merge

theano/compile/module.py

@@ -370,9 +370,9 @@ class Method(Component):
                 return memo[r]
             except KeyError:
                 if require:
-                    raise AllocationError('There is no storage associated to %s used by %s.'
+                    raise AllocationError('There is no storage associated to %s used by %s = %s.'
                                           ' Verify that it is indeed a Member of the'
-                                          ' enclosing module or of one of its submodules.' % (r, self))
+                                          ' enclosing module or of one of its submodules.' % (r, self.name, self))
                 else:
                     return io.In(result = r, value = gof.Container(r, storage = [None]), mutable = False)
         # Wrap the inputs in In instances. TODO: allow the inputs to _be_ In instances
@@ -639,6 +639,10 @@ class ComponentList(Composite):
             raise TypeError('ComponentList may only contain Components.', c, type(c))
         self._components.append(c)
 
+    def extend(self, other):
+        for o in other:
+            self.append(o)
+
     def __add__(self, other):
         if isinstance(other, (list, tuple)):
             return ComponentList(self._components + map(wrap,other))

theano/tensor/basic.py

@@ -581,6 +581,9 @@ class _tensor_py_operators:
     def __dot__(left, right): return dot(left, right)
     def __rdot__(right, left): return dot(left, right)
 
+    def sum(self, axis=None):
+        return elemwise.Sum(axis)(self)
+    
 
 class TensorResult(Result, _tensor_py_operators):
     """Subclass to add the tensor operators to the basic `Result` class."""
@@ -809,9 +812,16 @@ class MaxAndArgmax(Op):
 #        gMax * dMax/dx + gArgMax * dArgMax/dx, gMax * dMax/daxis + gArgMax * dArgMax/daxis
 #       g_max has one less dimension than x, so you need to complete g_max to x's shape
 #        when axis=0 the broadcasting mechanism does it automatically
-        assert axis.data == 0
-        g_x = eq(max(x, axis), x) * g_max
+        
+        assert axis.data == 0 or axis.data == x.ndim-1
+        g_max_pad = shape_padleft(g_max) if axis.data==0 else \
+                    shape_padright(g_max)
+        xmax = max(x, axis)
+        xmax_pad = shape_padleft(xmax) if axis.data==0 else \
+                   shape_padright(xmax)
+        g_x = eq(xmax_pad, x) * g_max_pad
         return g_x, None
+
 @_redefine_asRoutine(MaxAndArgmax())
 def max_and_argmax(a):
     pass
@@ -1624,7 +1634,7 @@ pprint.assign(lambda pstate, r: r.owner and isinstance(r.owner.op, Join),
 
 
 @constructor
-def shape_padleft(tensor, n_ones):
+def shape_padleft(tensor, n_ones=1):
     """Reshape `tensor` by left-padding the shape with `n_ones` 1s
     
     See also: `shape_padright` and `Dimshuffle`
@@ -1639,7 +1649,7 @@ def rightpad_shape(tensor, n_ones):
     return DimShuffle(tensor.broadcastable, pattern)(tensor)
 
 @constructor
-def shape_padright(tensor, n_ones):
+def shape_padright(tensor, n_ones=1):
     """Reshape `tensor` by right-padding the shape with `n_ones` 1s
     
     See also: `shape_padleft` and `Dimshuffle`

theano/tensor/raw_random.py

@@ -212,7 +212,7 @@ class RandomKit(SymbolicInputKit):
         return out
 
     def distribute(self, value, indices, containers):
-        rg = partial(numpy.random.RandomState(value).randint, 2**30)
+        rg = partial(numpy.random.RandomState(int(value)).randint, 2**30)
         elems = deque(zip(indices, containers))
         i = 0
         while elems: