use sphinx section names

theano/compile/nanguardmode.py

@@ -189,8 +189,8 @@ class NanGuardMode(Mode):
     big_is_error : bool
         If True, raise an error when a value greater than 1e10 is encountered.
 
-    Note
-    ----
+    Notes
+    -----
         We ignore the linker parameter
     """
 

theano/compile/ops.py

@@ -680,8 +680,8 @@ class Rebroadcast(gof.Op):
     -----
     Works inplace and works for CudaNdarrayType.
 
-    Example
-    -------
+    Examples
+    --------
     `Rebroadcast((0, True), (1, False))(x)` would make `x` broadcastable in
     axis 0 and not broadcastable in axis 1.
 

theano/gof/cmodule.py

@@ -65,8 +65,8 @@ def debug_counter(name, every=1):
 
     This is a utility function one may use when debugging.
 
-    Example
-    -------
+    Examples
+    --------
     debug_counter('I want to know how often I run this line')
 
     """

theano/gof/graph.py

@@ -866,8 +866,8 @@ def clone(i, o, copy_inputs=True, copy_orphans=None):
     object
         The inputs and outputs of that copy.
 
-    Note
-    ----
+    Notes
+    -----
 
     A constant, if in the ``i`` list is not an orpha. So it will be
     copied depending of the ``copy_inputs`` parameter. Otherwise it

theano/gof/link.py

@@ -393,8 +393,8 @@ class Linker(object):
         operate in the same storage the fgraph uses, else independent storage
         will be allocated for the function.
 
-        Example
-        -------
+        Examples
+        --------
         e = x + y
         fgraph = FunctionGraph([x, y], [e])
         fn = MyLinker(fgraph).make_function(inplace)

theano/gof/op.py

@@ -187,8 +187,8 @@ class CLinkerObject(object):
         Optional: Return a list of compile args recommended to compile the
         code returned by other methods in this class.
 
-        Example
-        -------
+        Examples
+        --------
         return ['-ffast-math']
 
         Compiler arguments related to headers, libraries and search paths should

theano/gpuarray/dnn.py

@@ -1748,8 +1748,8 @@ class GpuDnnPoolDesc(Op):
     pad : tuple
         (padX, padY) or (padX, padY, padZ)
 
-    Note
-    ----
+    Notes
+    -----
     Not used anymore. Only needed to reload old pickled files.
     """
 

theano/gradient.py

@@ -1782,8 +1782,8 @@ def verify_grad(
     no_debug_ref : bool
         Don't use DebugMode for the numerical gradient function.
 
-    Note
-    ----
+    Notes
+    -----
     This function does not support multiple outputs. In
     tests/test_scan.py there is an experimental verify_grad that
     covers that case as well by using random projections.
@@ -2380,8 +2380,8 @@ def grad_clip(x, lower_bound, upper_bound):
     >>> print(f(2.0))
     [array(1.0), array(4.0)]
 
-    Note
-    ----
+    Notes
+    -----
     We register an opt in tensor/opt.py that remove the GradClip.
     So it have 0 cost in the forward and only do work in the grad.
 

theano/scan_module/scan_utils.py

@@ -1393,8 +1393,8 @@ def forced_replace(out, x, y):
     x := sigmoid(wu)
     forced_replace(out, x, y) := y*(1-y)
 
-    Note
-    ----
+    Notes
+    -----
     When it find a match, it don't continue on the corresponding inputs.
     """
     if out is None:

theano/tensor/basic.py

@@ -225,8 +225,8 @@ def constant(x, name=None, ndim=None, dtype=None):
     ValueError
         `x` could not be expanded to have ndim dimensions.
 
-    Note
-    ----
+    Notes
+    -----
     We create a small cache of frequently used constant.
     This speed up the Merge optimization for big graph.
     We want to cache all scalar to don't merge as frequently constants.
@@ -4792,8 +4792,8 @@ def shape_padright(t, n_ones=1):
 def shape_padaxis(t, axis):
     """Reshape `t` by inserting 1 at the dimension `axis`.
 
-    Example
-    -------
+    Examples
+    --------
     >>> tensor = theano.tensor.tensor3()
     >>> theano.tensor.shape_padaxis(tensor, axis=0)
     DimShuffle{x,0,1,2}.0

theano/tensor/elemwise.py

@@ -80,8 +80,8 @@ class DimShuffle(COp):
     inplace : bool, optional
         If True (default), the output will be a view of the input.
 
-    Note
-    ----
+    Notes
+    -----
     If `j = new_order[i]` is an index, the output's ith dimension
     will be the input's jth dimension.
     If `new_order[i]` is `x`, the output's ith dimension will
@@ -91,8 +91,6 @@ class DimShuffle(COp):
     If `input.broadcastable[i] == False` then `i` must be found in new_order.
     Broadcastable dimensions, on the other hand, can be discarded.
 
-    Note
-    ----
     .. code-block:: python
 
         DimShuffle((False, False, False), ['x', 2, 'x', 0, 1])
@@ -115,8 +113,8 @@ class DimShuffle(COp):
     If the tensor has shape (1, 20), the resulting tensor will have shape
     (20, ).
 
-    Example
-    -------
+    Examples
+    --------
     .. code-block:: python
 
         DimShuffle((), ['x'])  # make a 0d (scalar) into a 1d vector
@@ -399,8 +397,8 @@ class Elemwise(OpenMPOp):
         variable number of inputs), whereas the numpy function may
         not have varargs.
 
-    Note
-    ----
+    Notes
+    -----
     | Elemwise(add) represents + on tensors (x + y)
     | Elemwise(add, {0 : 0}) represents the += operation (x += y)
     | Elemwise(add, {0 : 1}) represents += on the second argument (y += x)
@@ -1330,8 +1328,8 @@ class CAReduce(Op):
         - List of dimensions that we want to reduce
         - If None, all dimensions are reduced
 
-    Note
-    ----
+    Notes
+    -----
     .. code-block:: python
 
         CAReduce(add)      # sum (ie, acts like the numpy sum operation)

theano/tensor/slinalg.py

@@ -303,8 +303,6 @@ solve = Solve()
 Solves the equation ``a x = b`` for x, where ``a`` is a matrix and
 ``b`` can be either a vector or a matrix.
 
-Note
-
 Parameters
 ----------
 a : `(M, M) symbolix matrix`