Merge pull request #242 from nouiz/max_default

theano/compile/sharedvalue.py

@@ -18,13 +18,6 @@ from theano.gof import Container, Variable, generic
 
 _logger = logging.getLogger('theano.compile.sharedvalue')
 
-AddConfigVar('shared.value_borrows',
-        ("DEPRECATED. You should not use the 'value' property of shared"
-            " variables, but use the .get_value() and .set_value() methods."
-            " False: shared variables 'value' property is guaranteed to not"
-            " alias theano-managed memory. True: no guarantee, but faster."),
-        BoolParam(True),
-        in_c_key=False)
 
 class SharedVariable(Variable):
     """
@@ -125,29 +118,6 @@ class SharedVariable(Variable):
         cp.tag = copy.copy(self.tag)
         return cp
 
-    def _value_get(self):
-        warnings.warn(("The .value property of shared variables is deprecated."
-            " You should use the .get_value() method instead."),
-            stacklevel=2)
-        return self.get_value(borrow=config.shared.value_borrows, return_internal_type=False)
-    def _value_set(self, new_value):
-        warnings.warn(("The .value property of shared variables is deprecated."
-            " You should use the .set_value() method instead."),
-            stacklevel=2)
-        return self.set_value(new_value, borrow=config.shared.value_borrows)
-
-    #TODO: USE A CONFIG VARIABLE TO set these get/set methods to the non-borrowing versions
-    #      Semantically things are clearer when using non-borrow versions.  That should be the
-    #      default.  The default support transparently (if slowly) when the 'raw' value is in a
-    #      different memory space (e.g. GPU or other machine).
-    value = property(_value_get, _value_set,
-            doc=("DEPRECATED. Shortcut for self.get_value() and "
-                 "self.set_value(). "
-                 "The `borrow` argument to these methods is read from "
-                 "`theano.config.shared.value_borrows`. "
-                 "You should call get_value() and set_value() directly."))
-
-
     def filter_update(self, update):
         """
         When this shared variable is updated by a pfunc, the update value will be run through this function.

theano/tensor/opt_uncanonicalize.py

@@ -57,7 +57,7 @@ class MaxAndArgmaxOptimizer(Optimizer):
                     if len(node.outputs[1].clients)==0:
                         try:
                             axis=get_constant_value(node.inputs[1])
-                        except ValueError:
+                        except (ValueError, TypeError), e:
                             return False
 
                         new = CAReduce(scal.maximum,axis)(node.inputs[0])

theano/tensor/tests/test_opt_uncanonicalize.py

@@ -13,91 +13,95 @@ from theano.tests import unittest_tools as utt
 
 class T_max_and_argmax(unittest.TestCase):
     def test_optimization(self):
-        #If we use only the max output, we should replace this op with a faster one.
-        mode = theano.compile.mode.get_default_mode().including('canonicalize','fast_run')
+        #If we use only the max output, we should replace this op with
+        #a faster one.
+        mode = theano.compile.mode.get_default_mode().including(
+            'canonicalize', 'fast_run')
 
-        data = numpy.asarray(numpy.random.rand(2,3),dtype=config.floatX)
-        n = tensor.matrix()
+        for axis in [0, 1, -1]:
+            data = numpy.asarray(numpy.random.rand(2, 3), dtype=config.floatX)
+            n = tensor.matrix()
 
-        f = function([n], tensor.max_and_argmax(n,0)[0], mode=mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==1
-        assert isinstance(topo[0].op, CAReduce)
+            f = function([n], tensor.max_and_argmax(n, axis)[0], mode=mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 1
+            assert isinstance(topo[0].op, CAReduce)
 
-        f = function([n], tensor.max_and_argmax(n,0), mode=mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==1
-        assert isinstance(topo[0].op, tensor.MaxAndArgmax)
+            f = function([n], tensor.max_and_argmax(n, axis), mode=mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 1
+            assert isinstance(topo[0].op, tensor.MaxAndArgmax)
 
 
 class T_min_max(unittest.TestCase):
     def setUp(self):
         utt.seed_rng()
-        self.mode = theano.compile.mode.get_default_mode().including('canonicalize','fast_run')
+        self.mode = theano.compile.mode.get_default_mode().including(
+            'canonicalize', 'fast_run')
 
     def test_optimization_max(self):
-        data = numpy.asarray(numpy.random.rand(2,3),dtype=config.floatX)
+        data = numpy.asarray(numpy.random.rand(2, 3), dtype=config.floatX)
         n = tensor.matrix()
 
