* Fix shape crash inserted by Scan optimization. The gradient of some recursive scan was making the PushOutSeqScan optimization insert crash during the execution of a Theano function. (Frédéric B., reported by Hugo Larochelle)
We recommand everybody to update to this version.
Highlights:
Highlights:
* Python 3.3 compatibility with buildbot.
* Python 3.3 compatibility with buildbot test for it.
* Full advanced indexing support.
* Full advanced indexing support.
* Better Windows 64 bit support.
* Better Windows 64 bit support.
* New profiler.
* New profiler.
* Better error messages that help debugging.
* Better error messages that help debugging.
* Better support of newer NumPy version(remove useless warning/crash) (Frederic B., Huy Nguyen)
* Better support of newer NumPy version (remove useless warning/crash).
* Faster optimization/compilation for big graph (Frederic B.)
* Faster optimization/compilation for big graph.
* Move in Theano the Conv3d2d implementation (James Bergstra, Frederic B., Pascal L.)
* Move in Theano the Conv3d2d implementation.
* Better SymPy/Theano bridge: Make an Theano op from SymPy expression and use SymPy c code generator (Matthew Rocklin)
* Better SymPy/Theano bridge: Make an Theano op from SymPy expression and use SymPy c code generator.
* Bug fixes.
Committers for this rc4 only:
Installation:
Installation:
* Canopy support (direct link to MKL):
* Canopy support (direct link to MKL):
...
@@ -39,31 +40,30 @@ Installation:
...
@@ -39,31 +40,30 @@ Installation:
* Anaconda instructions (Pascal L., Frederic B.)
* Anaconda instructions (Pascal L., Frederic B.)
* Doc Ubuntu 13.04 (Frederic B.)
* Doc Ubuntu 13.04 (Frederic B.)
* Better support of newer NumPy version(remove useless warning/crash) (Frederic B., Huy Nguyen)
Committers for this rc3 only:
Bug fixes:
Bug fixes:
* Disable the optimization that introduce the CSMGradC op as it don't work correctly with unsorted indices. (Frederic B.)
* Scan: if a scan node was cloned (by theano.clone) with different inputs, and if both the initial and the cloned nodes are used in the function being compiled, the value of the outputs of one would be replaced with the outputs of the other one. (Pascal L.)
* Fix wrong result of GpuDownsampleFactorMaxGrad on Mac OSX. (Pascal L.)
* Sparse: Disable the optimization that introduce the CSMGradC op as it don't work correctly with unsorted indices. (Frederic B.)
* Auto-Detect and work around a bug in BLAS on MacOS X (Pascal L.)
* Mac: Fix wrong result of GpuDownsampleFactorMaxGrad on Mac OSX. (Pascal L.)
* Work around bug in MacOS X. If 2 compiled modules had the same name, the OS or Python was not always the right one even when we used the right handle to it. (Pascal L.)
* Mac: Auto-Detect and work around a bug in BLAS on MacOS X (Pascal L.)
* Mac: Work around bug in MacOS X. If 2 compiled modules had the same name, the OS or Python was not always the right one even when we used the right handle to it. (Pascal L.)
Use this hash in the Python module, and in %(nodename)s, so that different helper functions in the support code for different Ops will always have different names.
Use this hash in the Python module, and in %(nodename)s, so that different helper functions in the support code for different Ops will always have different names.
* Fix infinite loop related to Scan on the GPU. (Pascal L.)
* Fix infinite loop when the compiledir is full. (Frederic B.)
* Fix a shape cycle crash in the optimizer (Pascal L., Frédéric B., reported by Cho KyungHyun)
* Fix command not returning with recent mingw64 on Windows (Pascal L., reported by many people)
* Fix ConstructSparseFromList.infer_shape (Pascal L., reported by Rami Al-Rfou')
* (introduced in the development version after 0.6rc3 release) (Frederic B.)
* (introduced in the development version after 0.6rc3 release) (Frederic B.)
Reduction that upcasts the input on no axis (ex: call theano.sum() on a scalar when the original dtype isn't float64 or [u]int64). It produced bad results as we don't upcast the inputs in the code, we just copy them.
Reduction that upcasts the input on no axis (ex: call theano.sum() on a scalar when the original dtype isn't float64 or
[u]int64). It produced bad results as we did not upcasted the inputs in the code, we just copy them.
