Merge pull request Theano#5545 from notoraptor/prepare-rc-0-9

Prepare release 0.9.0rc1.

.mailmap

@@ -33,6 +33,8 @@ Benjamin Scellier <[email protected]> Benjamin Scellier <scellier@bart4.
 Benjamin Scellier <[email protected]> Benjamin Scellier <scellier@bart5>
 Benjamin Scellier <[email protected]> Benjamin Scellier <[email protected]>
 Benjamin Scellier <[email protected]> Benjamin Scellier <[email protected]>
+Benjamin Scellier <[email protected]> Benjamin Scellier <[email protected]>
+Benjamin Scellier <[email protected]> bscellier <[email protected]>
 Bogdan Budescu <[email protected]> bbudescu <[email protected]>
 Brian Cheung <[email protected]> briancheung <[email protected]>
 Caglar <[email protected]> Caglar <[email protected]>
@@ -61,6 +63,7 @@ Ethan Buchman <[email protected]> ebuchman <[email protected]>
 Evelyn Mitchell <[email protected]> evelynmitchell <[email protected]>
 Faruk Ahmed <[email protected]> Faruk Ahmed <[email protected]>
 Faruk Ahmed <[email protected]> Faruk Ahmed <[email protected]>
+Faruk Ahmed <[email protected]> Faruk Ahmed <[email protected]>
 Fei Wang <[email protected]> fay <[email protected]>
 Francesco Visin <[email protected]> Francesco <[email protected]>
 Francesco Visin <[email protected]> fvisin <[email protected]>

NEWS.txt

@@ -3,12 +3,82 @@ Release Notes
 =============
 
 
+Theano 0.9.0rc1 (20th of February, 2017)
+========================================
+
+This release extends the 0.9.0beta1 and announces the upcoming final release 0.9.
+
+Highlights (since 0.9.0beta1):
+ - Better integration of Theano+libgpuarray packages into conda distribution
+ - Better handling of Windows end-lines into C codes
+ - Better compatibility with NumPy 1.12
+ - Faster scan optimizations
+ - Fixed broadcast checking in scan
+ - Bug fixes related to merge optimizer and shape inference
+ - many other bug fixes and improvements
+ - Updated documentation
+
+ - New GPU back-end:
+
+   - Value of a shared variable is now set inplace
+
+A total of 26 people contributed to this release since 0.9.0beta1 and 118 since 0.8.0, see the list at the bottom.
+
+Interface changes:
+ - In MRG, replaced method `multinomial_wo_replacement()` with new method `choice()`
+
+Convolution updates:
+ - Implement conv2d_transpose convenience function
+
+GPU:
+ - GPUMultinomialFromUniform op now supports multiple dtypes
+
+New features:
+ - OpFromGraph now allows gradient overriding for every input
+ - Added Abstract Ops for batch normalization that use cuDNN when available and pure Theano CPU/GPU alternatives otherwise
+ - Added new Theano flag cuda.enabled
+ - Added new Theano flag print_global_stats to print some global statistics (time spent) at the end
+
+Others:
+ - Split op now has C code for CPU and GPU
+ - "theano-cache list" now includes compilation times
+
+
+Committers since 0.9.0beta1:
+ - Frederic Bastien
+ - Benjamin Scellier
+ - khaotik
+ - Steven Bocco
+ - Arnaud Bergeron
+ - Pascal Lamblin
+ - Gijs van Tulder
+ - Reyhane Askari
+ - Chinnadhurai Sankar
+ - Vincent Dumoulin
+ - Alexander Matyasko
+ - Cesar Laurent
+ - Nicolas Ballas
+ - affanv14
+ - Faruk Ahmed
+ - Anton Chechetka
+ - Alexandre de Brebisson
+ - Amjad Almahairi
+ - Dimitar Dimitrov
+ - Fuchai
+ - Jan Schlüter
+ - Jonas Degrave
+ - Mathieu Germain
+ - Rebecca N. Palmer
+ - Simon Lefrancois
+ - valtron
+
+
 Theano 0.9.0beta1 (24th of January, 2017)
 =========================================
 
 This release contains a lot of bug fixes and improvements + new features, to prepare the upcoming release candidate.
 
