Merge pull request #1355 from delallea/minor

doc/extending/optimization.txt

@@ -21,9 +21,10 @@ In this section we will define a couple optimizations on doubles.
 
 .. note::
 
-   There is the optimization tag `cxx_only` that tell this
-   optimization will insert Op that only have c code. So we should not
-   run them when we don't have a c++ compiler.
+   The optimization tag `cxx_only` is used for optimizations that insert
+   Ops which have no Python implementation (so they only have C code).
+   Optimizations with this tag are skipped when there is no C++ compiler
+   available.
 
 Global and local optimizations
 ==============================

doc/install.txt

@@ -442,9 +442,9 @@ correctly (for example, for MKL this might be ``-lmkl -lguide -lpthread`` or
 
     If you have problems linking with MKL, `Intel Line Advisor
     <http://software.intel.com/en-us/articles/intel-mkl-link-line-advisor>`_
-    and `MKL User Guide
+    and the `MKL User Guide
     <http://software.intel.com/sites/products/documentation/doclib/mkl_sa/11/mkl_userguide_lnx/index.htm>`_
-    can help you find the correct flag to use.
+    can help you find the correct flags to use.
 
 .. _gpu_linux:
 

doc/library/config.txt

@@ -410,17 +410,18 @@ import theano and print the config variable, as in:
 .. attribute:: config.cxx
 
 
-    Default: 'g++' if g++ is present. '' Otherwise.
+    Default: 'g++' if g++ is present. Empty string otherwise.
 
-    Tell the c++ compiler to use. If empty, don't compile c++ code.
-    We automatically detect if g++ is present and disable it if not
-    present.
+    Indicates which C++ compiler to use. If empty, no C++ code is compiled.
+    Theano automatically detects whether g++ is present and disables
+    C++ compilation when it is not.
 
     We print a warning if we detect that g++ is not present. It is
-    recommended to run with c++ compilation as Theano will be much
+    recommended to run with C++ compilation as Theano will be much
     slower otherwise.
 
-    Currently only g++ is supported, but supporting others is easy.
+    Currently only g++ is supported, but supporting other compilers should
+    not be too difficult.
 
 .. attribute:: optimizer_excluding
 
@@ -636,12 +637,12 @@ import theano and print the config variable, as in:
 
     Bool value, default: False
 
-    If True, will remove -O* parameter passed to g++.
-    This is useful to debug in gdb module compiled by Theano.
+    If True, will remove the -O* parameter passed to g++.
+    This is useful to debug in gdb modules compiled by Theano.
     The parameter -g is passed by default to g++.
 
 .. attribute:: cmodule.compilation_warning
 
     Bool value, default: False
 
-    If True, will print compilation warning.
+    If True, will print compilation warnings.

doc/library/scan.txt

@@ -295,25 +295,26 @@ the following:
 
 .. code-block:: python
 
- W = theano.shared ( W_values ) # we assume that ``W_values`` contains the
+ W = theano.shared(W_values) # we assume that ``W_values`` contains the
                              # initial values of your weight matrix
 
- bvis = theano.shared( bvis_values)
- bhid = theano.shared( bhid_values)
+ bvis = theano.shared(bvis_values)
+ bhid = theano.shared(bhid_values)
 
  trng = T.shared_randomstreams.RandomStreams(1234)
 
- def OneStep( vsample) :
-    hmean   = T.nnet.sigmoid( theano.dot( vsample, W) + bhid)
-    hsample = trng.binomial( size = hmean.shape, n = 1, prob = hmean)
-    vmean   = T.nnet.sigmoid( theano.dot( hsample. W.T) + bvis)
-    return trng.binomial( size = vsample.shape, n = 1, prob = vsample)
+ def OneStep(vsample) :
+    hmean = T.nnet.sigmoid(theano.dot(vsample, W) + bhid)
+    hsample = trng.binomial(size=hmean.shape, n=1, p=hmean)
+    vmean = T.nnet.sigmoid(theano.dot(hsample, W.T) + bvis)
+    return trng.binomial(size=vsample.shape, n=1, p=vmean,
+                         dtype=theano.config.floatX)
 
  sample = theano.tensor.vector()
 
- values, updates = theano.scan( OneStep, outputs_info = sample, n_steps = 10 )
+ values, updates = theano.scan(OneStep, outputs_info=sample, n_steps=10)
 
- gibbs10 = theano.function([sample], values[-1], updates = updates)
+ gibbs10 = theano.function([sample], values[-1], updates=updates)
 
 
 Note that if we use shared variables ( ``W``, ``bvis``, ``bhid``) but
@@ -335,7 +336,7 @@ afterwards. Look at this example :
 .. code-block:: python
 
  a = theano.shared(1)
- values,updates = theano.scan( lambda : {a:a+1}, n_steps = 10 )
+ values, updates = theano.scan(lambda: {a: a+1}, n_steps=10)
 
 In this case the lambda expression does not require any input parameters
 and returns an update dictionary which tells how ``a`` should be updated
@@ -343,9 +344,9 @@ after each step of scan. If we write :
 
 .. code-block:: python
 
-  b = a+1
+  b = a + 1
   c = updates[a] + 1
-  f = theano.function([], [b,c], updates = updates)
+  f = theano.function([], [b, c], updates=updates)
 
   print b
   print c

doc/library/tensor/basic.txt

@@ -646,30 +646,30 @@ dimensions, see :meth:`_tensor_py_operators.dimshuffle`.
     >>> x = T.concatenate([x0, x1[0], T.shape_padright(x2)], axis=1)
     >>> # x.ndim == 2
 
-.. function:: stacklist(tensor_list)
+.. function:: stacklists(tensor_list)
 
-    :type tensor_list: an iterable that contain tensors or iterable
-        with at the end tensors.
-    :param tensor_list: tensors to be
-        stackend together.
+    :type tensor_list: an iterable that contains either tensors or other
+        iterables of the same type as `tensor_list` (in other words, this
+        is a tree whose leaves are tensors).
+    :param tensor_list: tensors to be stacked together.
 
-    Recursivly stack lists of tensors to maintain similar structure.
+    Recursively stack lists of tensors to maintain similar structure.
 
-    This function can create a tensor from a shaped list of scalars
+    This function can create a tensor from a shaped list of scalars:
 
     >>> from theano.tensor import stacklists, scalars, matrices
     >>> from theano import function
-    >>> a,b,c,d = scalars('abcd')
+    >>> a, b, c, d = scalars('abcd')
     >>> X = stacklists([[a, b], [c, d]])
     >>> f = function([a, b, c, d], X)
     >>> f(1, 2, 3, 4)
     >>> # array([[ 1.,  2.], [ 3.,  4.]], dtype=float32)
 
     We can also stack arbitrarily shaped tensors. Here we stack matrices into
-    a 2 by 2 grid.
+    a 2 by 2 grid:
 
     >>> from numpy import ones
-    >>> a,b,c,d, = matrices('abcd')
+    >>> a, b, c, d = matrices('abcd')
     >>> X = stacklists([[a, b], [c, d]])
     >>> f = function([a, b, c, d], X)
     >>> x = ones((4, 4), 'float32')

doc/optimizations.txt

@@ -20,8 +20,8 @@ If you would like to add an additional optimization, refer to
 
     This list is partial.
 
-    The print_summary method allow several OpDBs and optimizers to list the optimization executed.
-    This allow to have an up-to-date list.
+    The print_summary method allows several OpDBs and optimizers to list the executed optimizations.
+    This makes it possible to have an up-to-date list.
 
     python -c 'import theano; theano.compile.FAST_RUN.optimizer.print_summary()'
 
@@ -255,6 +255,6 @@ Optimization                                              FAST_RUN  FAST_COMPILE
 
     local_log_softmax
         This is a stabilization optimization.
-        It can happen due to rounding problem that the softmax probability of one value get to 0.
-        Taking the log of 0, would generate -inf that will probably generate NaN later.
+        It can happen due to rounding errors that the softmax probability of one value gets to 0.
+        Taking the log of 0 would generate -inf that will probably generate NaN later.
         We return a closer answer.

doc/tutorial/extending_theano.txt

@@ -397,8 +397,8 @@ have to be jointly optimized explicitly in the code.)
 SciPy
 -----
 
-We can wrap SciPy function in Theano. But Scipy is an optional dependency.
-Here is some code that allow to make the op Optional:
+We can wrap SciPy functions in Theano. But SciPy is an optional dependency.
+Here is some code that allows the Op to be optional:
 
 .. code-block:: python
 
@@ -413,17 +413,19 @@ Here is some code that allow to make the op Optional:
         ...
         def make_node(self, x):
             assert imported_scipy, (
-            "Scipy not available. Scipy is needed for the SomeOp op.")
+            "SciPy not available. SciPy is needed for the SomeOp op.")
+            ...
 
     from nose.plugins.skip import SkipTest
-    class test_Solve(utt.InferShapeTester):
+    class test_SomeOp(utt.InferShapeTester):
         ...
         def test_infer_shape(self):
             if not imported_scipy:
-                raise SkipTest("Scipy needed for the Cholesky op.")
+                raise SkipTest("SciPy needed for the SomeOp op.")
+            ...
 
-Random number in tests
-----------------------
+Random numbers in tests
+-----------------------
 
 Making tests errors more reproducible is a good practice. To make your
 tests more reproducible, you need a way to get the same random
@@ -449,7 +451,7 @@ tutorial :ref:`Extending Theano<extending>`
 See :ref:`metadocumentation`, for some information on how to generate
 the documentation.
 
-Here is an example how to add docstring to an class.
+Here is an example how to add docstring to a class.
 
 .. code-block:: python
 
@@ -460,7 +462,7 @@ Here is an example how to add docstring to an class.
 
     :param x: input tensor.
 
-    :return: a tensor of the shape shape and dtype as the input with all
+    :return: a tensor of the same shape and dtype as the input with all
         values doubled.
 
     :note:
@@ -473,7 +475,8 @@ Here is an example how to add docstring to an class.
     .. versionadded:: 0.6
     """
 
-This is how it will show up for file that we auto list in the library documentation:
+This is how it will show up for files that we auto-list in the library
+documentation:
 
 
 .. automodule:: theano.misc.doubleop

doc/tutorial/faq.txt

@@ -24,7 +24,8 @@ internals cannot be modified.
 Faster gcc optimization
 -----------------------
 
-You can enable faster gcc optimization with the ``cxxflags``. This list of flags was suggested on the mailing list::
+You can enable faster gcc optimization with the ``cxxflags`` option.
+This list of flags was suggested on the mailing list::
 
     -O3 -ffast-math -ftree-loop-distribution -funroll-loops -ftracer
 

theano/compile/debugmode.py

@@ -886,8 +886,8 @@ def _lessbroken_deepcopy(a):
     """
     :param a: any object
 
-    Returns a copy of `a` that shares no internal storage with the original.
-    A deep copy.
+    Returns a copy of `a` that shares no internal storage with the original
+    (a deep copy).
     This function handles numpy arrays specially, because copy.deepcopy()
     called on a 0-d array will return a numpy scalar, not an array.
     """
@@ -2199,22 +2199,29 @@ class _Maker(FunctionMaker):  # inheritance buys a few helper functions
                     raise StochasticOrder(infolog.getvalue())
                 else:
                     if self.verbose:
-                        print >> sys.stderr, "OPTCHECK: optimization", i, "of", len(li), "events was stable."
+                        print >> sys.stderr, "OPTCHECK: optimization", i, \
+                                 "of", len(li), "events was stable."
             else:
                 fgraph0 = fgraph
 
-
         del fgraph0
         self.fgraph = fgraph
         #equivalence_tracker.printstuff()
 
         linker = _Linker(self)
 
+        # the 'no_borrow' outputs are the ones for which that we can't return
+        # the internal storage pointer.
 
