQUARES = False
+try:
+    from scipy.optimize import least_squares
+    HAS_LEAST_SQUARES = True
+except ImportError:
+    pass
+
 # check for EMCEE
 HAS_EMCEE = False
 try:
@@ -44,22 +63,6 @@
 except ImportError:
     pass
 
-# check for scipy.optimize.minimize
-HAS_SCALAR_MIN = False
-try:
-    from scipy.optimize import minimize as scipy_minimize
-    HAS_SCALAR_MIN = True
-except ImportError:
-    pass
-
-# check for scipy.opitimize.least_squares
-HAS_LEAST_SQUARES = False
-try:
-    from scipy.optimize import least_squares
-    HAS_LEAST_SQUARES = True
-except ImportError:
-    pass
-
 from .parameter import Parameter, Parameters
 
 # use locally modified version of uncertainties package
@@ -528,8 +531,6 @@ def scalar_minimize(self, method='Nelder-Mead', params=None, **kws):
         .. versionchanged:: 0.9.0
            return value changed to :class:`MinimizerResult`
         """
-        if not HAS_SCALAR_MIN:
-            raise NotImplementedError
 
         result = self.prepare_fit(params=params)
         result.method = method
@@ -1232,17 +1233,12 @@ def minimize(self, method='leastsq', params=None, **kws):
             function = self.leastsq
         elif user_method.startswith('least_s'):
             function = self.least_squares
-        elif HAS_SCALAR_MIN:
+        else:
             function = self.scalar_minimize
             for key, val in SCALAR_METHODS.items():
                 if (key.lower().startswith(user_method) or
                     val.lower().startswith(user_method)):
                     kwargs['method'] = val
-        elif (user_method.startswith('nelder') or
-              user_method.startswith('fmin')):
-            function = self.fmin
-        elif user_method.startswith('lbfgsb'):
-            function = self.lbfgsb
         return function(**kwargs)
 
 

requirements.txt

@@ -1,2 +1,2 @@
 numpy>=1.5
-scipy>=0.13
+scipy>=0.14