Merge pull request #365 from tBuLi/py3_migration

Py3 final migration

.github/workflows/test.yaml

@@ -7,7 +7,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [ubuntu-latest, macos-latest, windows-latest]
+        os: [ubuntu-latest]
         python-version: [3.7, 3.8, 3.9, "3.10", 3.11]
     steps:
     - uses: actions/checkout@v3

symfit/contrib/interactive_guess/interactive_guess.py

@@ -6,7 +6,7 @@
 
 from ... import ODEModel, Derivative, latex
 from ...core.fit import TakesData
-from ...core.support import keywordonly, key2str, deprecated
+from ...core.support import key2str, deprecated
 
 import itertools
 
@@ -21,8 +21,7 @@ class InteractiveGuess(TakesData):
     """A class that provides an graphical, interactive way of guessing initial
     fitting parameters."""
 
-    @keywordonly(n_points=50, log_contour=True, percentile=(5, 95))
-    def __init__(self, *args, **kwargs):
+    def __init__(self, *args, n_points=50, log_contour=True, percentile=(5, 95), **kwargs):
         """Create a matplotlib window with sliders for all parameters
         in this model, so that you may graphically guess initial fitting
         parameters. n_points is the number of points drawn for the plot.
@@ -51,9 +50,8 @@ def __init__(self, *args, **kwargs):
             corresponds to a 90% percentile. Should be a list of 2 numbers.
         :type percentile: list
         """
-        self.log_contour = kwargs.pop('log_contour')
-        n_points = kwargs.pop('n_points')
-        self.percentile = kwargs.pop('percentile')
+        self.log_contour = log_contour
+        self.percentile = percentile
         super(InteractiveGuess, self).__init__(*args, **kwargs)
         if len(self.independent_data) > 1:
             self._dimension_strategy = StrategynD(self)
@@ -99,8 +97,7 @@ def __init__(self, *args, **kwargs):
         self._set_up_sliders()
         self._update_plot(None)
 
-    @keywordonly(show=True, block=True)
-    def execute(self, **kwargs):
+    def execute(self, *, show=True, block=True, **kwargs):
         """
         Execute the interactive guessing procedure.
 
@@ -112,7 +109,7 @@ def execute(self, **kwargs):
         Any additional keyword arguments are passed to
         matplotlib.pyplot.show().
         """
-        show = kwargs.pop('show')
+        kwargs.update(block=block)
         if show:
             # self.fig.show()  # Apparently this does something else,
             # see https://github.com/matplotlib/matplotlib/issues/6138

symfit/core/fit.py

@@ -4,13 +4,12 @@
 
 from collections import OrderedDict
 from collections.abc import Sequence
-import sys
 
 import sympy
 import numpy as np
 
 from symfit.core.argument import Variable
-from .support import keywordonly, key2str
+from .support import key2str
 from .minimizers import (
     BFGS, SLSQP, LBFGSB, BaseMinimizer, GradientMinimizer, HessianMinimizer,
     ConstrainedMinimizer, MINPACK, ChainedMinimizer, BasinHopping
@@ -21,19 +20,15 @@
 )
 from .models import BaseModel, Model, BaseNumericalModel, CallableModel
 
-if sys.version_info >= (3,0):
-    import inspect as inspect_sig
-else:
-    import funcsigs as inspect_sig
+import inspect
 
 class TakesData(object):
     """
     An base class for everything that takes data. Most importantly, it takes care
     of linking the provided data to variables. The allowed variables are extracted
     from the model.
     """
-    @keywordonly(absolute_sigma=None)
-    def __init__(self, model, *ordered_data, **named_data):
+    def __init__(self, model, *ordered_data, absolute_sigma=None, **named_data):
         """
         :param model: (dict of) sympy expression or ``Model`` object.
         :param bool absolute_sigma: True by default. If the sigma is only used
@@ -51,7 +46,6 @@ def __init__(self, model, *ordered_data, **named_data):
         with sigma\_. For example, let x be a Variable. Then sigma_x will give the
         stdev in x.
         """
-        absolute_sigma = named_data.pop('absolute_sigma')
         if isinstance(model, BaseModel):
             self.model = model
         else:
@@ -130,7 +124,7 @@ def _make_signature(self):
         """
         parameters = self._make_parameters(self.model)
         parameters = sorted(parameters, key=lambda p: p.default is None)
-        return inspect_sig.Signature(parameters=parameters)
+        return inspect.Signature(parameters=parameters)
 
