merge nc

HISTORY.txt

@@ -5,6 +5,141 @@
 Release Notes
 =============
 
+Theano 0.4.0 (2011-06-13)
+=========================
+
+Change in output memory storage for Ops:
+ If you implemented custom Ops, with either C or Python implementation,
+ this will concern you.
+
+ The contract for memory storage of Ops has been changed. In particular,
+ it is no longer guaranteed that output memory buffers are either empty,
+ or allocated by a previous execution of the same Op.
+
+ Right now, here is the situation:
+  * For Python implementation (perform), what is inside output_storage
+    may have been allocated from outside the perform() function, for
+    instance by another node (e.g., Scan) or the Mode. If that was the
+    case, the memory can be assumed to be C-contiguous (for the moment).
+  * For C implementations (c_code), nothing has changed yet.
+
+ In a future version, the content of the output storage, both for Python and C
+ versions, will either be NULL, or have the following guarantees:
+  * It will be a Python object of the appropriate Type (for a Tensor variable,
+    a numpy.ndarray, for a GPU variable, a CudaNdarray, for instance)
+  * It will have the correct number of dimensions, and correct dtype
+ However, its shape and memory layout (strides) will not be guaranteed.
+
+ When that change is made, the config flag DebugMode.check_preallocated_output
+ will help you find implementations that are not up-to-date.
+
+Deprecation:
+ * tag.shape attribute deprecated (#633)
+ * CudaNdarray_new_null is deprecated in favour of CudaNdarray_New
+ * Dividing integers with / is deprecated: use // for integer division, or
+   cast one of the integers to a float type if you want a float result (you may
+   also change this behavior with config.int_division).
+ * Removed (already deprecated) sandbox/compile module
+ * Removed (already deprecated) incsubtensor and setsubtensor functions,
+   inc_subtensor and set_subtensor are to be used instead.
+
+Bugs fixed:
+ * In CudaNdarray.__{iadd,idiv}__, when it is not implemented, return the error.
+ * THEANO_FLAGS='optimizer=None' now works as expected
+ * Fixed memory leak in error handling on GPU-to-host copy
+ * Fix relating specifically to Python 2.7 on Mac OS X
+ * infer_shape can now handle Python longs
+ * Trying to compute x % y with one or more arguments being complex now
+   raises an error.
+ * The output of random samples computed with uniform(..., dtype=...) is
+   guaranteed to be of the specified dtype instead of potentially being of a
+   higher-precision dtype.
+ * The perform() method of DownsampleFactorMax did not give the right result
+   when reusing output storage. This happen only if you use the Theano flags 
+   'linker=c|py_nogc' or manually specify the mode to be 'c|py_nogc'.
+
+Crash fixed:
+ * Work around a bug in gcc 4.3.0 that make the compilation of 2d convolution
+   crash.
+ * Some optimizations crashed when the "ShapeOpt" optimization was disabled.
+
+Optimization:
+ * Optimize all subtensor followed by subtensor.
+
+GPU:
+ * Move to the gpu fused elemwise that have other dtype then float32 in them
+   (except float64) if the input and output are float32.
+   * This allow to move elemwise comparisons to the GPU if we cast it to
+     float32 after that.
+ * Implemented CudaNdarray.ndim to have the same interface in ndarray.
+ * Fixed slowdown caused by multiple chained views on CudaNdarray objects
+ * CudaNdarray_alloc_contiguous changed so as to never try to free
+   memory on a view: new "base" property
+ * Safer decref behaviour in CudaNdarray in case of failed allocations
+ * New GPU implementation of tensor.basic.outer
+ * Multinomial random variates now available on GPU
+
+New features:
+ * ProfileMode
+    * profile the scan overhead
+    * simple hook system to add profiler
+    * reordered the output to be in the order of more general to more specific
+ * DebugMode now checks Ops with different patterns of preallocated memory,
+   configured by config.DebugMode.check_preallocated_output.
+ * var[vector of index] now work, (grad work recursively, the direct grad
+   work inplace, gpu work)
+    * limitation: work only of the outer most dimensions.
+ * New way to test the graph as we build it. Allow to easily find the source
+   of shape mismatch error:
+   `http://deeplearning.net/software/theano/tutorial/debug_faq.html#interactive-debugger`__
+ * cuda.root inferred if nvcc is on the path, otherwise defaults to
+   /usr/local/cuda
+ * Better graph printing for graphs involving a scan subgraph
+ * Casting behavior can be controlled through config.cast_policy,
+   new (experimental) mode.
+ * Smarter C module cache, avoiding erroneous usage of the wrong C
+   implementation when some options change, and avoiding recompiling the
+   same module multiple times in some situations.
+ * The "theano-cache clear" command now clears the cache more thoroughly.
+ * More extensive linear algebra ops (CPU only) that wrap scipy.linalg
+   now available in the sandbox.
+ * CUDA devices 4 - 16 should now be available if present.
+ * infer_shape support for the View op, better infer_shape support in Scan
+ * infer_shape supported in all case of subtensor
+ * tensor.grad now gives an error by default when computing the gradient
+   wrt a node that is disconnected from the cost (not in the graph, or
+   no continuous path from that op to the cost).
+ * New tensor.isnan and isinf functions.
+
+Documentation:
+ * Better commenting of cuda_ndarray.cu
+ * Fixes in the scan documentation: add missing declarations/print statements
+ * Better error message on failed __getitem__
+ * Updated documentation on profile mode
+ * Better documentation of testing on Windows
+ * Better documentation of the 'run_individual_tests' script
+
+Unit tests:
+ * More strict float comparaison by default
+ * Reuse test for subtensor of tensor for gpu tensor(more gpu test)
+ * Tests that check for aliased function inputs and assure appropriate copying
+   (#374)
+ * Better test of copies in CudaNdarray
+ * New tests relating to the new base pointer requirements
+ * Better scripts to run tests individually or in batches
+ * Some tests are now run whenever cuda is available and not just when it has
+   been enabled before
+ * Tests display less pointless warnings.
+
+Other:
+ * Correctly put the broadcast flag to True in the output var of
+   a Reshape op when we receive an int 1 in the new shape.
+ * pydotprint: high contrast mode is now the default, option to print
+   more compact node names.
+ * pydotprint: How trunk label that are too long.
+ * More compact printing (ignore leading "Composite" in op names)
+
+
 Theano 0.3.1 (2011-02-21)
 =========================
 

