提交 · 06cdb1fa35e28fabd6bec1cf4487941d5eaf4913 · testgroup / pytensor

31 1月, 2014 1 次提交
- Merge pull request #1710 from nouiz/gpu_reduce · 06cdb1fa
  由 abergeron 提交于 1月 30, 2014
```
new Gpu reduce based on the version in Theano that is much faster.
```
  06cdb1fa
30 1月, 2014 6 次提交
- Rename GpuCAReduce to GpuCAReduceCuda and reuse the name GpuCAReduce for… · 1121fe45
  由 Frederic 提交于 1月 30, 2014
```
Rename GpuCAReduce to GpuCAReduceCuda and reuse the name GpuCAReduce for GpuCAReduceCPY to allow reloading older pickled files.
```
  1121fe45
- Port typo fix from gpuarray files by Olivier D. to the old back-end. · a2afb9dd
  由 Frederic 提交于 1月 30, 2014
  
  a2afb9dd
- Fix opt crash with other dtype then float32, the code isn't ready. · 111400e4
  由 Frederic 提交于 1月 29, 2014
  
  111400e4
- Merge pull request #1700 from MarcCote/cumsum · 262f59a2
  由 Frédéric Bastien 提交于 1月 29, 2014
```
Added the cumsum and cumprod functions similar to numpy's ones.
```
  262f59a2
- remove GpuShape and associated opt as now we reuse the Shape Op with specific c code. · 047c8913
  由 Frederic 提交于 1月 29, 2014
  
  047c8913
- Remove check_preallocated_output and use default mode for theano.function. · 5c504099
  由 Marc-Alexandre Cote 提交于 1月 29, 2014
  
  5c504099
29 1月, 2014 8 次提交
- Fix crashes introduced in this PR. · b68d4a87
  由 Frederic 提交于 1月 29, 2014
  
  b68d4a87
- Fix crash and comment. · 452b6e2b
  由 Frederic 提交于 1月 28, 2014
  
  452b6e2b
- Use the simpler interface to CUDA extension. · 2f252daa
  由 Frederic 提交于 1月 28, 2014
  
  2f252daa
- Simplify unit tests using check_preallocated_output · 09d7318f
  由 Marc-Alexandre Cote 提交于 1月 28, 2014
  
  09d7318f
- Fix compilation crash introduced in this PR. · 571f3368
  由 Frederic 提交于 1月 28, 2014
  
  571f3368
- Fix comment. · d7a67b73
  由 Frederic 提交于 1月 28, 2014
  
  d7a67b73
- Implemented cumprod function similar to numpy's one. · 195bee05
  由 Marc-Alexandre Cote 提交于 1月 28, 2014
  
  195bee05
- Reuse allocated memory when possible. · 100d336e
  由 Marc-Alexandre Cote 提交于 1月 28, 2014
  
  100d336e
28 1月, 2014 4 次提交
- Merge pull request #1712 from nouiz/nonetype · e02ce0d8
  由 abergeron 提交于 1月 27, 2014
```
Add C code to NoneTypeT and fix MaxAndArgmax c_code to work with it.
```
  e02ce0d8
- Fix compilation just introduced. · 6545c206
  由 Frederic 提交于 1月 27, 2014
  
  6545c206
- Merge pull request #1713 from MarcCote/scalar_cdeclare · 3dad804a
  由 Frédéric Bastien 提交于 1月 27, 2014
```
Uniformize the typedef name of Scalar's dtype.
```
  3dad804a
- Uniformize the typedef name of Scalar's dtype. · c8c0bbc5
  由 Marc-Alexandre Cote 提交于 1月 27, 2014
  
  c8c0bbc5
27 1月, 2014 1 次提交
- Add C code to NoneTypeT and fix MaxAndArgmax c_code to work with it. · 2278e82d
  由 Frederic 提交于 1月 26, 2014
  
  2278e82d
26 1月, 2014 2 次提交
- Changed PyArray_SHAPE to PyArray_DIMS. · 2dddf4ac
  由 Marc-Alexandre Cote 提交于 1月 25, 2014
  
  2dddf4ac
- Added a C version for the CumsumOp. · c9d6cfe0
  由 Marc-Alexandre Cote 提交于 1月 25, 2014
  
  c9d6cfe0
25 1月, 2014 18 次提交
- Add GpuCAReduce more cases and remove warning about shape of 0 in infer shape. · c4ea6761
  由 Frederic 提交于 1月 24, 2014
  
  c4ea6761
- Last back of compilation fix to GpuCAReduce · 907a465f
  由 Frederic 提交于 1月 24, 2014
  
  907a465f
- skip test error due to notimplemented cases. · e37e6f9d
  由 Frederic 提交于 1月 24, 2014
  
  e37e6f9d
- Test with clinker too as now the Shape Op have c code · 1390deb4
  由 Frederic 提交于 1月 24, 2014
  
  1390deb4
- Disable some GpuCAReduce test cases that aren't implemented. · 6f1c4d22
  由 Frederic 提交于 1月 24, 2014
  
  6f1c4d22
- better error message. · b67bb9e0
  由 Frederic 提交于 1月 24, 2014
  
  b67bb9e0
- register c code for Shape op with GpuArrayVariable and doc the register fct(). · b467c97f
  由 Frederic 提交于 1月 24, 2014
  
  b467c97f
- Move Shape Op. · 7b1d7d2f
  由 Frederic 提交于 1月 24, 2014
  
  7b1d7d2f
- gpucareduce port continue. · 30d2c4c3
  由 Frederic 提交于 1月 19, 2014
  
  30d2c4c3
- Port gpu reduce opt to make reshape the input if needed. · a482cc6f
  由 Frederic 提交于 1月 19, 2014
  
  a482cc6f
- Port sync · 85faa716
  由 Frederic 提交于 1月 19, 2014
  
  85faa716
- Test with dtype specified by the tests. · c908977c
  由 Frederic 提交于 1月 19, 2014
  
  c908977c
- Convert some of the old GpuCAReduce code to the new interface. · d76b9684
  由 Frederic 提交于 1月 19, 2014
  
  d76b9684
- Override perform() (otherwise DebugMode and py linker would use it) · a8c03627
  由 Frederic 提交于 1月 19, 2014
  
  a8c03627
- First step of conversion to the new back-end GpuCAReduce. · de775205
  由 Frederic 提交于 1月 10, 2014
```
Added tests and updated the opt to use it. It should not compile for now.
```
  de775205
- Rename GpuCAReduce to GpuCAReduceCPY · c0cca58a
  由 Frederic 提交于 1月 10, 2014
  
  c0cca58a
- Merge pull request #1709 from yosinski/master · 440255b0
  由 Olivier Delalleau 提交于 1月 24, 2014
```
Fix for string substitution / implicity tuple bug.
```
  440255b0
- Added missing parentheses around tuple · 3e755c3d
  由 Jason Yosinski 提交于 1月 24, 2014
  
  3e755c3d