     @staticmethod
     def _make_parameters(model, none_allowed=None):
@@ -149,10 +143,10 @@ def _make_parameters(model, none_allowed=None):
         if none_allowed is None:
             none_allowed = model.sigmas.values()
         parameters = [
-            inspect_sig.Parameter(
+            inspect.Parameter(
                 var.name,
-                kind=inspect_sig.Parameter.POSITIONAL_OR_KEYWORD,
-                default=None if var in none_allowed else inspect_sig.Parameter.empty
+                kind=inspect.Parameter.POSITIONAL_OR_KEYWORD,
+                default=None if var in none_allowed else inspect.Parameter.empty
             )
             for var in model.vars
         ]
@@ -338,9 +332,8 @@ class Fit(HasCovarianceMatrix):
         fit_result = fit.execute(minimizer_kwargs=[dict(xatol=0.1), {}])
     """
 
-    @keywordonly(objective=None, minimizer=None, constraints=None,
-                 absolute_sigma=None)
-    def __init__(self, model, *ordered_data, **named_data):
+    def __init__(self, model, *ordered_data, objective=None, minimizer=None,
+                 constraints=None, absolute_sigma=None, **named_data):
         """
 
         :param model: (dict of) sympy expression(s) or ``Model`` object.
@@ -364,10 +357,6 @@ def __init__(self, model, *ordered_data, **named_data):
         :param named_data: assign dependent, independent and sigma variables
             data by name.
         """
-        objective = named_data.pop('objective')
-        minimizer = named_data.pop('minimizer')
-        constraints = named_data.pop('constraints')
-        absolute_sigma = named_data.pop('absolute_sigma')
         # Should be a list of Constraint objects
         constraints = [] if constraints is None else constraints
 
@@ -439,7 +428,7 @@ def _make_signature(self):
                 unique_parameters.append(par)
 
         parameters = sorted(unique_parameters, key=lambda p: p.default is None)
-        return inspect_sig.Signature(parameters=parameters)
+        return inspect.Signature(parameters=parameters)
 
     def _determine_minimizer(self):
         """

symfit/core/fit_results.py

@@ -9,7 +9,6 @@
 from symfit.core.objectives import (
     LeastSquares, VectorLeastSquares, LogLikelihood
 )
-from symfit.core.support import keywordonly
 
 class FitResults(object):
     """
@@ -26,8 +25,7 @@ class FitResults(object):
     their optimized values. Can be `**` unpacked when evaluating
     :class:`~symfit.core.models.Model`'s.
     """
-    @keywordonly(constraints=None)
-    def __init__(self, model, popt, covariance_matrix, minimizer, objective, message, **minimizer_output):
+    def __init__(self, model, popt, covariance_matrix, minimizer, objective, message, *, constraints=None, **minimizer_output):
         """
         :param model: :class:`~symfit.core.models.Model` that was fit to.
         :param popt: best fit parameters, same ordering as in model.params.
@@ -37,7 +35,6 @@ def __init__(self, model, popt, covariance_matrix, minimizer, objective, message
         :param message: Status message returned by the minimizer.
         :param \**minimizer_output: Raw output as given by the minimizer.
         """
-        constraints = minimizer_output.pop('constraints')
         self.constraints = constraints if constraints is not None else []
         self.minimizer_output = minimizer_output
         self.model = model