doc/cifarSC2011/advanced_theano.txt

 
 .. _advanced_theano:
 
-
 ***************
 Advanced Theano
 ***************
 
+Conditions
+----------
+**IfElse**
+
+- Build condition over symbolic variables.
+- IfElse Op takes a boolean condition and two variables to compute as input.
+- While Switch Op evaluates both 'output' variables, IfElse Op is lazy and only
+  evaluates one variable respect to the condition.
+
+**IfElse Example: Comparison with Switch**
+
+.. code-block:: python
+
+  from theano import tensor as T
+  from theano.lazycond import ifelse
+  import theano, time, numpy
+
+  a,b = T.scalars('a','b')
+  x,y = T.matrices('x','y')
+  
+  z_switch = T.switch(T.lt(a,b), T.mean(x), T.mean(y))
+  z_lazy = ifelse(T.lt(a,b), T.mean(x), T.mean(y))
+
+  f_switch = theano.function([a,b,x,y], z_switch, 
+                      mode=theano.Mode(linker='vm'))
+  f_lazyifelse = theano.function([a,b,x,y], z_lazy,
+                      mode=theano.Mode(linker='vm'))
+
+  val1 = 0.
+  val2 = 1.
+  big_mat1 = numpy.ones((10000,1000))
+  big_mat2 = numpy.ones((10000,1000))
+
+  n_times = 10
+
+  tic = time.clock()
+  for i in xrange(n_times):
+      f_switch(val1, val2, big_mat1, big_mat2)
+  print 'time spent evaluating both values %f sec'%(time.clock()-tic)
+
+  tic = time.clock()
+  for i in xrange(n_times):
+      f_lazyifelse(val1, val2, big_mat1, big_mat2)
+  print 'time spent evaluating one value %f sec'%(time.clock()-tic)
+
+IfElse Op spend less time (about an half) than Switch since it computes only
+one variable instead of both.
+
+>>> python ifelse_switch.py
+time spent evaluating both values 0.6700 sec
+time spent evaluating one value 0.3500 sec
+
+Note that IfElse condition is a boolean while Switch condition is a tensor, so
+Switch is more general.
+
+It is actually important to use  ``linker='vm'`` or ``linker='cvm'``,
+otherwise IfElse will compute both variables and take the same computation
+time as the Switch Op. The linker is not currently set by default to 'cvm' but
+it will be in a near future.
+
+Loops
+-----
+
+**Scan**
+
+- General form of **recurrence**, which can be used for looping.
+- **Reduction** and **map** (loop over the leading dimensions) are special cases of Scan
+- You 'scan' a function along some input sequence, producing an output at each time-step
+- The function can see the **previous K time-steps** of your function
+- ``sum()`` could be computed by scanning the z + x(i) function over a list, given an initial state of ``z=0``.
+- Often a for-loop can be expressed as a ``scan()`` operation, and ``scan`` is the closest that Theano comes to looping.
+- The advantage of using ``scan`` over for loops
+  
+  - The number of iterations to be part of the symbolic graph
+  - Minimizes GPU transfers if GPU is involved
+  - Compute gradients through sequential steps
+  - Slightly faster then using a for loop in Python with a compiled Theano function
+  - Can lower the overall memory usage by detecting the actual amount of memory needed
+
+**Scan Example: Computing pow(A,k)**
+
+.. code-block:: python
+
+  import theano
+  import theano.tensor as T
+
+  k = T.iscalar("k"); A = T.vector("A")
+
+  def inner_fct(prior_result, A): return prior_result * A
+  # Symbolic description of the result
+  result, updates = theano.scan(fn=inner_fct,
+                              outputs_info=T.ones_like(A),
+                              non_sequences=A, n_steps=k)
+
+  # Scan has provided us with A**1 through A**k.  Keep only the last
+  # value. Scan notices this and does not waste memory saving them.
+  final_result = result[-1]
+  
+  power = theano.function(inputs=[A,k], outputs=final_result,
+                        updates=updates)
+  
+  print power(range(10),2)
+  #[  0.   1.   4.   9.  16.  25.  36.  49.  64.  81.]
+
+
+**Scan Example: Calculating a Polynomial**
+
+.. code-block:: python
+
+  import theano
+  import theano.tensor as T
+
+  coefficients = theano.tensor.vector("coefficients")
+  x = T.scalar("x"); max_coefficients_supported = 10000
+
+  # Generate the components of the polynomial
+  full_range=theano.tensor.arange(max_coefficients_supported)
+  components, updates = theano.scan(fn=lambda coeff, power, free_var:
+                                     coeff * (free_var ** power),
+                                  outputs_info=None,
+                                  sequences=[coefficients, full_range],
+                                  non_sequences=x)
+  polynomial = components.sum()
+  calculate_polynomial = theano.function(inputs=[coefficients, x],
+                                       outputs=polynomial)
+
+  test_coeff = numpy.asarray([1, 0, 2], dtype=numpy.float32)
+  print calculate_polynomial(test_coeff, 3)
+  # 19.0
+
+
+
+Exercise 4
+-----------
+
+- Run both examples 
+- Modify and execute the polynomial example to have the reduction done by scan
+
+
+
 Compilation pipeline
 --------------------
 
