Release 0.10.0beta2

8fcb9eef · notoraptor · 9b3f6351 · 8fcb9eef · 8fcb9eef · 8fcb9eef
--- a/.mailmap
+++ b/.mailmap
@@ -167,6 +167,7 @@ Markus Roth <markus.roth@herr-biber.de> Markus Roth <mail@rothmark.us>
 Mathieu Germain <mathieu.germain@gmail.com> Mathieu Germain <mathieu.germain2@usherbrooke.ca>
 Mehdi Mirza <memirzamo@gmail.com> Mehdi Mirza <memimo@users.noreply.github.com>
 Mehdi Mirza <memirzamo@gmail.com> memimo <memirzamo@gmail.com>
+Mohammed Affan <affanv14@gmail.com> affan <affanv14@gmail.com>
 Mohammed Affan <affanv14@gmail.com> affanv14 <affanv14@gmail.com>
 Moslem Kazemi <moslemk@gmail.com> Moslem Kazemi <moslemk@users.noreply.github.com>
 Moslem Kazemi <moslemk@gmail.com> Mo <moslemk@gmail.com>

--- a/HISTORY.txt
+++ b/HISTORY.txt
@@ -5,6 +5,187 @@
 Old Release Notes
 =================

+Theano 0.10.0beta1 (9th of August, 2017)
+========================================
+
+This release contains a lot of bug fixes, improvements and new features to prepare the upcoming release candidate.
+
+We recommend that every developer updates to this version.
+
+Highlights:
+ - Moved Python 3.* minimum supported version from 3.3 to 3.4
+ - Replaced deprecated package ``nose-parameterized`` with up-to-date package ``parameterized`` for Theano requirements
+ - Theano now internally uses ``sha256`` instead of ``md5`` to work on systems that forbide ``md5`` for security reason
+ - Removed old GPU backend ``theano.sandbox.cuda``. New backend ``theano.gpuarray`` is now the official GPU backend
+ - Support more debuggers for ``PdbBreakpoint``
+
+ - Scan improvements
+
+   - Speed up Theano scan compilation and gradient computation
+   - Added meaningful message when missing inputs to scan
+
+ - Speed up graph toposort algorithm
+ - Faster C compilation by massively using a new interface for op params
+ - Faster optimization step
+ - Documentation updated and more complete
+ - Many bug fixes, crash fixes and warning improvements
+
+A total of 65 people contributed to this release since 0.9.0, see list below.
+
+Interface changes:
+ - Merged duplicated diagonal functions into two ops: ``ExtractDiag`` (extract a diagonal to a vector),
+   and ``AllocDiag`` (set a vector as a diagonal of an empty array)
+ - Renamed ``MultinomialWOReplacementFromUniform`` to ``ChoiceFromUniform``
+
+ - Removed or deprecated Theano flags:
+
+   - ``cublas.lib``
+   - ``cuda.enabled``
+   - ``enable_initial_driver_test``
+   - ``gpuarray.sync``
+   - ``home``
+   - ``lib.cnmem``
+   - ``nvcc.*`` flags
+   - ``pycuda.init``
+
+ - Changed ``grad()`` method to ``L_op()`` in ops that need the outputs to compute gradient
+
+Convolution updates:
+ - Extended Theano flag ``dnn.enabled`` with new option ``no_check`` to help speed up cuDNN importation
+ - Implemented separable convolutions
+ - Implemented grouped convolutions
+
+GPU:
+ - Prevent GPU initialization when not required
+ - Added disk caching option for kernels
+ - Added method ``my_theano_function.sync_shared()`` to help synchronize GPU Theano functions
+ - Added useful stats for GPU in profile mode
+ - Added Cholesky op based on ``cusolver`` backend
+ - Added GPU ops based on `magma library <http://icl.cs.utk.edu/magma/software/>`_:
+   SVD, matrix inverse, QR, cholesky and eigh
+ - Added ``GpuCublasTriangularSolve``
+ - Added atomic addition and exchange for ``long long`` values in ``GpuAdvancedIncSubtensor1_dev20``
+ - Support log gamma function for all non-complex types
+ - Support GPU SoftMax in both OpenCL and CUDA
+ - Support offset parameter ``k`` for ``GpuEye``
+ - ``CrossentropyCategorical1Hot`` and its gradient are now lifted to GPU
+
+ - Better cuDNN support
+
+   - Official support for ``v5.*`` and ``v6.*``
+   - Better support and loading on Windows and Mac
+   - Support cuDNN v6 dilated convolutions
