PY23: optimize: callable() + exec + builtins

scipy/optimize/_minimize.py

@@ -14,6 +14,9 @@
 from warnings import warn
 
 from numpy import any
+
+from scipy.lib.six import callable
+
 # unconstrained minimization
 from .optimize import (_minimize_neldermead, _minimize_powell, _minimize_cg,
                       _minimize_bfgs, _minimize_newtoncg,
@@ -26,7 +29,6 @@
 from .tnc import _minimize_tnc
 from .cobyla import _minimize_cobyla
 from .slsqp import _minimize_slsqp
-import collections
 
 def minimize(fun, x0, args=(), method='BFGS', jac=None, hess=None,
              hessp=None, bounds=None, constraints=(), tol=None,
@@ -315,7 +317,7 @@ def minimize(fun, x0, args=(), method='BFGS', jac=None, hess=None,
              RuntimeWarning)
 
     # fun also returns the jacobian
-    if not isinstance(jac, collections.Callable):
+    if not callable(jac):
         if bool(jac):
             fun = MemoizeJac(fun)
             jac = fun.derivative

scipy/optimize/_root.py

@@ -10,12 +10,13 @@
 
 import numpy as np
 
+from scipy.lib.six import callable
+
 from warnings import warn
 
 from .optimize import MemoizeJac, Result, _check_unknown_options
 from .minpack import _root_hybr, leastsq
 from . import nonlin
-import collections
 
 def root(fun, x0, args=(), method='hybr', jac=None, tol=None, callback=None,
          options=None):
@@ -146,7 +147,7 @@ def root(fun, x0, args=(), method='hybr', jac=None, tol=None, callback=None,
              RuntimeWarning)
 
     # fun also returns the jacobian
-    if not isinstance(jac, collections.Callable) and meth in ('hybr', 'lm'):
+    if not callable(jac) and meth in ('hybr', 'lm'):
         if bool(jac):
             fun = MemoizeJac(fun)
             jac = fun.derivative

scipy/optimize/cobyla.py

@@ -11,10 +11,10 @@
 """
 
 import numpy as np
+from scipy.lib.six import callable
 from scipy.optimize import _cobyla
 from .optimize import Result, _check_unknown_options
 from warnings import warn
-import collections
 
 __all__ = ['fmin_cobyla']
 
@@ -138,14 +138,14 @@ def fmin_cobyla(func, x0, cons, args=(), consargs=None, rhobeg=1.0, rhoend=1e-4,
     try:
         m = len(cons)
     except TypeError:
-        if isinstance(cons, collections.Callable):
+        if callable(cons):
             m = 1
             cons = [cons]
         else:
             raise TypeError(err)
     else:
         for thisfunc in cons:
-            if not isinstance(thisfunc, collections.Callable):
+            if not callable(thisfunc):
                 raise TypeError(err)
 
     if consargs is None:

scipy/optimize/nonlin.py

@@ -113,6 +113,7 @@ def residual(P):
 
 import sys
 import numpy as np
+from scipy.lib.six import callable, exec_
 from scipy.linalg import norm, solve, inv, qr, svd, lstsq, LinAlgError
 from numpy import asarray, dot, vdot
 import scipy.sparse.linalg
@@ -588,7 +589,7 @@ def asjacobian(J):
                         setup=getattr(J, 'setup'),
                         dtype=J.dtype,
                         shape=J.shape)
-    elif isinstance(J, collections.Callable):
+    elif callable(J):
         # Assume it's a function J(x) that returns the Jacobian
         class Jac(Jacobian):
             def update(self, x, F):
@@ -1492,7 +1493,7 @@ def %(name)s(F, xin, iter=None %(kw)s, verbose=False, maxiter=None,
                              kwkw=kwkw_str)
     ns = {}
     ns.update(globals())
-    exec(wrapper, ns)
+    exec_(wrapper, ns)
     func = ns[name]
     func.__doc__ = jac.__doc__
     _set_doc(func)

scipy/optimize/optimize.py

@@ -25,12 +25,12 @@
 
 import warnings
 import numpy
+from scipy.lib.six import callable
 from numpy import atleast_1d, eye, mgrid, argmin, zeros, shape, \
      squeeze, vectorize, asarray, absolute, sqrt, Inf, asfarray, isinf
 from .linesearch import \
      line_search_BFGS, line_search_wolfe1, line_search_wolfe2, \
      line_search_wolfe2 as line_search
-import collections
 
 
 # standard status messages of optimizers
@@ -143,7 +143,8 @@ def is_array_scalar(x):
     return len(atleast_1d(x) == 1)
 
 abs = absolute
-import builtins
+
+from scipy.lib.six.moves import builtins
 pymin = builtins.min
 pymax = builtins.max
 __version__ = "0.7"
@@ -2253,7 +2254,7 @@ def _scalarfunc(*params):
     if (N == 1):
         grid = grid[0]
         xmin = xmin[0]
-    if isinstance(finish, collections.Callable):
+    if callable(finish):
         vals = finish(func, xmin, args=args, full_output=1, disp=0)
         xmin = vals[0]
         Jmin = vals[1]