-        #the 'no_borrow' outputs are the ones for which that we can't return the internal storage pointer.
-        no_borrow = [output for output, spec in zip(fgraph.outputs, outputs+additional_outputs) if not spec.borrow]
+        no_borrow = [
+                output
+                for output, spec in izip(fgraph.outputs,
+                                         outputs + additional_outputs)
+                if not spec.borrow]
         if no_borrow:
-            self.linker = linker.accept(fgraph, no_recycling = infer_reuse_pattern(fgraph, no_borrow))
+            self.linker = linker.accept(
+                    fgraph,
+                    no_recycling=infer_reuse_pattern(fgraph, no_borrow))
         else:
             self.linker = linker.accept(fgraph)
 

theano/compile/mode.py

@@ -86,7 +86,7 @@ def register_linker(name, linker):
 # If a string is passed as the optimizer argument in the constructor
 # for Mode, it will be used as the key to retrieve the real optimizer
 # in this dictionary
-exclude=[]
+exclude = []
 if not theano.config.cxx:
     exclude = ['cxx_only']
 OPT_FAST_RUN = gof.Query(include=['fast_run'], exclude=exclude)
@@ -120,7 +120,7 @@ def register_optimizer(name, opt):
 class AddDestroyHandler(gof.Optimizer):
     """This optimizer performs two important functions:
 
-    1) it has a 'requirement' of the destroyhandler.  This means that the fgraph
+    1) It has a 'requirement' of the destroyhandler. This means that the fgraph
     will include it as a feature for this optimization, and keep this feature
     enabled for subsequent optimizations.  All optimizations that work inplace
     on any of their inputs must run *after* this optimization to ensure that

theano/configdefaults.py

@@ -131,9 +131,10 @@ else:
     enum = EnumStr("")
 
 AddConfigVar('cxx',
-             "The c++ compiler to use. Currently only g++ is"
-             " supported. But supporting more is easy if someone want this."
-             "If it is empty, we don't compile c++ code.",
+             "The C++ compiler to use. Currently only g++ is"
+             " supported, but supporting additional compilers should not be "
+             "too difficult. "
+             "If it is empty, no C++ code is compiled.",
              enum,
              in_c_key=False)
 del enum

theano/gof/cmodule.py

@@ -45,13 +45,13 @@ AddConfigVar('cmodule.warn_no_version',
              in_c_key=False)
 
 AddConfigVar('cmodule.remove_gxx_opt',
-             "If True, will remove -O* parameter passed to g++."
-             "This is useful to debug in gdb module compiled by Theano."
+             "If True, will remove the -O* parameter passed to g++."
+             "This is useful to debug in gdb modules compiled by Theano."
              "The parameter -g is passed by default to g++",
              BoolParam(False))
 
 AddConfigVar('cmodule.compilation_warning',
-             "If True, will print compilation warning.",
+             "If True, will print compilation warnings.",
              BoolParam(False))
 
 
@@ -162,13 +162,15 @@ static struct PyModuleDef moduledef = {{
       MyMethods,
 }};
 """.format(name=self.hash_placeholder)
-            print >> stream, "PyMODINIT_FUNC PyInit_%s(void) {" % self.hash_placeholder
+            print >> stream, ("PyMODINIT_FUNC PyInit_%s(void) {" %
+                              self.hash_placeholder)
             for block in self.init_blocks:
                 print >> stream, '  ', block
             print >> stream, "    PyObject *m = PyModule_Create(&moduledef);"
             print >> stream, "    return m;"
         else:
-            print >> stream, "PyMODINIT_FUNC init%s(void){" % self.hash_placeholder
+            print >> stream, ("PyMODINIT_FUNC init%s(void){" %
+                              self.hash_placeholder)
             for block in self.init_blocks:
                 print >> stream, '  ', block
             print >> stream, '  ', ('(void) Py_InitModule("%s", MyMethods);'

theano/gradient.py

@@ -869,17 +869,19 @@ def _populate_grad_dict(var_to_app_to_idx,
                 for o, og in zip(node.outputs, output_grads):
                     o_dt = getattr(o.type, 'dtype', None)
                     og_dt = getattr(og.type, 'dtype', None)
-                    if o_dt not in theano.tensor.discrete_dtypes and og_dt and o_dt != og_dt:
+                    if (o_dt not in theano.tensor.discrete_dtypes and
+                        og_dt and o_dt != og_dt):
                         new_output_grads.append(og.astype(o_dt))
                     else:
                         new_output_grads.append(og)
 