+   - Support cuDNN v6 reductions
+   - Added new Theano flags ``cuda.include_path``, ``dnn.base_path`` and ``dnn.bin_path``
+     to help configure Theano when CUDA and cuDNN can not be found automatically.
+
+ - Updated ``float16`` support
+
+   - Added documentation for GPU float16 ops
+   - Support ``float16`` for ``GpuGemmBatch``
+   - Started to use ``float32`` precision for computations that don't support ``float16`` on GPU
+
+New features:
+ - Added a wrapper for `Baidu's CTC <https://github.com/baidu-research/warp-ctc>`_ cost and gradient functions
+ - Added scalar and elemwise CPU ops for modified Bessel function of order 0 and 1 from ``scipy.special``.
+ - Added Scaled Exponential Linear Unit (SELU) activation
+ - Added sigmoid_binary_crossentropy function
+ - Added tri-gamma function
+ - Added modes ``half`` and ``full`` for ``Images2Neibs`` ops
+ - Implemented gradient for ``AbstractBatchNormTrainGrad``
+ - Implemented gradient for matrix pseudoinverse op
+ - Added new prop `replace` for ``ChoiceFromUniform`` op
+ - Added new prop ``on_error`` for CPU ``Cholesky`` op
+ - Added new Theano flag ``deterministic`` to help control how Theano optimize certain ops that have deterministic versions.
+   Currently used for subtensor Ops only.
+ - Added new Theano flag ``cycle_detection`` to speed-up optimization step by reducing time spending in inplace optimizations
+ - Added new Theano flag ``check_stack_trace`` to help check the stack trace during optimization process
+ - Added new Theano flag ``cmodule.debug`` to allow a debug mode for Theano C code. Currently used for cuDNN convolutions only.
+
+Others:
+ - Added deprecation warning for the softmax and logsoftmax vector case
+ - Added a warning to announce that C++ compiler will become mandatory in next Theano release ``0.11``
+
+Other more detailed changes:
+ - Removed useless warning when profile is manually disabled
+ - Added tests for abstract conv
+ - Added options for `disconnected_outputs` to Rop
+ - Removed ``theano/compat/six.py``
+ - Removed ``COp.get_op_params()``
+ - Support of list of strings for ``Op.c_support_code()``, to help not duplicate support codes
+ - Macro names provided for array properties are now standardized in both CPU and GPU C codes
+ - Started to move C code files into separate folder ``c_code`` in every Theano module
+ - Many improvements for Travis CI tests (with better splitting for faster testing)
+ - Many improvements for Jenkins CI tests: daily testings on Mac and Windows in addition to Linux
+
+Commiters since 0.9.0:
+ - Frederic Bastien
+ - Arnaud Bergeron
+ - amrithasuresh
+ - João Victor Tozatti Risso
+ - Steven Bocco
+ - Pascal Lamblin
+ - Mohammed Affan
+ - Reyhane Askari
+ - Alexander Matyasko
+ - Simon Lefrancois
+ - Shawn Tan
+ - Thomas George
+ - Faruk Ahmed
+ - Zhouhan LIN
+ - Aleksandar Botev
+ - jhelie
+ - xiaoqie
+ - Tegan Maharaj
+ - Matt Graham
+ - Cesar Laurent
+ - Gabe Schwartz
+ - Juan Camilo Gamboa Higuera
+ - AndroidCloud
+ - Saizheng Zhang
+ - vipulraheja
+ - Florian Bordes
+ - Sina Honari
+ - Vikram
+ - erakra
+ - Chiheb Trabelsi
+ - Shubh Vachher
+ - Daren Eiri
+ - Gijs van Tulder
+ - Laurent Dinh
+ - Mohamed Ishmael Diwan Belghazi
+ - mila
+ - Jeff Donahue
+ - Ramana Subramanyam
+ - Bogdan Budescu
+ - Ghislain Antony Vaillant
+ - Jan Schlüter
+ - Xavier Bouthillier
+ - fo40225
+ - Aarni Koskela
+ - Adam Becker
+ - Adam Geitgey
+ - Adrian Keet
+ - Adrian Seyboldt
+ - Andrei Costinescu
+ - Anmol Sahoo
+ - Chong Wu
+ - Holger Kohr
+ - Jayanth Koushik
+ - Jenkins
+ - Lilian Besson
+ - Lv Tao
+ - Michael Manukyan
+ - Murugesh Marvel
+ - NALEPA
+ - Ubuntu
+ - Zotov Yuriy
+ - dareneiri
+ - lrast
+ - morrme
+ - yikang
+
+
 Theano 0.9.0 (20th of March, 2017)
 ==================================