-.. image:: pics/pipeline.png
+.. image:: ../hpcs2011_tutorial/pics/pipeline.png
    :width: 400 px
 
 Inplace optimization
@@ -113,7 +252,7 @@ Theano output:
       - Try the Theano flag floatX=float32
     """
 
-Exercise 4
+Exercise 5
 -----------
 
 - In the last exercises, do you see a speed up with the GPU?
@@ -167,19 +306,19 @@ Elemwise{Composite{neg,{sub,{{scalar_sigmoid,GT},neg}}}} [@183160204] ''   2
 
 >>> theano.printing.pydotprint_variables(prediction)
 
-.. image:: pics/logreg_pydotprint_prediction.png
+.. image:: ../hpcs2011_tutorial/pics/logreg_pydotprint_prediction.png
    :width: 800 px
 
 All pydotprint* requires graphviz and pydot
 
 >>> theano.printing.pydotprint(predict)
 
-.. image:: pics/logreg_pydotprint_predic.png
+.. image:: ../hpcs2011_tutorial/pics/logreg_pydotprint_predic.png
    :width: 800 px
 
 >>> theano.printing.pydotprint(train) # This is a small train example!
 
-.. image:: pics/logreg_pydotprint_train.png
+.. image:: ../hpcs2011_tutorial/pics/logreg_pydotprint_train.png
    :width: 1500 px
 
 
@@ -206,85 +345,6 @@ Debugging
   - Few optimizations
   - Run Python code (better error messages and can be debugged interactively in the Python debugger)
 
-
-Loops
------
-
-**Scan**
-
-- General form of **recurrence**, which can be used for looping.
-- **Reduction** and **map** (loop over the leading dimensions) are special cases of Scan
-- You 'scan' a function along some input sequence, producing an output at each time-step
-- The function can see the **previous K time-steps** of your function
-- ``sum()`` could be computed by scanning the z + x(i) function over a list, given an initial state of ``z=0``.
-- Often a for-loop can be expressed as a ``scan()`` operation, and ``scan`` is the closest that Theano comes to looping.
-- The advantage of using ``scan`` over for loops
-  
-  - The number of iterations to be part of the symbolic graph
-  - Minimizes GPU transfers if GPU is involved
-  - Compute gradients through sequential steps
-  - Slightly faster then using a for loop in Python with a compiled Theano function
-  - Can lower the overall memory usage by detecting the actual amount of memory needed
-
-**Scan Example: Computing pow(A,k)**
-
-.. code-block:: python
-
-  import theano
-  import theano.tensor as T
-
-  k = T.iscalar("k"); A = T.vector("A")
-
-  def inner_fct(prior_result, A): return prior_result * A
-  # Symbolic description of the result
-  result, updates = theano.scan(fn=inner_fct,
-                              outputs_info=T.ones_like(A),
-                              non_sequences=A, n_steps=k)
-
-  # Scan has provided us with A**1 through A**k.  Keep only the last
-  # value. Scan notices this and does not waste memory saving them.
-  final_result = result[-1]
-  
-  power = theano.function(inputs=[A,k], outputs=final_result,
-                        updates=updates)
-  
-  print power(range(10),2)
-  #[  0.   1.   4.   9.  16.  25.  36.  49.  64.  81.]
-
-
-**Scan Example: Calculating a Polynomial**
-
-.. code-block:: python
-
-  import theano
-  import theano.tensor as T
-
-  coefficients = theano.tensor.vector("coefficients")
-  x = T.scalar("x"); max_coefficients_supported = 10000
-
-  # Generate the components of the polynomial
-  full_range=theano.tensor.arange(max_coefficients_supported)
-  components, updates = theano.scan(fn=lambda coeff, power, free_var:
-                                     coeff * (free_var ** power),
-                                  outputs_info=None,
-                                  sequences=[coefficients, full_range],
-                                  non_sequences=x)
-  polynomial = components.sum()
-  calculate_polynomial = theano.function(inputs=[coefficients, x],
-                                       outputs=polynomial)
-
-  test_coeff = numpy.asarray([1, 0, 2], dtype=numpy.float32)
-  print calculate_polynomial(test_coeff, 3)
-  # 19.0
-
-
-
-Exercise 5
------------
-
-- Run both examples 
-- Modify and execute the polynomial example to have the reduction done by scan
-
 Known limitations
 -----------------
 