-                # Make sure that, if new_output_grads[i] has a floating point dtype,
-                # it is the same dtype as outputs[i]
+                # Make sure that, if new_output_grads[i] has a floating point
+                # dtype, it is the same dtype as outputs[i]
                 for o, ng in zip(node.outputs, new_output_grads):
                     o_dt = getattr(o.type, 'dtype', None)
                     ng_dt = getattr(ng.type, 'dtype', None)
-                    if ng_dt is not None and o_dt not in theano.tensor.discrete_dtypes:
+                    if (ng_dt is not None and
+                        o_dt not in theano.tensor.discrete_dtypes):
                         assert ng_dt == o_dt
 
                 # Someone who had obviously not read the Op contract tried
@@ -890,7 +892,8 @@ def _populate_grad_dict(var_to_app_to_idx,
                 # 2) Talk to Ian Goodfellow
                 # (Both of these sources will tell you not to do it)
                 for ng in new_output_grads:
-                    assert getattr(ng.type, 'dtype', None) not in theano.tensor.discrete_dtypes
+                    assert (getattr(ng.type, 'dtype', None)
+                            not in theano.tensor.discrete_dtypes)
 
                 input_grads = node.op.grad(inputs, new_output_grads)
 
@@ -908,7 +911,6 @@ def _populate_grad_dict(var_to_app_to_idx,
 
             # Do type checking on the result
 
-
             # List of bools indicating if each input only has integer outputs
             only_connected_to_int = [(True not in
                 [in_to_out and out_to_cost and not out_int
@@ -916,7 +918,6 @@ def _populate_grad_dict(var_to_app_to_idx,
                     zip(in_to_outs, outputs_connected, output_is_int)])
                 for in_to_outs in connection_pattern]
 
-
             for i, term in enumerate(input_grads):
 
                 # Disallow Nones
@@ -933,7 +934,6 @@ def _populate_grad_dict(var_to_app_to_idx,
                             'the grad_undefined or grad_unimplemented helper '
                             'functions.') % node.op)
 
-
                 if not isinstance(term.type,
                         (NullType, DisconnectedType)):
                     if term.type.dtype not in theano.tensor.float_dtypes:
@@ -973,8 +973,8 @@ def _populate_grad_dict(var_to_app_to_idx,
                             msg += "evaluate to zeros, but it evaluates to"
                             msg += "%s."
 
-                            msg % (str(node.op), str(term), str(type(term)),
-                                    i, str(theano.get_scalar_constant_value(term)))
+                            msg % (node.op, term, type(term), i,
+                                   theano.get_scalar_constant_value(term))
 
                             raise ValueError(msg)
 
@@ -1010,8 +1010,6 @@ def _populate_grad_dict(var_to_app_to_idx,
             #cache the result
             term_dict[node] = input_grads
 
-
-
         return term_dict[node]
 
     # populate grad_dict[var] and return it
@@ -1040,7 +1038,7 @@ def _populate_grad_dict(var_to_app_to_idx,
                         if isinstance(term.type, DisconnectedType):
                             continue
 