@@ -1365,7 +1546,7 @@ Interface Change:
 New Interface (reuse existing functionality):
 * tensor_var.sort() as a shortcut for theano.tensor.sort. (Jeremiah Lowin)
   We where already doing this for argsort.
- * Add theano.tensor.take() and a_tensor_var.take() to support NumPy syntax. (abalkin) 
+ * Add theano.tensor.take() and a_tensor_var.take() to support NumPy syntax. (abalkin)
 * Add a_tensor_variable.{dot,std,argmin,argmax,argsort,clip,conj,conjugate,repeat,round,trace,real,imag}. (abalkin)

 New debug feature:
@@ -1403,7 +1584,7 @@ Crash Fixes:
 * Make CrossentropySoftmax1HotWithBiasDx and CrossentropySoftmaxArgmax1HotWithBias support uint* dtype. (Frederic B., reported by Mark Fenner)
 * Fix GpuSoftmax and GpuSoftmaxWithBias crash on GTX285. (Frederic B.)
 * Fix crash due to a race condition when importing theano. (Ian G.)
- * Fix crash from path problem with `theano-nose --batch`. (Abalkin) 
+ * Fix crash from path problem with `theano-nose --batch`. (Abalkin)
 * Fix crash with tensor.roll(Var, iscalar). (Frederic B., reported by Jeremiah Lowin)
 * Fix compilation crash with llvm on Mac. (Abalkin)
 * Fix the grad of Scan that told wrongly that there is no connection between cost and parameters. (Razvan P.)
@@ -2268,7 +2449,7 @@ Optimizations:
 * IncSubtensor(x, zeros, idx) -> x
 * SetSubtensor(x, x[idx], idx) -> x (when x is a constant)
 * subtensor(alloc,...) -> alloc
- * Many new scan optimization 
+ * Many new scan optimization

   * Lower scan execution overhead with a Cython implementation
   * Removed scan double compilation (by using the new Op.make_thunk mechanism)
@@ -2331,10 +2512,10 @@ Deprecation (will be removed in Theano 0.5, warning generated if you use them):

        [outputs], [updates], [condition]. One can skip any of the three if not
        used, but the order has to stay unchanged.
- * tensor.grad(cost, wrt) will return an object of the "same type" as wrt 
+ * tensor.grad(cost, wrt) will return an object of the "same type" as wrt
   (list/tuple/TensorVariable).

-   * Currently tensor.grad return a type list when the wrt is a list/tuple of 
+   * Currently tensor.grad return a type list when the wrt is a list/tuple of
     more than 1 element.

 Decrecated in 0.4.0(Reminder, warning generated if you use them):
@@ -2445,7 +2626,7 @@ Bugs fixed:
   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 
+   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:

--- a/NEWS.txt
+++ b/NEWS.txt
@@ -3,182 +3,61 @@ Release Notes
 =============


-Theano 0.10.0beta1 (9th of August, 2017)
-========================================
+Theano 0.10.0beta2 (25th of August, 2017)
+=========================================

-This release contains a lot of bug fixes, improvements and new features to prepare the upcoming release candidate.
+This release contains new features, improvements and bug fixes to prepare the upcoming release candidate.

 We recommend that every developer updates to this version.