-Highlight:
+Highlights:
  - Many computation and compilation speed up
  - More numerical stability by default for some graph
  - Jenkins (gpu tests run on PR in addition to daily buildbot)
@@ -23,14 +93,15 @@ Highlight:
  - scan with checkpoint (trade off between speed and memory usage, useful for long sequences)
  - Added a bool dtype
 
- - New back-end:
+ - New GPU back-end:
+
    - float16 storage
    - better mapping between theano device number and nvidia-smi number, using the PCI bus ID of graphic cards
-   - More pooling support on GPU when cuDNN isn't there.
-   - ignore_border=False is now implemented for pooling.
+   - More pooling support on GPU when cuDNN isn't there
+   - ignore_border=False is now implemented for pooling
 
 
-A total of 112 people contributed to this release, see the list at the bottom.
+A total of 112 people contributed to this release since 0.8.0, see the list at the bottom.
 
 
 Interface changes:
@@ -41,7 +112,7 @@ Interface changes:
  - Move softsign out of sandbox to theano.tensor.nnet.softsign
  - Roll make the shift be modulo the size of the axis we roll on
  - Merge CumsumOp/CumprodOp into CumOp
- - round() default to the same as NumPy: half_to_even.
+ - round() default to the same as NumPy: half_to_even
 
 Convolution updates:
  - Multi-cores convolution and pooling on CPU
@@ -57,7 +128,7 @@ GPU:
  - Support for solve (using cusolver), erfinv and erfcinv
  - cublas gemv workaround when we reduce on an axis with a dimensions size of 0
  - Warn user that some cuDNN algorithms may produce unexpected results in certain environments
-   for convolution backward filter operations.
+   for convolution backward filter operations
 
 New features:
  - Add gradient of solve, tensorinv (CPU), tensorsolve (CPU) searchsorted (CPU)

NEWS_DEV.txt

@@ -15,7 +15,15 @@ git shortlog -sn rel-0.8.0..
 
 TODO: better Theano conv doc
 
-Highlight:
+Highlights:
+ - Better integration of Theano+libgpuarray packages into conda distribution
+ - Better handling of Windows end-lines into C codes
+ - Better compatibility with NumPy 1.12
+ - Faster scan optimizations
+ - Fixed broadcast checking in scan
+ - Bug fixes related to merge optimizer and shape inference
+ - many other bug fixes and improvements
+ - Updated documentation
  - Many computation and compilation speed up
  - More numerical stability by default for some graph
  - Jenkins (gpu tests run on PR in addition to daily buildbot)
@@ -30,29 +38,34 @@ Highlight:
  - scan with checkpoint (trade off between speed and memory usage, useful for long sequences)
  - Added a bool dtype
 
- - New back-end:
+ - New GPU back-end:
+
+   - Value of a shared variable is now set inplace
    - float16 storage
    - better mapping between theano device number and nvidia-smi number, using the PCI bus ID of graphic cards
-   - More pooling support on GPU when cuDNN isn't there.
-   - ignore_border=False is now implemented for pooling.
+   - More pooling support on GPU when cuDNN isn't there
+   - ignore_border=False is now implemented for pooling
 
 Interface changes:
+ - In MRG, replaced method `multinomial_wo_replacement()` with new method `choice()`
  - New pooling interface
  - Pooling parameters can change at run time
  - When converting empty list/tuple, now we use floatX dtype
  - The MRG random generator now try to infer the broadcast pattern of its output
  - Move softsign out of sandbox to theano.tensor.nnet.softsign
  - Roll make the shift be modulo the size of the axis we roll on
  - Merge CumsumOp/CumprodOp into CumOp
- - round() default to the same as NumPy: half_to_even.
+ - round() default to the same as NumPy: half_to_even
 
 Convolution updates:
+ - Implement conv2d_transpose convenience function
  - Multi-cores convolution and pooling on CPU
  - New abstract 3d convolution interface similar to the 2d convolution interface
  - Dilated convolution
 