@@ -304,5 +364,3 @@ Known limitations
   - Disabling a few optimizations can speed up compilation
   - Usually too many nodes indicates a problem with the graph
 
-- Lazy evaluation in a branch (We will try to merge this summer)
-

doc/cifarSC2011/boot_camp_overview.txt

@@ -13,13 +13,13 @@ on the afternoons of Aug 2, 3, 5, and 6 (but not Aug 4th).
 Day 1
 -----
 
- * Show of hands - what is your background?
+* Show of hands - what is your background?
 
- * Python & Numpy in a nutshell
+* Python & Numpy in a nutshell
 
- * Theano basics
+* Theano basics
 
- * Quick tour through Deep Learning Tutorials (think about projects)
+* Quick tour through Deep Learning Tutorials (think about projects)
 
 .. :
     day 1:
@@ -38,35 +38,31 @@ Day 1
 Day 2
 -----
 
-  * Loop/Condition in Theano (10-20m)
+* Loop/Condition in Theano (10-20m)
 
-  * Propose/discuss projects
+* Propose/discuss projects
 
-  * Form groups and start projects!
+* Form groups and start projects!
 
 Day 3
 -----
 
- * Advanced Theano (30 minutes)
+* Advanced Theano (30 minutes)
 
-   * Debugging, profiling, compilation pipeline
+ * Debugging, profiling, compilation pipeline
 
- * Projects / General hacking / code-sprinting.
+* Projects / General hacking / code-sprinting.
 
 Day 4
 -----
 
- * *You choose* (we can split the group)
+* *You choose* (we can split the group)
 
-   * Extending Theano
+ * Extending Theano
 
-     * How to write an Op
+  * How to write an Op
 
-     * How to use pycuda code in Theano
+  * How to use pycuda code in Theano
 
-   * Projects / General hacking / code-sprinting.
+* Projects / General hacking / code-sprinting.
 
-Note - the schedule here is a guideline.
-We can adapt it in reponse to developments in the hands-on work.
-The point is for you to learn something about the practice of machine
-learning.

doc/cifarSC2011/extending_theano.txt

@@ -14,7 +14,7 @@ Theano graphs
 
 Inputs and Outputs are lists of Theano variables
 
-.. image:: pics/apply_node.png
+.. image:: ../hpcs2011_tutorial/pics/apply_node.png
     :width: 500 px
 
 Op contract

doc/cifarSC2011/index.txt

@@ -21,9 +21,10 @@ What does it do?
 It complements the Python numeric/scientific software stack (e.g. numpy, scipy,
 scikits, matplotlib, PIL.)
 
-Design and feature set has been driven by research in the machine learning group at the University of
-Montreal (Yoshua Bengio, Pascal Vincent, Douglas Eck).
-Result: a very good library for doing research in deep
+Design and feature set has been driven by machine learning research
+at the University of
+Montreal (groups of Yoshua Bengio, Pascal Vincent, Douglas Eck).
+The result is a very good library for doing research in deep
 learning and neural network training, and a flexible framework for
 many other models and algorithms in machine learning more generally.
 
@@ -53,7 +54,7 @@ calculations on other data structures.
 Contents
 --------
 
-The structured part of the course will be a walk-through of the following
+The structured part of these lab sessions will be a walk-through of the following
 material. Interleaved with this structured part will be blocks of time for
 individual or group work.  The idea is that you can try out Theano and get help
 from gurus on hand if you get stuck.

doc/conf.py

@@ -51,9 +51,9 @@ copyright = '2008--2011, LISA lab'
 # other places throughout the built documents.
 #
 # The short X.Y version.
-version = '0.4'
+version = '0.4.1'
 # The full version, including alpha/beta/rc tags.
-release = '0.4.0'
+release = '0.4.1rc1'
 
 # There are two options for replacing |today|: either, you set today to some
 # non-false value, then it is used:

doc/internal/how_to_release.txt

@@ -51,7 +51,7 @@ You will need to commit the previous changes, tag the resulting version, and
 push that into the original repository. The syntax is something like the
 following::
 
-    hg commit -m"modifications for 0.X release" setup.py doc/conf.py NEWS.txt
+    hg commit -m"modifications for 0.X release" setup.py doc/conf.py NEWS.txt HISTORY.txt theano/configdefaults.py doc/library/config.txt
     hg tag 0.X
     hg push
 

doc/library/config.txt

@@ -245,7 +245,7 @@ import theano and print the config variable, as in:
 
 .. attribute:: config.warn.ignore_bug_before
 
-    String value: 'None', 'all', '0.3', '0.4'
+    String value: 'None', 'all', '0.3', '0.4', '0.4.1'
 
     Default: 'None'
 

doc/tutorial/gradients.txt

@@ -98,7 +98,7 @@ In order to compute the Jacobian of some function ``y`` with respect to some
 parameter ``x`` we need to use the ``scan``. What we do is to loop over the
 entries in ``y`` and compute the gradient of ``y[i]`` with respect to ``x``.
 