-                        if hasattr(var,'ndim') and term.ndim != var.ndim:
+                        if hasattr(var, 'ndim') and term.ndim != var.ndim:
                             raise ValueError(("%s.grad returned a term with"
                                 " %d dimensions, but %d are required.") % (
                                     str(node.op), term.ndim, var.ndim))
@@ -1058,8 +1056,8 @@ def _populate_grad_dict(var_to_app_to_idx,
                 if cost_name is not None and var.name is not None:
                     grad_dict[var].name = '(d%s/d%s)' % (cost_name, var.name)
             else:
-                # this variable isn't connected to the cost in the computational
-                # graph
+                # this variable isn't connected to the cost in the
+                # computational graph
                 grad_dict[var] = DisconnectedType()()
         # end if cache miss
         return grad_dict[var]
@@ -1068,6 +1066,7 @@ def _populate_grad_dict(var_to_app_to_idx,
 
     return rval
 
+
 def _float_zeros_like(x):
     """ Like zeros_like, but forces the object to have a
     a floating point dtype """
@@ -1317,9 +1316,9 @@ def verify_grad(fun, pt, n_tests=2, rng=None, eps=None,
     :param eps: stepsize used in the Finite Difference Method (Default
         None is type-dependent)
         Raising the value of eps can raise or lower the absolute and
-        relative error of the verification depending of the
-        Op. Raising the eps do not lower the verification quality. It
-        is better to raise eps then raising abs_tol or rel_tol.
+        relative errors of the verification depending on the
+        Op. Raising eps does not lower the verification quality. It
+        is better to raise eps than raising abs_tol or rel_tol.
     :param out_type: dtype of output, if complex (i.e. 'complex32' or
         'complex64')
     :param abs_tol: absolute tolerance used as threshold for gradient
@@ -1599,6 +1598,7 @@ def hessian(cost, wrt, consider_constant=None,
         hessians.append(hess)
     return format_as(using_list, using_tuple, hessians)
 
+
 def _is_zero(x):
     """
     Returns 'yes', 'no', or 'maybe' indicating whether x

theano/misc/doubleop.py

@@ -8,7 +8,7 @@ class DoubleOp(theano.Op):
 
     :param x: input tensor.
 
-    :return: a tensor of the shape shape and dtype as the input with all
+    :return: a tensor of the same shape and dtype as the input with all
         values doubled.
 
     :note:
@@ -46,8 +46,8 @@ class DoubleOp(theano.Op):
 
     def R_op(self, inputs, eval_points):
         # R_op can receive None as eval_points.
-        # That mean there is no diferientiable path through that input
-        # If this imply that you cannot compute some outputs,
+        # That means there is no differentiable path through that input.
+        # If this implies that you cannot compute some outputs,
         # return None for those.
         if eval_points[0] is None:
             return eval_points

theano/sandbox/cuda/__init__.py

@@ -244,7 +244,8 @@ class GpuOp(theano.gof.Op):
         return super(GpuOp, self).make_thunk(node, storage_map,
                                              compute_map, no_recycling)
 
-theano.compile.debugmode.default_make_thunk.append(get_unbound_function(GpuOp.make_thunk))
+theano.compile.debugmode.default_make_thunk.append(
+                                        get_unbound_function(GpuOp.make_thunk))
 
 # We must do those import to be able to create the full doc when
 # nvcc is not available
@@ -271,7 +272,8 @@ if cuda_available:
     shared_constructor = float32_shared_constructor
 
     import basic_ops
-    from basic_ops import (GpuFromHost, HostFromGpu, GpuElemwise,
+    from basic_ops import (
+            GpuFromHost, HostFromGpu, GpuElemwise,
             GpuDimShuffle, GpuCAReduce, GpuReshape, GpuContiguous,
             GpuSubtensor, GpuIncSubtensor,
             GpuAdvancedSubtensor1, GpuAdvancedIncSubtensor1,
@@ -388,16 +390,17 @@ def use(device,
                 cuda_enabled = True
 
             if config.print_active_device:
-                print >> sys.stderr, "Using gpu device %d: %s" %(
+                print >> sys.stderr, "Using gpu device %d: %s" % (
                         active_device_number(), active_device_name())
             if device_properties(use.device_number)['regsPerBlock'] < 16384:
                 # We will try to use too much register per bloc at many places
                 # when there is only 8k register per multi-processor.
-                _logger.warning("You are probably using an old GPU."
-                                " We didn't optimize nor we support those GPU."
-                                " This mean GPU code will be slow AND will"
-                                " crash when we try to use feature/properties"
-                                " that your GPU don't support.")
+                _logger.warning(
+                        "You are probably using an old GPU, that Theano"
+                        " does not support."
+                        " This means GPU code will most likely be slow AND may"
+                        " crash when we try to use features"
+                        " that your GPU does not support.")
 