 Highlights:
- - Moved Python 3.* minimum supported version from 3.3 to 3.4
- - Replaced deprecated package ``nose-parameterized`` with up-to-date package ``parameterized`` for Theano requirements
- - Theano now internally uses ``sha256`` instead of ``md5`` to work on systems that forbide ``md5`` for security reason
- - Removed old GPU backend ``theano.sandbox.cuda``. New backend ``theano.gpuarray`` is now the official GPU backend
- - Support more debuggers for ``PdbBreakpoint``
+ - Support NumPy ``1.13``
+ - Faster optimization step with new destroy handler
+ - Updated documentation
+ - Bug fixes, crash fixes and warning improvements

- - Scan improvements
-
-   - Speed up Theano scan compilation and gradient computation
-   - Added meaningful message when missing inputs to scan
-
- - Speed up graph toposort algorithm
- - Faster C compilation by massively using a new interface for op params
- - Faster optimization step
- - Documentation updated and more complete
- - Many bug fixes, crash fixes and warning improvements
-
-A total of 65 people contributed to this release since 0.9.0, see list below.
-
-Interface changes:
- - Merged duplicated diagonal functions into two ops: ``ExtractDiag`` (extract a diagonal to a vector),
-   and ``AllocDiag`` (set a vector as a diagonal of an empty array)
- - Renamed ``MultinomialWOReplacementFromUniform`` to ``ChoiceFromUniform``
-
- - Removed or deprecated Theano flags:
-
-   - ``cublas.lib``
-   - ``cuda.enabled``
-   - ``enable_initial_driver_test``
-   - ``gpuarray.sync``
-   - ``home``
-   - ``lib.cnmem``
-   - ``nvcc.*`` flags
-   - ``pycuda.init``
-
- - Changed ``grad()`` method to ``L_op()`` in ops that need the outputs to compute gradient
+A total of 14 people contributed to this release since 0.10.0beta1, see list below.

 Convolution updates:
- - Extended Theano flag ``dnn.enabled`` with new option ``no_check`` to help speed up cuDNN importation
- - Implemented separable convolutions
- - Implemented grouped convolutions
+ - Added 3D separable convolutions
+ - Added 3D grouped convolutions
+ - Removed old ``conv3d`` interface
+ - Deprecated old ``conv2d`` interface
+ - Updated ``conv`` documentation

 GPU:
- - Prevent GPU initialization when not required
- - Added disk caching option for kernels
- - Added method ``my_theano_function.sync_shared()`` to help synchronize GPU Theano functions
- - Added useful stats for GPU in profile mode
- - Added Cholesky op based on ``cusolver`` backend
- - Added GPU ops based on `magma library <http://icl.cs.utk.edu/magma/software/>`_:
-   SVD, matrix inverse, QR, cholesky and eigh
- - Added ``GpuCublasTriangularSolve``
- - Added atomic addition and exchange for ``long long`` values in ``GpuAdvancedIncSubtensor1_dev20``
- - Support log gamma function for all non-complex types
- - Support GPU SoftMax in both OpenCL and CUDA
- - Support offset parameter ``k`` for ``GpuEye``
- - ``CrossentropyCategorical1Hot`` and its gradient are now lifted to GPU
-
- - Better cuDNN support
+ - Added a meta-optimizer to select the fastest GPU implementations for convolutions

-   - Official support for ``v5.*`` and ``v6.*``
-   - Better support and loading on Windows and Mac
-   - Support cuDNN v6 dilated convolutions
-   - Support cuDNN v6 reductions
-   - Added new Theano flags ``cuda.include_path``, ``dnn.base_path`` and ``dnn.bin_path``
-     to help configure Theano when CUDA and cuDNN can not be found automatically.
+ - cuDNN:

- - Updated ``float16`` support
-
-   - Added documentation for GPU float16 ops
-   - Support ``float16`` for ``GpuGemmBatch``
-   - Started to use ``float32`` precision for computations that don't support ``float16`` on GPU
+   - Official support for ``v6.*`` and ``v7.*``, support for ``v5.*`` will be removed in next release
+   - Added spatial transformation operation based on cuDNN
+   - Updated and improved caching system for runtime-chosen cuDNN convolution algorithms
+   - Support cuDNN v7 tensor core operations for convolutions
+   - Restricted cuDNN reductions to contiguous inputs