-.. node::
+.. note::
     
     ``scan`` is a generic op in Theano that allows writting in a symbolic
     manner all kind of recurrent equations. While in principle, creating

doc/tutorial/index.txt

@@ -32,6 +32,7 @@ you out.
     symbolic_graphs
     modes
     aliasing
+    loop
     using_gpu
     pycuda
     shape_info

doc/tutorial/loop.txt

+.. _tutloop:
+
+====
+Loop
+====
+
+You can use :ref:`Scan <lib_scan>` to do all type of loop in Theano. All the documentation about it is in the library for now.

setup.py

@@ -47,8 +47,8 @@ AUTHOR_EMAIL        = "theano-dev@googlegroups.com"
 PLATFORMS           = ["Windows", "Linux", "Solaris", "Mac OS-X", "Unix"]
 MAJOR               = 0
 MINOR               = 4
-MICRO               = 0
-SUFFIX              = ""  # Should be blank except for rc's, betas, etc.
+MICRO               = 1
+SUFFIX              = "rc1"  # Should be blank except for rc's, betas, etc.
 ISRELEASED          = False
 
 VERSION             = '%d.%d.%d%s' % (MAJOR, MINOR, MICRO, SUFFIX)

theano/compile/function_module.py

@@ -814,7 +814,7 @@ def insert_deepcopy(env, wrapped_inputs, wrapped_outputs):
 
     assert len(wrapped_inputs) == len(env.inputs)
     assert len(wrapped_outputs) == len(env.outputs)
-
+    reason = "insert_deepcopy"
     updated_env_inputs = [env_i for i, env_i in zip(wrapped_inputs, env.inputs) if getattr(i, 'update', False)]
 
     # We can't use env.inputs as this don't include Constant Value.
@@ -830,9 +830,11 @@ def insert_deepcopy(env, wrapped_inputs, wrapped_outputs):
             # and not(wrapped_outputs[i].borrow and wrapped_outputs[j].borrow):
             if env.outputs[j] in views_of_output_i:
                 if wrapped_outputs[i].borrow and wrapped_outputs[j].borrow:
-                    env.change_input('output',i, view_op(env.outputs[i]))
+                    env.change_input('output',i, view_op(env.outputs[i]),
+                                     reason=reason)
                 else:
-                    env.change_input('output', i, deep_copy_op(env.outputs[i]))
+                    env.change_input('output', i, deep_copy_op(env.outputs[i]),
+                                     reason=reason)
                 copied = True
                 break
 
@@ -850,16 +852,20 @@ def insert_deepcopy(env, wrapped_inputs, wrapped_outputs):
                     if input_j in env.inputs:
                         j = env.inputs.index(input_j)
                         if wrapped_outputs[i].borrow and wrapped_inputs[j].borrow:
-                            env.change_input('output',i, view_op(env.outputs[i]))
+                            env.change_input('output',i, view_op(env.outputs[i]),
+                                             reason="insert_deepcopy")
                             break
                         else:
-                            env.change_input('output', i, deep_copy_op(env.outputs[i]))
+                            env.change_input('output', i, deep_copy_op(env.outputs[i]),
+                                             reason="insert_deepcopy")
                             break
                     elif wrapped_outputs[i].borrow:
-                        env.change_input('output',i, view_op(env.outputs[i]))
+                        env.change_input('output',i, view_op(env.outputs[i]),
+                                         reason="insert_deepcopy")
                         break
                     else:
-                        env.change_input('output', i, deep_copy_op(env.outputs[i]))
+                        env.change_input('output', i, deep_copy_op(env.outputs[i]),
+                                         reason="insert_deepcopy")
                         break
 
 NODEFAULT = ['NODEFAULT']

theano/configdefaults.py

@@ -223,7 +223,7 @@ AddConfigVar('numpy.seterr_invalid',
 ###
 AddConfigVar('warn.ignore_bug_before',
              "If 'None', we warn about all Theano bugs found by default. If 'all', we don't warn about Theano bugs found by default. If a version, we print only the warnings relative to Theano bugs found after that version. Warning for specific bugs can be configured with specific [warn] flags.",
-             EnumStr('None', 'all', '0.3','0.4', allow_override=False),
+             EnumStr('None', 'all', '0.3','0.4', '0.4.1',allow_override=False),
              in_c_key=False)
 
 default_0_3 = True

theano/gof/tests/test_compute_test_value.py

@@ -283,7 +283,10 @@ class TestComputeTestValue(unittest.TestCase):
                         n_steps=k)
                 assert False
             except ValueError, e:
-                assert e.message.startswith("shape mismatch")
+                # The first message is for numpy before 1.6
+                # The second is a new message in numpy 1.6
+                assert (e.message.startswith("shape mismatch") or
+                        e.message.startswith("operands could not be broadcast together with shapes"))
 
         finally:
             theano.config.compute_test_value = orig_compute_test_value

theano/gof/tests/test_lazy.py

@@ -84,11 +84,6 @@ class IfElseIfElseIf(PureOp):
 class NotImplementedOp(PureOp):
     class E(Exception): pass
 
