Commits · 08e80a04ade9901a6f177c19beacfb2153356576 · ift / NIFTy

24 Apr, 2016 2 commits
- fix all and any · 08e80a04
  csongor authored Apr 24, 2016
  
  08e80a04
- fix nanmin and nanmax · 34279eb5
  csongor authored Apr 24, 2016
  
  34279eb5
22 Apr, 2016 1 commit
- Merge branch 'tests-for-axis' into 'add_axis_keyword_to_d2o' · 8cf39e8f
  Theo Steininger authored Apr 22, 2016
```
Tests for axis



See merge request !9
```
  8cf39e8f
20 Apr, 2016 3 commits
- Replaced the sendrecv by a Sendrecv in the slicing_distributor. · d72b7996
  theos authored Apr 20, 2016
  
  d72b7996
- Fix median, add axis for argmin etc. · facb6103
  csongor authored Apr 20, 2016
  
  facb6103
- Add axis keyword to point_space and test it · f399b0da
  csongor authored Apr 19, 2016
  
  f399b0da
19 Apr, 2016 1 commit
- Add tests for testing d2o operations with axis keyword · 51428d74
  csongor authored Apr 19, 2016
  
  51428d74
18 Apr, 2016 2 commits
- Solve median and std · 528d7a73
  csongor authored Apr 18, 2016
  
  528d7a73
- obj.unique() now is built on top of MPI.comm.allreduce with a custom MPI.Op · 7f09a39a
  theos authored Apr 18, 2016
  
  7f09a39a
17 Apr, 2016 2 commits
- Reapplied speedup from 680d82c0 · ec62ebd2
  theos authored Apr 17, 2016
  
  ec62ebd2
- Added axis functionality to d2o methods. · c09199ba
  theos authored Apr 17, 2016
```
-> Therefore modified the MPI_dummy.py

Fixed bincount method of _slicing_distributor (now uses allreduce instead of allgather+sum)
```
  c09199ba
07 Apr, 2016 3 commits
- Speeded up the contraction_helper and the distributed_data_object.__init__ · 680d82c0
  theos authored Apr 07, 2016
  
  680d82c0
- Fixed the import logic in the dependency_injector. · 7d3eca8a
  theos authored Apr 07, 2016
  
  7d3eca8a
- dependency_injector now is able to apply a check function to a loaded module before registering. · 6343088d
  theos authored Apr 07, 2016
```
Added a check for pyfftw.FFTW_MPI
```
  6343088d
06 Apr, 2016 3 commits
- Improved distributed_data_object's performance by avoiding unnecessary copies. · 86674115
  theos authored Apr 06, 2016
```
Added `copy` kwargs where missing.
```
  86674115
- Increased d2o test multiplicity. · 79d0153f
  theos authored Apr 06, 2016
  
  79d0153f
- Added dtype kwarg to __array__() method of distributed_data_object · d3fa8b86
  theos authored Apr 06, 2016
  
  d3fa8b86
05 Apr, 2016 1 commit
- Removed an unnecessary array copy in obj.copy() · 2aa97ad8
  theos authored Apr 05, 2016
  
  2aa97ad8
31 Mar, 2016 1 commit
- Fixed the setup.py after introducing d2o module. · 23345ccb
  theos authored Mar 31, 2016
  
  23345ccb
30 Mar, 2016 1 commit
- Moved d2o into dedicated subpackage. · b3cb661c
  theos authored Mar 30, 2016
  
  b3cb661c
29 Mar, 2016 1 commit
- Removed 'np'-datamodel from point_space, rg_space, los_response and power_operator. · 320ab0c7
  theos authored Mar 30, 2016
  
  320ab0c7
23 Mar, 2016 1 commit
- Added __array__ method to distributed_data_object. np.array(obj) is now fast! · 2d8eee09
  theos authored Mar 23, 2016
  
  2d8eee09
21 Mar, 2016 7 commits
- Merge branch 'issue/12/Remove-np-support' into 'master' · 7589b45b
  Theo Steininger authored Mar 21, 2016
```
Issue/12/remove np support



See merge request !6
```
  7589b45b
- Added small corrections. · 7068b092
  theos authored Mar 21, 2016
  
  7068b092
- Hp refactor with Theos changes · db16767c
  csongor authored Mar 21, 2016
  
  db16767c
- lm_space, gl_space: Fixed calc_power · 4d45da34
  theos authored Mar 21, 2016
```
gl_space: Fixed get_plot
```
  4d45da34
- lm_space: Fixed calc_transform and calc_smooth. · 6a059a6a
  theos authored Mar 21, 2016
  
  6a059a6a
- gl_space: Fixed calc_weight · 887c8e07
  theos authored Mar 21, 2016
  
  887c8e07
- nifty_lm.py: Fixed random number generators · 2aa4157a
  theos authored Mar 21, 2016
  
  2aa4157a
20 Mar, 2016 10 commits
- nifty_lm: Fixed plot routine · 30d17f0d
  theos authored Mar 21, 2016
  
  30d17f0d
- nifty_lm: Fixed np.array(d2o) casting. · 78ff4dbf
  theos authored Mar 21, 2016
```
Fixed dtype conversion.
```
  78ff4dbf
- nifty_lm: Adapted calc_smooth to d2o. · b37d01e4
  theos authored Mar 21, 2016
  
  b37d01e4
- lm_space: Fixed an error in the _cast_to_d2o · 45a8c42e
  theos authored Mar 21, 2016
  
  45a8c42e
- lm_space: Fixed calc_transform not doing np.array(d2o) casting · 468c8814
  theos authored Mar 21, 2016
  
  468c8814
- Updated lm_space.calc_dot in order to make it distribution_strategy independent. · 10735686
  theos authored Mar 20, 2016
```
Libsharp is not longer used.
```
  10735686
- nifty_lm fix: get_random_values returned numpy array for arg==None. · dcbb81ac
  theos authored Mar 20, 2016
```
Consolidated casting to the end.
```
  dcbb81ac
- lm_space: Fixed the cast method. · a965d4e8
  theos authored Mar 20, 2016
```
-> Switched naming to _cast_d2o
-> Removed use of np functions for complexity check!
```
  a965d4e8
- lm_space: Fixed the datamodel check · 504626ab
  theos authored Mar 20, 2016
```
nifty_power_indices: Fixed the methods for d2o compatiblity
```
  504626ab
- Only global_type distribution strategies are valid for hp/lm/gl at init time. · 22f9278e
  theos authored Mar 20, 2016
  
  22f9278e
19 Mar, 2016 1 commit
- Minor fixes · cbd33338
  csongor authored Mar 18, 2016
  
  cbd33338