         except (EnvironmentError, ValueError, RuntimeError), e:
             _logger.error(("ERROR: Not using GPU."

theano/scan_module/scan_op.py

@@ -228,7 +228,7 @@ class Scan(PureOp):
                    )
         err_msg2 = ('When compiling the inner function of scan the '
                     'following error has been encountered: The '
-                    'initial state (outputs_info in scan nomenclature)'
+                    'initial state (outputs_info in scan nomenclature) '
                     'of variable %s (argument number %d)'
                     ' has dtype %s and %d dimension(s), while the result '
                     'of the inner function for this output has dtype %s '
@@ -1387,6 +1387,7 @@ class Scan(PureOp):
                         self.inner_nitsot_outs(self_outputs))
         scan_node = outs[0].owner
         connection_pattern = self.connection_pattern(scan_node)
+
         def get_inp_idx(iidx):
             if iidx < self.n_seqs:
                 return 1 + iidx
@@ -1428,10 +1429,10 @@ class Scan(PureOp):
             odx = get_out_idx(self_outputs.index(y))
             wrt = [x for x in theano.gof.graph.inputs([y])
                    if (x in diff_inputs) and
-                    (connection_pattern[get_inp_idx(self_inputs.index(x))][odx])]
+                   connection_pattern[get_inp_idx(self_inputs.index(x))][odx]]
             grads = gradient.grad(
-                    cost = None,
-                    known_grads = {y : g_y },
+                    cost=None,
+                    known_grads={y: g_y},
                     wrt=wrt, consider_constant=wrt,
                     disconnected_inputs='ignore',
                     return_disconnected='None')

theano/tensor/basic.py

@@ -8236,13 +8236,14 @@ def diag(v, k=0):
 
 
 def stacklists(arg):
-    """ Recursivly stack lists of tensors to maintain similar structure
+    """
+    Recursively stack lists of tensors to maintain similar structure.
 
-    This function can create a tensor from a shaped list of scalars
+    This function can create a tensor from a shaped list of scalars:
 
     >>> from theano.tensor import stacklists, scalars, matrices
     >>> from theano import function
-    >>> a,b,c,d = scalars('abcd')
+    >>> a, b, c, d = scalars('abcd')
     >>> X = stacklists([[a, b], [c, d]])
     >>> f = function([a, b, c, d], X)
     >>> f(1, 2, 3, 4)
@@ -8250,10 +8251,10 @@ def stacklists(arg):
            [ 3.,  4.]], dtype=float32)
 
     We can also stack arbitrarily shaped tensors. Here we stack matrices into
-    a 2 by 2 grid.
+    a 2 by 2 grid:
 
     >>> from numpy import ones
-    >>> a,b,c,d, = matrices('abcd')
+    >>> a, b, c, d = matrices('abcd')
     >>> X = stacklists([[a, b], [c, d]])
     >>> f = function([a, b, c, d], X)
     >>> x = ones((4, 4), 'float32')

theano/tensor/nnet/Conv3D.py

@@ -560,7 +560,7 @@ conv3D = Conv3D()
 :param b: bias, shape == (W.shape[0],)
 :param d: strides when moving the filter over the input(dx, dy, dt)
 
-:note: The order of dimensions do not correspond with the one in `conv2d`.
+:note: The order of dimensions does not correspond to the one in `conv2d`.
        This is for optimization.
 """
 

theano/tests/run_tests_in_batch.py

@@ -103,7 +103,7 @@ def main(stdout=None, stderr=None, argv=None, theano_nose=None,
                 theano_nose = path
                 break
     if theano_nose is None:
-        raise Exception("Not able to find theano-nose")
+        raise Exception("Unable to find theano-nose")
     if batch_size is None:
         batch_size = 100
     stdout_backup = sys.stdout