 New features:
- - Added a wrapper for `Baidu's CTC <https://github.com/baidu-research/warp-ctc>`_ cost and gradient functions
- - Added scalar and elemwise CPU ops for modified Bessel function of order 0 and 1 from ``scipy.special``.
- - Added Scaled Exponential Linear Unit (SELU) activation
- - Added sigmoid_binary_crossentropy function
- - Added tri-gamma function
- - Added modes ``half`` and ``full`` for ``Images2Neibs`` ops
- - Implemented gradient for ``AbstractBatchNormTrainGrad``
- - Implemented gradient for matrix pseudoinverse op
- - Added new prop `replace` for ``ChoiceFromUniform`` op
- - Added new prop ``on_error`` for CPU ``Cholesky`` op
- - Added new Theano flag ``deterministic`` to help control how Theano optimize certain ops that have deterministic versions.
-   Currently used for subtensor Ops only.
- - Added new Theano flag ``cycle_detection`` to speed-up optimization step by reducing time spending in inplace optimizations
- - Added new Theano flag ``check_stack_trace`` to help check the stack trace during optimization process
- - Added new Theano flag ``cmodule.debug`` to allow a debug mode for Theano C code. Currently used for cuDNN convolutions only.
-
-Others:
- - Added deprecation warning for the softmax and logsoftmax vector case
- - Added a warning to announce that C++ compiler will become mandatory in next Theano release ``0.11``
+ - Added ``tensor6()`` and ``tensor7()`` in ``theano.tensor`` module
+ - Added boolean indexing for sub-tensors
+ - Added covariance matrix function ``theano.tensor.cov``
+ - Added new Theano flag ``pickle_test_value`` to help disable pickling test values

 Other more detailed changes:
- - Removed useless warning when profile is manually disabled
- - Added tests for abstract conv
- - Added options for `disconnected_outputs` to Rop
- - Removed ``theano/compat/six.py``
- - Removed ``COp.get_op_params()``
- - Support of list of strings for ``Op.c_support_code()``, to help not duplicate support codes
- - Macro names provided for array properties are now standardized in both CPU and GPU C codes
- - Started to move C code files into separate folder ``c_code`` in every Theano module
- - Many improvements for Travis CI tests (with better splitting for faster testing)
- - Many improvements for Jenkins CI tests: daily testings on Mac and Windows in addition to Linux
+ - Moved all C code files into separate folder ``c_code`` in every Theano module
+ - Improvements for Jenkins tests

-Commiters since 0.9.0:
- - Frederic Bastien
- - Arnaud Bergeron
- - amrithasuresh
+Commiters since 0.10.0beta1:
 - João Victor Tozatti Risso
- - Steven Bocco
- - Pascal Lamblin
 - Mohammed Affan
+ - Frederic Bastien
 - Reyhane Askari
- - Alexander Matyasko
- - Simon Lefrancois
- - Shawn Tan
- - Thomas George
- - Faruk Ahmed
- - Zhouhan LIN
- - Aleksandar Botev
- - jhelie
- - xiaoqie
- - Tegan Maharaj
- - Matt Graham
- - Cesar Laurent
- - Gabe Schwartz
- - Juan Camilo Gamboa Higuera
- - AndroidCloud
- - Saizheng Zhang
- - vipulraheja
- - Florian Bordes
- - Sina Honari
- - Vikram
- - erakra
- - Chiheb Trabelsi
- - Shubh Vachher
- - Daren Eiri
+ - Steven Bocco
 - Gijs van Tulder
- - Laurent Dinh
- - Mohamed Ishmael Diwan Belghazi
- - mila
- - Jeff Donahue
- - Ramana Subramanyam
- - Bogdan Budescu
- - Ghislain Antony Vaillant
- - Jan Schlüter
- - Xavier Bouthillier
- - fo40225
- - Aarni Koskela
- - Adam Becker
- - Adam Geitgey
- - Adrian Keet
- - Adrian Seyboldt
- - Andrei Costinescu
- - Anmol Sahoo
- - Chong Wu
- - Holger Kohr
- - Jayanth Koushik
- - Jenkins
- - Lilian Besson
- - Lv Tao
- - Michael Manukyan
- - Murugesh Marvel
- - NALEPA
- - Ubuntu
- - Zotov Yuriy
- - dareneiri
- - lrast
- - morrme
- - yikang
+ - Boris Fomitchev
+ - Arnaud Bergeron
+ - Joseph Paul Cohen
+ - Dzmitry Bahdanau
+ - Yikang Shen
+ - Faruk Ahmed
+ - Shawn Tan
+ - wyjw
--- a/NEWS_DEV.txt
+++ b/NEWS_DEV.txt
@@ -18,6 +18,7 @@ TODO: better Theano conv doc
 # NB: Following notes contains infos since 0.9.0.