-        f = function([n],tensor.max(n,0), mode=self.mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==1
-        assert isinstance(topo[0].op,CAReduce)
-        f(data)
-
-
-        f = function([n],tensor.max(-n,0), mode=self.mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==2
-        assert isinstance(topo[0].op, Elemwise)
-        assert isinstance(topo[0].op.scalar_op, scalar.Neg)
-        assert isinstance(topo[1].op,CAReduce)
-        f(data)
-
-        f = function([n],-tensor.max(n,0), mode=self.mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==2
-        assert isinstance(topo[0].op,CAReduce)
-        assert isinstance(topo[1].op, Elemwise)
-        assert isinstance(topo[1].op.scalar_op, scalar.Neg)
-        f(data)
-
-        f = function([n],-tensor.max(-n,0), mode=self.mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==1
-        assert isinstance(topo[0].op,CAReduce)#min
-        f(data)
+        for axis in [0, 1, -1]:
+            f = function([n], tensor.max(n, axis), mode=self.mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 1
+            assert isinstance(topo[0].op, CAReduce)
+            f(data)
+
+            f = function([n], tensor.max(-n, axis), mode=self.mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 2
+            assert isinstance(topo[0].op, Elemwise)
+            assert isinstance(topo[0].op.scalar_op, scalar.Neg)
+            assert isinstance(topo[1].op, CAReduce)
+            f(data)
+
+            f = function([n], -tensor.max(n, axis), mode=self.mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 2
+            assert isinstance(topo[0].op, CAReduce)
+            assert isinstance(topo[1].op, Elemwise)
+            assert isinstance(topo[1].op.scalar_op, scalar.Neg)
+            f(data)
+
+            f = function([n], -tensor.max(-n, axis), mode=self.mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 1
+            assert isinstance(topo[0].op, CAReduce)  # min
+            f(data)
 
     def test_optimization_min(self):
-        data = numpy.asarray(numpy.random.rand(2,3),dtype=config.floatX)
+        data = numpy.asarray(numpy.random.rand(2, 3), dtype=config.floatX)
         n = tensor.matrix()
 
-        f = function([n],tensor.min(n,0), mode=self.mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==1
-        assert isinstance(topo[0].op,CAReduce)
-        f(data)
-
-        #test variant with neg to make sure we optimize correctly
-        f = function([n],tensor.min(-n,0), mode=self.mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==2
-        assert isinstance(topo[0].op,CAReduce)#max
-        assert isinstance(topo[1].op, Elemwise)
-        assert isinstance(topo[1].op.scalar_op, scalar.Neg)
-        f(data)
-
-        f = function([n],-tensor.min(n,0), mode=self.mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==2
-        assert isinstance(topo[0].op, Elemwise)
-        assert isinstance(topo[0].op.scalar_op, scalar.Neg)
-        assert isinstance(topo[1].op,CAReduce)#max
-        f(data)
-
-        f = function([n],-tensor.min(-n,0), mode=self.mode)
-        topo = f.maker.env.toposort()
-        assert len(topo)==1
-        assert isinstance(topo[0].op,CAReduce)#max
-        f(data)
-
+        for axis in [0, 1, -1]:
+            f = function([n], tensor.min(n, axis), mode=self.mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 1
+            assert isinstance(topo[0].op, CAReduce)
+            f(data)
+
+            #test variant with neg to make sure we optimize correctly
+            f = function([n], tensor.min(-n, axis), mode=self.mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 2
+            assert isinstance(topo[0].op, CAReduce)  # max
+            assert isinstance(topo[1].op, Elemwise)
+            assert isinstance(topo[1].op.scalar_op, scalar.Neg)
+            f(data)
+
+            f = function([n], -tensor.min(n, axis), mode=self.mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 2
+            assert isinstance(topo[0].op, Elemwise)
+            assert isinstance(topo[0].op.scalar_op, scalar.Neg)
+            assert isinstance(topo[1].op, CAReduce)  # max
+            f(data)
+
+            f = function([n], -tensor.min(-n, axis), mode=self.mode)
+            topo = f.maker.env.toposort()
+            assert len(topo) == 1
+            assert isinstance(topo[0].op, CAReduce)  # max
+            f(data)

theano/tensor/tests/test_sharedvar.py

@@ -350,27 +350,6 @@ def makeSharedTester(shared_constructor_,
                 assert may_share_memory(old_data, x_shared.container.storage[0])
                 x_shared.get_value(borrow=True)
 
-            # Test by .value
-            # As we know that .value is deprecated, we filter out the warning
-            warnings.filterwarnings(
-                    action='ignore',
-                    message='The .value property of shared variables is deprecated.'
-                    )
-            try:
-                nd += 1
-                old_data = x_shared.container.storage[0]
-                x_shared.value = nd
-                assert numpy.allclose(self.ref_fct(x_shared.value), self.ref_fct(self.cast_value(nd)))
-                assert may_share_memory(old_data, x_shared.container.storage[0]) == self.set_value_inplace
-            finally:
-                # Restore the default behavior.
-                # TODO There is a cleaner way to do this in Python 2.6, once
-                # Theano drops support of Python 2.4 and 2.5.
-                warnings.filterwarnings(
-                    action='default',
-                    message='The .value property of shared variables is deprecated.'
-                    )
-
             # Test by set_value with borrow=False
             nd += 1
             old_data = x_shared.container.storage[0]