提交 · 884ed6beffafe44790d02b08adf0038a4bf9f1f5 · testgroup / pytensor

20 7月, 2016 1 次提交
- Merge pull request #4747 from nouiz/mixed_opt · 884ed6be
  由 abergeron 提交于 7月 19, 2016
```
Small mixed changes.
```
  884ed6be
19 7月, 2016 1 次提交

由提交于 7月 18, 2016

* flake 8 changes

* flake 8 changes

* add missing imports

* code cleanup

* code clean up

* flake 8 changes

* code cleanup

* code cleanup

* code cleanup

* code cleanup

* add missing import

d5944c96

16 7月, 2016 6 次提交
- flake8 · ad25509e
  由 Frederic Bastien 提交于 7月 15, 2016
  
  ad25509e
- Fix crash with more recent graphviz · aefd29db
  由 Frederic Bastien 提交于 7月 15, 2016
  
  aefd29db
- Add docstring · 9a925e49
  由 Frederic Bastien 提交于 7月 15, 2016
  
  9a925e49
- Doc about issue label NeedSomeoneToFinish · 3ca887f8
  由 Frederic Bastien 提交于 7月 15, 2016
  
  3ca887f8
- In GpuElemwise, don't call Elemwise.make_node, small speed up and help… · d8ea8892
  由 Frederic Bastien 提交于 7月 15, 2016
```
In GpuElemwise, don't call Elemwise.make_node, small speed up and help investigate. Also build directly a GpuDimShuffle instead of a DimShuffle that will be lifter later.
```
  d8ea8892
- Split local_abstractconv_cudnn into 3 opt to have better stats. · 02bde7ea
  由 Frederic Bastien 提交于 7月 15, 2016
  
  02bde7ea
15 7月, 2016 3 次提交
- Merge pull request #4743 from ruslanagit/ruslanagit · 74b4c807
  由 Frédéric Bastien 提交于 7月 15, 2016
```
Use nvcc flags to compile with cuDNN
```
  74b4c807
- Use nvcc flags to compile with cuDNN · b2177eae
  由 ruslanagit 提交于 7月 14, 2016
```
Fix for issue #4739
```
  b2177eae
- Use nvcc flags to compile with cuDNN · fb5a99d0
  由 ruslanagit 提交于 7月 14, 2016
```
Fix for issue #4739
```
  fb5a99d0
14 7月, 2016 3 次提交
- Merge pull request #4737 from huanzhang12/add-opencl-Images2Neibs · dc49660f
  由 Frédéric Bastien 提交于 7月 14, 2016
```
OpenCL support for Image2Neibs kernels
```
  dc49660f
- OpenCL support for Image2Neibs kernels · e38074ad
  由 Huan Zhang 提交于 7月 10, 2016
```
OpenCL support for Image2Neibs requires a bugfix for libgpuarray
in commit f8b8e156, otherwise python will crash in certain cases.
```
  e38074ad
- Merge pull request #4733 from nouiz/builtins · b5018dba
  由 Frédéric Bastien 提交于 7月 13, 2016
```
Use the six builtins
```
  b5018dba
13 7月, 2016 7 次提交
- Use the six builtins · 8be6057d
  由 Frederic Bastien 提交于 7月 13, 2016
  
  8be6057d
- Fix as_string docstring · 6b8d3f0d
  由 Francesco Visin 提交于 7月 13, 2016
  
  6b8d3f0d
- Merge pull request #4725 from Sentient07/faster-careduce · bdee24d6
  由 Frédéric Bastien 提交于 7月 12, 2016
```
Speeding up local_gpua_careduce
```
  bdee24d6
- Merge pull request #4726 from abergeron/fix_buildbot · c5da9237
  由 Frédéric Bastien 提交于 7月 12, 2016
```
Fix some problems in the buildbot
```
  c5da9237
- Fix problem with float64 for cuda. · 71c1bcbe
  由 Arnaud Bergeron 提交于 7月 12, 2016
  
  71c1bcbe
- Fix problem with testing unpickling. · f74a9d81
  由 Arnaud Bergeron 提交于 7月 12, 2016
  
  f74a9d81
- Fix deprecation warning about using non-integers. · a95f8af1
  由 Arnaud Bergeron 提交于 7月 12, 2016
  
  a95f8af1
12 7月, 2016 7 次提交
- Merge pull request #4683 from Faruk-Ahmed/gpuarray_tests · a96b5ded
  由 Frédéric Bastien 提交于 7月 12, 2016
```
gpuarray tests should pass when floatX=float32 and add this to the daily buildbot
```
  a96b5ded
- Speeding up local_gpua_careduce · 36a857de
  由 sentient07 提交于 7月 12, 2016
  
  36a857de
- Merge pull request #4570 from Sentient07/new_graph2gpu · a24fd9bb
  由 Frédéric Bastien 提交于 7月 12, 2016
```
New graph2gpu
```
  a24fd9bb
- Merge pull request #4712 from Morgan243/master · 2b371c6f
  由 Frédéric Bastien 提交于 7月 12, 2016
```
Decode bytes to string before split
```
  2b371c6f
- Portable version of decode; swap 'dir' variable · daeb22d3
  由 Morgan Stuart 提交于 7月 11, 2016
  
  daeb22d3
- Merge pull request #4716 from nouiz/doc_cudnn · 9bf778f8
  由 Frédéric Bastien 提交于 7月 11, 2016
```
Add in the doc cuDNN BN and new CPU parallelized operations
```
  9bf778f8
- Fix docstring · aab3f35f
  由 Frederic Bastien 提交于 7月 11, 2016
  
  aab3f35f
11 7月, 2016 2 次提交
- Add in the doc cuDNN BN and new CPU parallelized operations · 9b1d9969
  由 Frederic Bastien 提交于 7月 11, 2016
  
  9b1d9969
- Fixing compilation of graphs with only one host_from_gpu OP · 195f9b1d
  由 sentient07 提交于 7月 11, 2016
  
  195f9b1d
10 7月, 2016 2 次提交
- Decode bytes to string before split · 7045e6a2
  由 Morgan Stuart 提交于 7月 09, 2016
  
  7045e6a2
- Corrected few import statements in scan module and fixed error with elemwise, fft and nerv · 345d9024
  由 sentient07 提交于 7月 09, 2016
  
  345d9024
09 7月, 2016 8 次提交
- Added fft opts to new opt, fixed broadcasting test, few cleanups · 3bd237f5
  由 sentient07 提交于 7月 09, 2016
  
  3bd237f5
- standardized opt name and return type of gpualloc and gpueye · ebcf5615
  由 sentient07 提交于 7月 09, 2016
  
  ebcf5615
- Missed out comments · aa853330
  由 sentient07 提交于 7月 08, 2016
  
  aa853330
- Added test for special case of get_scalar_constant_value · 430f0b6a
  由 sentient07 提交于 7月 08, 2016
  
  430f0b6a
- Final Cleanup · 0797283e
  由 sentient07 提交于 7月 08, 2016
  
  0797283e
- Fixed few errors generated from Dilation PR · 6548ed04
  由 sentient07 提交于 7月 06, 2016
```
Simplification

added fd parameter to tcase
```
  6548ed04
- Generalized filter dialation parameter and reverted a change · 9a9e873b
  由 sentient07 提交于 7月 04, 2016
  
  9a9e873b
- Changed interface of gpu_alloc_empty and few cleanups · 6701568f
  由 sentient07 提交于 7月 03, 2016
  
  6701568f