 
 GPU:
+ - GPUMultinomialFromUniform op now supports multiple dtypes
  - cuDNN: support versoin 5.1 and wrap batch normalization (2d and 3d) and RNN functions
  - Multiple-GPU, synchrone update (via platoon, use NCCL)
  - GpuAdvancedSubtensor in new back-end
@@ -61,9 +74,13 @@ GPU:
  - Support for solve (using cusolver), erfinv and erfcinv
  - cublas gemv workaround when we reduce on an axis with a dimensions size of 0
  - Warn user that some cuDNN algorithms may produce unexpected results in certain environments
-   for convolution backward filter operations.
+   for convolution backward filter operations
 
 New features:
+ - OpFromGraph now allows gradient overriding for every input
+ - Added Abstract Ops for batch normalization that use cuDNN when available and pure Theano CPU/GPU alternatives otherwise
+ - Added new Theano flag cuda.enabled
+ - Added new Theano flag print_global_stats to print some global statistics (time spent) at the end
  - Add gradient of solve, tensorinv (CPU), tensorsolve (CPU) searchsorted (CPU)
  - Add Multinomial Without Replacement
  - conv3d2d support full and half mode (REMOVE?)
@@ -77,15 +94,15 @@ New features:
 
 
 Others:
+ - Split op now has C code for CPU and GPU
+ - "theano-cache list" now includes compilation times
  - Speed up argmax only on gpu (without also needing the max)
  - A few unfrequent bugfix 
  - More stack trace in error message
  - Speed up cholesky grad
  - log(sum(exp(...))) now get stability optimized
 
 
-
-
 Other more detailed changes:
  - Allow more then one output to be an destructive inplace
  - Add flag profiling.ignore_first_call, useful to profile the new gpu back-end
@@ -100,6 +117,58 @@ Other more detailed changes:
 
 
 ALL THE PR BELLOW HAVE BEEN CHECKED
+* https://github.com/Theano/Theano/pull/5559
+* https://github.com/Theano/Theano/pull/5568
+* https://github.com/Theano/Theano/pull/5553
+* https://github.com/Theano/Theano/pull/5561
+* https://github.com/Theano/Theano/pull/5558
+* https://github.com/Theano/Theano/pull/5544
+* https://github.com/Theano/Theano/pull/5552
+* https://github.com/Theano/Theano/pull/5547
+* https://github.com/Theano/Theano/pull/5542
+* https://github.com/Theano/Theano/pull/5451
+* https://github.com/Theano/Theano/pull/5520
+* https://github.com/Theano/Theano/pull/5539
+* https://github.com/Theano/Theano/pull/5532
+* https://github.com/Theano/Theano/pull/5457
+* https://github.com/Theano/Theano/pull/5477
+* https://github.com/Theano/Theano/pull/5499
+* https://github.com/Theano/Theano/pull/5518
+* https://github.com/Theano/Theano/pull/5527
+* https://github.com/Theano/Theano/pull/5522
+* https://github.com/Theano/Theano/pull/5505
+* https://github.com/Theano/Theano/pull/5523
+* https://github.com/Theano/Theano/pull/5516
+* https://github.com/Theano/Theano/pull/5511
+* https://github.com/Theano/Theano/pull/5443
+* https://github.com/Theano/Theano/pull/5255
+* https://github.com/Theano/Theano/pull/5508
+* https://github.com/Theano/Theano/pull/5479
+* https://github.com/Theano/Theano/pull/5462
+* https://github.com/Theano/Theano/pull/5490
+* https://github.com/Theano/Theano/pull/5480
+* https://github.com/Theano/Theano/pull/5497
+* https://github.com/Theano/Theano/pull/5489
+* https://github.com/Theano/Theano/pull/5474
+* https://github.com/Theano/Theano/pull/5454
+* https://github.com/Theano/Theano/pull/5469
+* https://github.com/Theano/Theano/pull/5458
+* https://github.com/Theano/Theano/pull/5481
+* https://github.com/Theano/Theano/pull/5190
+* https://github.com/Theano/Theano/pull/5473
+* https://github.com/Theano/Theano/pull/5456
+* https://github.com/Theano/Theano/pull/5398
+* https://github.com/Theano/Theano/pull/5468
+* https://github.com/Theano/Theano/pull/5459
+* https://github.com/Theano/Theano/pull/5452
+* https://github.com/Theano/Theano/pull/5298
+* https://github.com/Theano/Theano/pull/5442
+* https://github.com/Theano/Theano/pull/5450
+* https://github.com/Theano/Theano/pull/5435
+* https://github.com/Theano/Theano/pull/5446
+* https://github.com/Theano/Theano/pull/5447
+* https://github.com/Theano/Theano/pull/5445
+* https://github.com/Theano/Theano/pull/5323
 * https://github.com/Theano/Theano/pull/5421
 * https://github.com/Theano/Theano/pull/5343
 * https://github.com/Theano/Theano/pull/5437

