提交 · bd1bdc441adfcdde5f5a27c4eb01e01e15aa2bcf · testgroup / pytensor

05 5月, 2013 1 次提交
- Typo fixes related to new memory profiling · bd1bdc44
  由 Olivier Delalleau 提交于 5月 05, 2013
  
  bd1bdc44
04 5月, 2013 3 次提交
- Merge pull request #1277 from nouiz/memory_prof · 91a31814
  由 lamblin 提交于 5月 03, 2013
```
Memory profiler in the new profiler
```
  91a31814
- Add assert to detect futur potential scipy change. · 33ded753
  由 Frederic 提交于 5月 03, 2013
  
  33ded753
- Fix code revice comments and add a message. · fdb4028d
  由 Frederic 提交于 5月 03, 2013
  
  fdb4028d
03 5月, 2013 13 次提交
- Add doc/ref about the new Type method. · 57a65772
  由 Frederic 提交于 5月 03, 2013
  
  57a65772
- Support memory profiling of theano.scalar.Scalar · 6078b095
  由 Frederic 提交于 5月 03, 2013
  
  6078b095
- fix a printed profiler information and rename a variable. · a45a4692
  由 Frederic 提交于 5月 03, 2013
  
  a45a4692
- Add the theano flag profiling.{n_apply,n_ops,min_memory_size} · 7ea29681
  由 Frederic 提交于 5月 03, 2013
  
  7ea29681
- replace T. by tensor. · 7f25e754
  由 Frederic 提交于 5月 03, 2013
  
  7f25e754
- Enable memory profiling for sparse variable. · 56eb0c95
  由 Frederic 提交于 5月 03, 2013
  
  56eb0c95
- Add Type.get_size() fct to help during profiling. · 0cbdf291
  由 Frederic 提交于 5月 03, 2013
```
This add the size of the RandomType.
```
  0cbdf291
- Small change following code review. · 25a45105
  由 Frederic 提交于 5月 02, 2013
  
  25a45105
- Merge pull request #1361 from nouiz/get_scalar_constant_value · c4a3bd9f
  由 lamblin 提交于 5月 02, 2013
```
Get scalar constant value
```
  c4a3bd9f
- Added comment following code review. · c1b2c1b6
  由 Frederic 提交于 5月 02, 2013
  
  c1b2c1b6
- Merge pull request #1362 from nouiz/doc_adv_idx · c5f8cc3d
  由 lamblin 提交于 5月 02, 2013
```
Document that we now support Integer advanced indexing (not boolean mask)
```
  c5f8cc3d
- Merge pull request #1348 from nouiz/fixes · cb371ba7
  由 lamblin 提交于 5月 02, 2013
```
Small fixes
```
  cb371ba7
- Merge pull request #1303 from nouiz/gpusoftmax · ec3a90d4
  由 James Bergstra 提交于 5月 02, 2013
```
Fix GpuSoftmax[WithBias] Ops to work for large rows
```
  ec3a90d4
02 5月, 2013 9 次提交
- more pep8 · 2335f829
  由 Frederic 提交于 4月 29, 2013
  
  2335f829
- Fix pep8 and add a comment following code review. · e7455488
  由 Frederic 提交于 4月 29, 2013
  
  e7455488
- Add timming in check_blas.py · ec8a9d78
  由 Frederic 提交于 3月 30, 2013
  
  ec8a9d78
- Add link to a tool that profile memory · b8250943
  由 Frederic 提交于 3月 30, 2013
  
  b8250943
- pep8 and rename var to be the same as used elsewhere. · 5aa9892c
  由 Frederic 提交于 3月 30, 2013
  
  5aa9892c
- Make GpuSoftmaxWithBias work with bigger input. · 5b6f228e
  由 Frederic 提交于 3月 28, 2013
```
Also in kernel_codegen.py There is a fix then X is strided. The
variable loop_line wasn't checking the stride.
```
  5b6f228e
- Faster GpuSoftmax for big row. · bbd6a10e
  由 Frederic 提交于 3月 27, 2013
  
  bbd6a10e
- Make GpuSoftmax work with bigger row. · f3fc1f0f
  由 Frederic 提交于 3月 26, 2013
  
  f3fc1f0f
- pep8 · c2c34032
  由 Frederic 提交于 3月 26, 2013
  
  c2c34032
01 5月, 2013 11 次提交
- Warn only once. · 1591a674
  由 Frederic 提交于 4月 30, 2013
  
  1591a674
- don't delete when we don't loop. · cb94f4e6
  由 Frederic 提交于 4月 30, 2013
  
  cb94f4e6
- don't consider sparse variable as dense! · bcdaa379
  由 Frederic 提交于 4月 30, 2013
  
  bcdaa379
- remove var not used. · 859e8fea
  由 Frederic 提交于 4月 30, 2013
  
  859e8fea
- in memory profile, mark the CudaNdarray object strides as c when it is the case. · a24027b2
  由 Frederic 提交于 4月 30, 2013
  
  a24027b2
- Document that we now support Integer advanced indexing (not boolean mask). · 4dd2d1ba
  由 Frederic Bastien 提交于 4月 30, 2013
  
  4dd2d1ba
- Add code to test the example · 38f3f8c5
  由 Frederic 提交于 4月 30, 2013
  
  38f3f8c5
- Fix c code in example and add c_code_cache_version. · 5cfd421a
  由 Frederic 提交于 4月 30, 2013
  
  5cfd421a
- pep8 · 4662adeb
  由 Frederic 提交于 4月 30, 2013
  
  4662adeb
- Added an assert in an example to show people how to test that. · f278742c
  由 Frederic 提交于 4月 30, 2013
  
  f278742c
- Fix doc and associated test about get_scalar_constant_value · d861fd61
  由 Frederic 提交于 4月 30, 2013
```
fixes gh-1300
```
  d861fd61
30 4月, 2013 3 次提交
- Merge pull request #1360 from nouiz/revert_blas_windows · 97fff534
  由 Olivier Delalleau 提交于 4月 29, 2013
```
Revert "msvc doesn't understand -Wl,-rpath option"
```
  97fff534
- Revert "msvc doesn't understand -Wl,-rpath option" · 0e746782
  由 Frederic 提交于 4月 29, 2013
```
This reverts commit cc639be5.

This shouldn't be done on the platform, but on the compiler. Anyway,
we won't make GPU op link to BLAS in Pylearn2 for now, so I just
remove this.
```
  0e746782
- Remove not used var tval, simplify the test and test for the error! · a0344d4b
  由 Frederic 提交于 4月 29, 2013
  
  a0344d4b