 Highlights:
+ - Support NumPy ``1.13``
 - Moved Python 3.* minimum supported version from 3.3 to 3.4
 - Replaced deprecated package ``nose-parameterized`` with up-to-date package ``parameterized`` for Theano requirements
 - Theano now internally uses ``sha256`` instead of ``md5`` to work on systems that forbide ``md5`` for security reason
@@ -31,7 +32,7 @@ Highlights:

 - Speed up graph toposort algorithm
 - Faster C compilation by massively using a new interface for op params
- - Faster optimization step
+ - Faster optimization step with new destroy handler
 - Documentation updated and more complete
 - Many bug fixes, crash fixes and warning improvements

@@ -54,11 +55,15 @@ Interface changes:
 - Changed ``grad()`` method to ``L_op()`` in ops that need the outputs to compute gradient

 Convolution updates:
+ - Removed old ``conv3d`` interface
+ - Deprecated old ``conv2d`` interface
+ - Updated ``conv`` documentation
 - Extended Theano flag ``dnn.enabled`` with new option ``no_check`` to help speed up cuDNN importation
- - Implemented separable convolutions
- - Implemented grouped convolutions
+ - Implemented separable convolutions for 2D and 3D
+ - Implemented grouped convolutions for 2D and 3D

 GPU:
+ - Added a meta-optimizer to select the fastest GPU implementations for convolutions
 - Prevent GPU initialization when not required
 - Added disk caching option for kernels
 - Added method ``my_theano_function.sync_shared()`` to help synchronize GPU Theano functions
@@ -74,11 +79,13 @@ GPU:
 - ``CrossentropyCategorical1Hot`` and its gradient are now lifted to GPU

 - Better cuDNN support
-
-   - Official support for ``v5.*`` and ``v6.*``
+   - Official support for ``v6.*`` and ``v7.*``, support for ``v5.*`` will be removed in next release
+   - Added spatial transformation operation based on cuDNN
+   - Updated and improved caching system for runtime-chosen cuDNN convolution algorithms
+   - Support cuDNN v7 tensor core operations for convolutions
   - Better support and loading on Windows and Mac
   - Support cuDNN v6 dilated convolutions
-   - Support cuDNN v6 reductions
+   - Support cuDNN v6 reductions for contiguous inputs
   - Added new Theano flags ``cuda.include_path``, ``dnn.base_path`` and ``dnn.bin_path``
     to help configure Theano when CUDA and cuDNN can not be found automatically.

@@ -89,6 +96,9 @@ GPU:
   - Started to use ``float32`` precision for computations that don't support ``float16`` on GPU

 New features:
+ - Added ``tensor6()`` and ``tensor7()`` in ``theano.tensor`` module
+ - Added boolean indexing for sub-tensors
+ - Added covariance matrix function ``theano.tensor.cov``
 - Added a wrapper for `Baidu's CTC <https://github.com/baidu-research/warp-ctc>`_ cost and gradient functions
 - Added scalar and elemwise CPU ops for modified Bessel function of order 0 and 1 from ``scipy.special``.
 - Added Scaled Exponential Linear Unit (SELU) activation
@@ -104,6 +114,7 @@ New features:
 - Added new Theano flag ``cycle_detection`` to speed-up optimization step by reducing time spending in inplace optimizations
 - Added new Theano flag ``check_stack_trace`` to help check the stack trace during optimization process
 - Added new Theano flag ``cmodule.debug`` to allow a debug mode for Theano C code. Currently used for cuDNN convolutions only.
+ - Added new Theano flag ``pickle_test_value`` to help disable pickling test values