doc/conf.py

@@ -74,7 +74,7 @@
 # The short X.Y version.
 version = '0.9'
 # The full version, including alpha/beta/rc tags.
-release = '0.9.0beta1'
+release = '0.9.0rc1'
 
 # There are two options for replacing |today|: either, you set today to some
 # non-false value, then it is used:

doc/index.txt

@@ -21,6 +21,8 @@ learning/machine learning <https://mila.umontreal.ca/en/cours/>`_ classes).
 News
 ====
 
+* 2017/02/20: Release of Theano 0.9.0rc1, many improvements and bugfixes, final release to coming.
+
 * 2017/01/24: Release of Theano 0.9.0beta1, many improvements and bugfixes, release candidate to coming.
 
 * 2016/05/09: New technical report on Theano:

doc/introduction.txt

@@ -165,11 +165,13 @@ Note: There is no short term plan to support multi-node computation.
 Theano Vision State
 ===================
 
-Here is the state of that vision as of January 24th, 2017 (after Theano 0.9.0beta1):
+Here is the state of that vision as of February 20th, 2017 (after Theano 0.9.0rc1):
 
 * We support tensors using the `numpy.ndarray` object and we support many operations on them.
 * We support sparse types by using the `scipy.{csc,csr,bsr}_matrix` object and support some operations on them.
 * We have implementing/wrapping more advanced linear algebra operations. Still many more possible.
+* We have basic support for the creation of new operations from graphs at runtime. It supports well gradient overload
+  for every input and inlining at the start of compilation. We don't cover well the case when it is not inlined.
 * We have many graph transformations that cover the 4 categories listed above.
 * We can improve the graph transformation with better storage optimization
   and instruction selection.
@@ -195,7 +197,9 @@ Here is the state of that vision as of January 24th, 2017 (after Theano 0.9.0bet
 
 * No multi-node support.
 * Most, but not all NumPy functions/aliases are implemented.
+
   * https://github.com/Theano/Theano/issues/1080
+
 * Wrapping an existing Python function in easy and documented.
 * We know how to separate the shared variable memory
   storage location from its object type (tensor, sparse, dtype, broadcast

setup.py

@@ -53,7 +53,7 @@
 MAJOR               = 0
 MINOR               = 9
 MICRO               = 0
-SUFFIX              = "beta1"  # Should be blank except for rc's, betas, etc.
+SUFFIX              = "rc1"  # Should be blank except for rc's, betas, etc.
 ISRELEASED          = False
 
 VERSION             = '%d.%d.%d%s' % (MAJOR, MINOR, MICRO, SUFFIX)

theano/gpuarray/__init__.py

@@ -45,9 +45,9 @@ def init_dev(dev, name=None):
     global pygpu_activated
     if not config.cxx:
         raise RuntimeError("The new gpu-backend need a c++ compiler.")
-    if (pygpu.version.major, pygpu.version.minor) < (0, 6):
+    if (pygpu.version.major, pygpu.version.minor, pygpu.version.patch) < (0, 6, 1):
         raise ValueError(
-            "Your installed version of pygpu is too old, please upgrade to 0.6 or later")
+            "Your installed version of pygpu is too old, please upgrade to 0.6.1 or later")
     # This is for the C headers API, we need to match the exact version.
     if pygpu.gpuarray.api_version()[0] != 1:
         raise ValueError(