* Fix some cases of theano.clone() when we get a replacement of x that is a function of x. (Razvan P., reported by Akio Takano)
* Fix some cases of theano.clone() when we get a replacement of x that is a function of x. (Razvan P., reported by Akio Takano)
* Fix grad of Alloc when we unbroadcast the value value and it isn't a scalar. (Frederic B., reported Ian G.)
* Fix grad of Alloc when we unbroadcast the value value and it isn't a scalar. (Frederic B., reported Ian G.)
* I some cases (I think most cases), there was an exception raised in the theano.tensor.grad() method. But in theory, there could be bad shapes produced in the unbroadcasted dimensions.
* I some cases (I think most cases), there was an exception raised in the theano.tensor.grad() method.
But in theory, there could be bad shapes produced in the unbroadcasted dimensions.
New Features:
New Features:
* ???Do I add it here??? First version of the new GPU back-end available (Arnaud Bergeron, Frederic B.)
* Better SymPy/Theano bridge: Make an Theano op from SymPy expression and use SymPy c code generator (Matthew Rocklin)
* Move in Theano the Conv3d2d implementation (James Bergstra, Frederic B., Pascal L.)
* First version of the new GPU back-end available (Arnaud Bergeron, Frederic B.)
* Not all Ops have been converted to this new back-end.
* Not all Ops have been converted to this new back-end.
To use, use Theano flag device=cudaN or device=openclN, where N is a integer.
To use, use Theano flag device=cudaN or device=openclN, where N is a integer.
* Python 3.3 compatible (abalkin, Gabe Schwartz, Frederic B., Pascal L.)
* Python 3.3 compatible (abalkin, Gabe Schwartz, Frederic B., Pascal L.)
* A new profiler (Frederic B.)
* A new profiler (Frederic B.)
The new profiler now can profile the memory with the Theano flag profile_memory=True.
The new profiler now can profile the memory with the Theano flag profile_memory=True.
...
@@ -168,7 +168,12 @@ Speed-ups:
...
@@ -168,7 +168,12 @@ Speed-ups:
* A fix that removes a local_setsubtensor_of_allocs optimization warning and enables it in that case. (Frederic B., reported by John Salvatier)
* A fix that removes a local_setsubtensor_of_allocs optimization warning and enables it in that case. (Frederic B., reported by John Salvatier)
* Make inv_as_solve optimization work (Matthew Rocklin)
* Make inv_as_solve optimization work (Matthew Rocklin)
Crash fixes:
Crash/no return fixes:
* Fix shape crash inserted by Scan optimization. The gradient of some recursive scan was making the PushOutSeqScan optimization insert crash during the execution of a Theano function. (Frédéric B., reported by Hugo Larochelle)
* Fix command not returning with recent mingw64 on Windows (Pascal L., reported by many people)
* Fix infinite loop related to Scan on the GPU. (Pascal L.)
* Fix infinite loop when the compiledir is full. (Frederic B.)
* Fix a shape cycle crash in the optimizer (Pascal L., Frédéric B., reported by Cho KyungHyun)
* Fix MRG normal now accept to generate scalar. (Pascal L.)
* Fix MRG normal now accept to generate scalar. (Pascal L.)
* Fix some GPU compilation issue on Mac (John Yani, Frédéric B.)
* Fix some GPU compilation issue on Mac (John Yani, Frédéric B.)
* Fix crash when building symbolic random variables with a mix of symbolic and numeric scalar in the "size" parameter. (Pascal L., Reported by Wu Zhen Zhou)
* Fix crash when building symbolic random variables with a mix of symbolic and numeric scalar in the "size" parameter. (Pascal L., Reported by Wu Zhen Zhou)
...
@@ -193,8 +198,8 @@ Crash fixes:
...
@@ -193,8 +198,8 @@ Crash fixes:
* Fix crash in the elemwise python code for some big shape with power of 2. (Sina Honari, Pascal L.)
* Fix crash in the elemwise python code for some big shape with power of 2. (Sina Honari, Pascal L.)
* Fix compile and import errors on Windows including for the GPU. (Bogdan Budescu)
* Fix compile and import errors on Windows including for the GPU. (Bogdan Budescu)