-    def __eq__(self, other):
-        return type(self) == type(other)
-    def __hash__(self):
-        return hash(type(self))
-
     def make_node(self, x):
         return Apply(self, [x], [x.type()])
     def make_thunk(self, node, storage_map, compute_map, no_recycling):

theano/misc/check_blas.py

@@ -65,7 +65,9 @@ def execute(execute=True, verbose=True):
         t1=time.time()
     if verbose and execute:
         print
-        print 'this execution time took %.2fs'%(t1-t0)
+        print 'This execution time took %.2fs'%(t1-t0)
+        print
+        print 'Try to run this script a few times. Experience show that the first time is not as fast as followings call. The difference is not big, but consistent.'
     return t1-t0
 
 
@@ -103,7 +105,7 @@ if __name__ == "__main__":
             * manually compiled numpy and ATLAS with 2 threads
             * goto 1.26 with 1, 2, 4 and 8 threads.
             * goto2 1.13 compiled with multiple thread enabled.
-            
+
                           Xeon   Xeon   Xeon  Core2 i7    i7     Xeon   Xeon
         lib/nb threads    E5345  E5430  E5450 E8500 930   950    X5560  X5550
 
@@ -139,6 +141,8 @@ if __name__ == "__main__":
         M2070/3.2         0.32s
         GTX470/3.0        0.34s
         GTX285/3.0        0.40s
+        C1060/3.2         0.46s
+        GTX550Ti/4.0      0.57s
         GT220/3.2RC       3.80s
         8500GT/3.0       10.68s
         """

theano/misc/tests/test_pycuda_theano_simple.py

@@ -19,7 +19,7 @@ if not theano.misc.pycuda_init.pycuda_available:
 
 if cuda_ndarray.cuda_available == False:
     from nose.plugins.skip import SkipTest
-    raise SkipTest('Optional package cuda disabled')
+    raise SkipTest('Optional theano package cuda disabled')
 
 import pycuda
 import pycuda.driver as drv

theano/printing.py

@@ -424,6 +424,9 @@ def pydotprint(fct, outfile=None,
                 file to which the name of the scan op is concatenated and
                 the index in the toposort of the scan.
                 This index can be printed in the graph with the option with_ids.
+    :param var_with_name_simple: If true and a variable have a name,
+                we will print only the variable name.
+                Otherwise, we concatenate the type to the var name.
 
     In the graph, box are an Apply Node(the execution of an op) and ellipse are variable.
     If variable have name they are used as the text(if multiple var have the same name, they will be merged in the graph).

theano/sandbox/blas_ger.py

-"""
-
-Optimization to specialize gemm -> ger are not written
-
-Scipy implementation is not written
-
-We need to call scipy.linalg.blas.[cf]blas.[sdcz]ger here in order not to lose speed against the old Outer op.
-Here is the scipy signature: ger(alpha,x,y,incx=1,incy=1,a=0.0,overwrite_x=1,overwrite_y=1,overwrite_a=0)
-
-http://www.scipy.org/doc/api_docs/SciPy.lib.blas.info.html
-
-
-
-C implementation is not written.
-
-Tests are not written.
-"""
-class GER(Op):
-    """
-    General rank-1 update
-    A <- A + a x' y
-
-    For matrix A, vectors x, y, and scalar a.
-    """
-    def __init__(self, inplace):
-        self.inplace = bool(inplace)
-        if self.inplace:
-            self.destroy_map = {0: [0]}
-
-    def __hash__(self):
-        return hash((type(self), self.inplace))
-
-    def __eq__(self, other):
-        return hash((type(self), self.inplace))
-
-    def make_node(self, *inputs):
-        inputs = map(as_tensor_variable, inputs)
-        A, a, x, y = inputs
-
-        nx = x.type.ndim
-        ny = y.type.ndim
-
-        if nx != 1: raise TypeError('non-vector arg0 to outer()', x)
-        if ny != 1: raise TypeError('not-vector arg1 to outer()', y)
-
-        if A.dtype != a.dtype:
-            raise TypeError('dtype mismatch', (A.dtype, a.dtype))
-        if A.dtype != x.dtype:
-            raise TypeError('dtype mismatch', (A.dtype, x.dtype))
-        if A.dtype != y.dtype:
-            raise TypeError('dtype mismatch', (A.dtype, y.dtype))
-
-        return Apply(self, inputs, [A.type()])
-
-    def perform(self, node, inp, out):
-        A, a, x, y = inp
-        if not self.inplace:
-            A = A.copy()
-        A += a * numpy.outer(x, y)
-        out[0][0] = A
-
-    # grad not needed because this is put in during optimization
-
-    def __str__(self):
-        return "GER"

theano/scan_module/scan.py

@@ -297,6 +297,14 @@ def scan( fn
         loop are done (see ``theano.function`` for details about
         possible values and their meaning).
 