 Others:
 - Added deprecation warning for the softmax and logsoftmax vector case
@@ -117,11 +128,41 @@ Other more detailed changes:
 - Removed ``COp.get_op_params()``
 - Support of list of strings for ``Op.c_support_code()``, to help not duplicate support codes
 - Macro names provided for array properties are now standardized in both CPU and GPU C codes
- - Started to move C code files into separate folder ``c_code`` in every Theano module
+ - Moved all C code files into separate folder ``c_code`` in every Theano module
 - Many improvements for Travis CI tests (with better splitting for faster testing)
 - Many improvements for Jenkins CI tests: daily testings on Mac and Windows in addition to Linux

 ALL THE PR BELLOW HAVE BEEN CHECKED
+* https://github.com/Theano/Theano/pull/6301
+* https://github.com/Theano/Theano/pull/6333
+* https://github.com/Theano/Theano/pull/6341
+* https://github.com/Theano/Theano/pull/6332
+* https://github.com/Theano/Theano/pull/6319
+* https://github.com/Theano/Theano/pull/6302
+* https://github.com/Theano/Theano/pull/6300
+* https://github.com/Theano/Theano/pull/6323
+* https://github.com/Theano/Theano/pull/6324
+* https://github.com/Theano/Theano/pull/5817
+* https://github.com/Theano/Theano/pull/6312
+* https://github.com/Theano/Theano/pull/6061
+* https://github.com/Theano/Theano/pull/6305
+* https://github.com/Theano/Theano/pull/6059
+* https://github.com/Theano/Theano/pull/6315
+* https://github.com/Theano/Theano/pull/6295
+* https://github.com/Theano/Theano/pull/6252
+* https://github.com/Theano/Theano/pull/6267
+* https://github.com/Theano/Theano/pull/6207
+* https://github.com/Theano/Theano/pull/6309
+* https://github.com/Theano/Theano/pull/6307
+* https://github.com/Theano/Theano/pull/6000
+* https://github.com/Theano/Theano/pull/6293
+* https://github.com/Theano/Theano/pull/6292
+* https://github.com/Theano/Theano/pull/6299
+* https://github.com/Theano/Theano/pull/6143
+* https://github.com/Theano/Theano/pull/6296
+* https://github.com/Theano/Theano/pull/6280
+* https://github.com/Theano/Theano/pull/6289
+* https://github.com/Theano/Theano/pull/6285
 * https://github.com/Theano/Theano/pull/6275
 * https://github.com/Theano/Theano/pull/6218
 * https://github.com/Theano/Theano/pull/6271

--- a/doc/conf.py
+++ b/doc/conf.py
@@ -74,7 +74,7 @@ copyright = '2008--2017, LISA lab'
 # The short X.Y version.
 version = '0.10'
 # The full version, including alpha/beta/rc tags.
-release = '0.10.0beta1'
+release = '0.10.0beta2'

 # There are two options for replacing |today|: either, you set today to some
 # non-false value, then it is used:

--- a/doc/index.txt
+++ b/doc/index.txt
@@ -21,6 +21,8 @@ learning/machine learning <https://mila.umontreal.ca/en/cours/>`_ classes).
 News
 ====

+* 2017/08/25: Release of Theano 0.10.0beta2, new features and many bugfixes, release candidate to coming.
+
 * 2017/08/09: Release of Theano 0.10.0beta1, many improvements and bugfixes, release candidate to coming.

 * Removed support for the old (device=gpu) backend.  Use the new

--- a/setup.py
+++ b/setup.py
@@ -53,7 +53,7 @@ PLATFORMS           = ["Windows", "Linux", "Solaris", "Mac OS-X", "Unix"]
 MAJOR               = 0
 MINOR               = 10
 MICRO               = 0
-SUFFIX              = "beta1"  # Should be blank except for rc's, betas, etc.
+SUFFIX              = "beta2"  # Should be blank except for rc's, betas, etc.
 ISRELEASED          = False

 VERSION             = '%d.%d.%d%s' % (MAJOR, MINOR, MICRO, SUFFIX)