+    :param profile:
+        Flag or string. If true, or different from the empty string, a
+        profile object will be created and attached to the inner graph of
+        scan. In case ``profile`` is True, the profile object will have the
+        name of the scan instance, otherwise it will have the passed string.
+        Profile object collect (and print) information only when running the
+        inner graph with the new cvm linker ( with default modes,
+        other linkers this argument is useless)
 
     :rtype: tuple
     :return: tuple of the form (outputs, updates); ``outputs`` is either a

theano/scan_module/scan_opt.py

@@ -17,17 +17,16 @@ import numpy
 import sys
 
 import theano
-from theano import tensor, scalar
-from theano.tensor import opt, TensorType, get_constant_value
+from theano import tensor
+from theano.tensor import opt, get_constant_value
 from theano import gof
 from theano.compile import optdb
-from theano.gof.opt import EquilibriumOptimizer
 from theano import config
 from theano.compile.function_module import deep_copy_op
 
 import scan_op
 import scan_utils
-from scan_utils import clone, equal_computations, find_up, scan_args
+from scan_utils import equal_computations, find_up, scan_args
 from theano.gof.opt import pre_constant_merge, pre_greedy_local_optimizer
 
 # Logging function for sending warning or info
@@ -454,12 +453,12 @@ class ScanSaveMem(gof.Optimizer):
                     if i > op.n_mit_mot:
                         try:
                             length = shape_of[out][0]
-                        except:
+                        except Exception:
                             length = node.inputs[0] + init_l[i]
                     else:
                         try:
                             length = shape_of[out][0]
-                        except:
+                        except Exception:
                             length = out.shape[0]
                     cf_slice = tensor.basic.get_canonical_form_slice(
                                                     this_slice[0], length)
@@ -556,7 +555,7 @@ class ScanSaveMem(gof.Optimizer):
                     else:
                         try:
                             length = shape_of[out][0]
-                        except:
+                        except Exception:
                             length = out.shape[0]
                     cf_slice = tensor.basic.get_canonical_form_slice(
                                                     this_slice[0],length)
@@ -1124,5 +1123,3 @@ scan_seqopt.register('scanOp_merge_inouts',
                      3,
                      'fast_run',
                      'scan')
-
-

theano/tensor/basic.py

@@ -158,7 +158,7 @@ def as_tensor_variable(x, name=None, ndim=None):
     except TypeError:
         try:
             str_x = str(x)
-        except:
+        except Exception, e:
             str_x = repr(x)
         raise TypeError("Cannot convert %s to TensorType" % str_x, type(x))
 
@@ -340,7 +340,7 @@ def constant_or_value(x, rtype, name=None, ndim=None, dtype=None):
         else:
             # leave the shape out of the type
             return rtype(TensorType(dtype = x_.dtype, broadcastable = bcastable), x_, name=name)
-    except:
+    except Exception:
         raise TypeError("Could not convert %s to TensorType" % x, type(x))
 
 def constant(x, name=None, ndim=None, dtype=None):
@@ -425,7 +425,7 @@ def get_constant_value(v):
         try:
             numpy.complex(data) #works for all numeric scalars
             return data
-        except:
+        except Exception:
             raise TypeError('v.data is non-numeric, non-scalar, or has more than one unique value', v)
     if v.owner:
         if isinstance(v.owner.op, Alloc):
@@ -1361,7 +1361,7 @@ class TensorConstantSignature(tuple):
             return False
         try:
             (t0, d0), (t1,d1) = self, other
-        except:
+        except Exception, e:
             return False
         #N.B. compare shape to ensure no broadcasting in ==
         if t0 != t1 or d0.shape != d1.shape:
@@ -1994,7 +1994,7 @@ def max(x, axis='DEFAULT'):
     try:
         const = get_constant_value(axis)
         return CAReduce(scal.maximum,list(const))(x)
-    except:
+    except Exception:
         return max_and_argmax(x,axis)[0]
 
 @constructor
@@ -2873,7 +2873,7 @@ def extract_constant(x):
     '''
     try:
         x = get_constant_value(x)
-    except:
+    except Exception:
         pass
     if isinstance(x, scal.ScalarVariable):
         if x.owner and isinstance(x.owner.op, ScalarFromTensor):
@@ -4398,7 +4398,7 @@ class Reshape(Op):
                     ', should be %i' % (len(shp), self.ndim), shp)
         try:
             out[0] = numpy.reshape(x, shp)
-        except:
+        except Exception, e:
             raise ValueError('Cannot reshape input of shape %s to shape %s' % (x.shape,shp))
     def grad(self, inp, grads):
         x, shp = inp
@@ -4593,7 +4593,7 @@ class ARange(Op):
             try:
                 v = get_constant_value(var)
                 return numpy.all(v == value)
-            except:
+            except Exception:
                 pass
             return False
 

theano/tensor/nnet/ConvTransp3D.py

@@ -12,7 +12,7 @@ class ConvTransp3D(theano.Op):
         return hash(type(self))
 
     def c_code_cache_version(self):
-        return (1,)
+        return (2,)
 
     def make_node(self, W, b, d, H, RShape = None):
         """
@@ -266,11 +266,11 @@ class ConvTransp3D(theano.Op):
 
                                        for (int i = 0; i < batchSize; i++) {
                                         for (int r = 0; r < videoHeight; r++) {
-                                         const int frc = std::max(0.0, ceil(float(r-filterHeight+1)/float(dr)));
+                                         const int frc = (int)std::max(0.0f, ceilf(float(r-filterHeight+1)/float(dr)));
                                          for (int c = 0; c < videoWidth; c++) {
-                                          const int fcc = std::max(0.0, ceil(float(c-filterWidth +1)/float(dc)));
+                                          const int fcc = (int)std::max(0.0f, ceilf(float(c-filterWidth +1)/float(dc)));
                                           for (int t = 0; t < videoDur; t++) {
-                                           const int ftc = std::max(0.0, ceil(float(t-filterDur +1)  /float(dt)));
+                                           const int ftc = (int)std::max(0.0f, ceilf(float(t-filterDur +1)  /float(dt)));
 
                                            long long Rpost = i * %(R)s->strides[0] + r * %(R)s->strides[1] + c * %(R)s->strides[2] + t * %(R)s->strides[3];
 

theano/tensor/tensor_grad.py

@@ -117,7 +117,7 @@ def Rop(f, wrt, eval_points):
         return rval
 
 
-def Lop(f, wrt, eval_points, consider_constant=[], warn_type=False,
+def Lop(f, wrt, eval_points, consider_constant=None, warn_type=False,
          disconnected_inputs='raise'):
     """
     Computes the L operation on `f` wrt to `wrt` evaluated at points given
@@ -140,6 +140,8 @@ def Lop(f, wrt, eval_points, consider_constant=[], warn_type=False,
         indices that specify both the position within a list and all
         coordinates of the tensor element in the last
     """
+    if consider_constant is None:
+        consider_constant = []
 
     if not isinstance(f, TensorVariable):
         raise TypeError('In tensor.Lop(), cost argument should be a TensorVariable.', f)
@@ -155,7 +157,6 @@ def Lop(f, wrt, eval_points, consider_constant=[], warn_type=False,
             list(inputs) + list(consider_constant),
             warn_type=warn_type)
 
-
     # Note : If p is not in gmap there can be several reasons, among which
     # is the fact that p might not be part of the computational graph. A
     # simple example is that for a+b for e.g. a[0] is not part of the graph,
@@ -196,7 +197,7 @@ def Lop(f, wrt, eval_points, consider_constant=[], warn_type=False,
 # Gradient
 #########################
 
-def grad(cost, wrt, g_cost=None, consider_constant=[], warn_type=False,
+def grad(cost, wrt, g_cost=None, consider_constant=None, warn_type=False,
          disconnected_inputs='raise'):
     """
     :type cost: Scalar (0-dimensional) `Variable`
@@ -228,6 +229,9 @@ def grad(cost, wrt, g_cost=None, consider_constant=[], warn_type=False,
     `theano.gradient.grad_sources_inputs``.
 
     """
+    if consider_constant is None:
+        consider_constant = []
+
     if not isinstance(cost, TensorVariable):
         raise TypeError('In tensor.grad(), cost argument should be a TensorVariable.', cost)
 

theano/tensor/tests/test_basic.py

@@ -7,6 +7,7 @@ import unittest
 from nose.plugins.skip import SkipTest
 import numpy
 from numpy.testing import dec
+from numpy.testing.noseclasses import KnownFailureTest
 
 from theano.tensor import *
 from theano.tensor import basic as tensor # for hidden symbols
@@ -4736,6 +4737,22 @@ class test_arithmetic_cast(unittest.TestCase):
                                     config.int_division == 'floatX'):
                                     assert theano_dtype == config.floatX
                                     continue
+                                if (cfg == 'numpy+floatX' and
+                                    a_type == 'complex128' and
+                                    b_type == 'float32' and
+                                    combo == ('scalar', 'array') and
+                                    numpy.__version__ == '1.6.0' and
+                                    theano_dtype == 'complex128' and
+                                    numpy_dtypes == ['complex64',
+                                                     'complex64']):
+                                    # In numpy 1.6.0 adding a complex128 with
+                                    # a float32 may result in a complex64. This
+                                    # may be a bug (investigation is currently
+                                    # in progress), so in the meantime we just
+                                    # mark this test as a known failure.
+                                    raise KnownFailureTest('Known issue with '
+                                            'numpy 1.6.0, see #761')
+
                                 # In any other situation: something wrong is
                                 # going on!
                                 assert False

theano/tensor/tests/test_elemwise.py

@@ -82,7 +82,7 @@ class test_Broadcast(unittest.TestCase):
 
             self.assertTrue((f(xv, yv) == zv).all())
 
-            #test CAReduce.infer_shape
+            #test Elemwise.infer_shape
             #the Shape op don't implement c_code!
             if isinstance(linker,gof.PerformLinker):
                 x = TensorType('float64', [(entry == 1) for entry in xsh])('x')
@@ -111,7 +111,7 @@ class test_Broadcast(unittest.TestCase):
             f(xv, yv)
 
             self.assertTrue((xv == zv).all())
-            #test CAReduce.infer_shape
+            #test Elemwise.infer_shape
             #the Shape op don't implement c_code!
             if isinstance(linker,gof.PerformLinker):
                 x = TensorType('float64', [(entry == 1) for entry in xsh])('x')