Commits · 84b6c1840fd55940b2bc442c7a3e7cd5b214950f · FSL / fdt · GitLab

Snippets Groups Projects

Apr 12, 2013
- Method to merge the result parts of Bedpostx. Call from post-proc script · 84b6c184
  Moises Fernandez authored 11 years ago
  
  84b6c184
- Change the log files structure · c345325c
  Moises Fernandez authored 11 years ago
  
  c345325c
Apr 11, 2013
- Fix: add end of method · d4a96533
  Moises Fernandez authored 11 years ago
  
  d4a96533
- Deleted structures not used · 7e28fdd4
  Moises Fernandez authored 11 years ago
  
  7e28fdd4
- Changed to use Dynamic Shared Memory · d498bdc5
  Moises Fernandez authored 11 years ago
  
  d498bdc5
- Postprocessing of the parts created by the BedpostX GPU version · 4a399c9a
  Moises Fernandez authored 11 years ago
  
  4a399c9a
- Changed to use fsl_sub and divide in X parts instead of slices · 8b5c17cf
  Moises Fernandez authored 11 years ago
  
  8b5c17cf
- Deleted set path to Dynamic library · f27dce7c
  Moises Fernandez authored 11 years ago
  
  f27dce7c
- Changed to give the chance to indicate the number of jobs to submit · 27be8bbc
  Moises Fernandez authored 11 years ago
  
  27be8bbc
- Changed to divide in subparts the jobs · 347095f7
  Moises Fernandez authored 11 years ago
  
  347095f7
- Changed to have a compiled version · 24733487
  Moises Fernandez authored 11 years ago
  
  24733487
- Changed to divide in subparts the jobs and to have a compiled version · db12f2c8
  Moises Fernandez authored 11 years ago
  
  db12f2c8
- Changed to use Dynamic Shared Memory · 14e23879
  Moises Fernandez authored 11 years ago
  
  14e23879
- Changed to save only the voxels of each new subpart · 28a169c5
  Moises Fernandez authored 11 years ago
  
  28a169c5
- Changed to have a compiled version. Don't need to compile each run time · 816b3031
  Moises Fernandez authored 11 years ago
  
  816b3031
- Changed to use Dynamic Shared Memory and fix curand with a number of randoms not multiplo of 2 · 17d8d4bc
  Moises Fernandez authored 11 years ago
  
  17d8d4bc
- Changed to use Dynamic Shared Memory and fix curand with a number of randoms not multiplo of 2 · 276947c0
  Moises Fernandez authored 11 years ago
  
  276947c0
- fix number of maximum number of parameters · 4d103eb9
  Moises Fernandez authored 11 years ago
  
  4d103eb9
- Changed to define maximums in the compiled version · 318e4dba
  Moises Fernandez authored 11 years ago
  
  318e4dba
- Changed to use Dynamic Shared Memory · 14fb5977
  Moises Fernandez authored 11 years ago
  
  14fb5977
- Changed to have a compiled version. Don't need to compile each run time · 2af5e1ba
  Moises Fernandez authored 11 years ago
  
  2af5e1ba
- Deleted dim_blocks.h: Deprecated, now in options/options.h · d27358fd
  Moises Fernandez authored 11 years ago
  
  d27358fd
- Changed to use Dynamic Shared Memory · 7f81c0cc
  Moises Fernandez authored 11 years ago
  
  7f81c0cc
- Deleted bedpostx_multigpu_SGE because now the MultiGPU version in SGE will be... · 557802c3
  Moises Fernandez authored 11 years ago
  
  Deleted bedpostx_multigpu_SGE because now the MultiGPU version in SGE will be controlled by fsl_sub and NGPU parameter in bedpostx_gpu
  557802c3
- Changed to use Dynamic Shared Memory · 0c2259f2
  Moises Fernandez authored 11 years ago
  
  0c2259f2
- Changed the way of divide set of voxels (before by slices). Now in X parts... · c65e6092
  Moises Fernandez authored 11 years ago
  
  Changed the way of divide set of voxels (before by slices). Now in X parts each one divided in subparts of a maximum number of voxels that are sent to the GPU in a iteretive fashion
  c65e6092
- Deleted splitter_gpu.cc splitter_multigpu.cc because now the division of... · c9d27fd6
  Moises Fernandez authored 11 years ago
  
  Deleted splitter_gpu.cc splitter_multigpu.cc because now the division of slices and voxels is make inside scripts and xfibres_gpu.cc
  c9d27fd6
Mar 06, 2013
- Now accept in the input full tensors (9 elements), not necessarily symmetric · f559cfcd
  Stamatios Sotiropoulos authored 12 years ago
  
  f559cfcd
Feb 28, 2013
- Change to don't repeat all slices in case of fail in one · 00ff6fdd
  Moises Fernandez authored 12 years ago
  
  00ff6fdd
- Changes in grad-nonlin functions · c38b202c
  Moises Fernandez authored 12 years ago
  
  c38b202c
- make faster mix parameters after repeat a vox because f0 · 38ab83a8
  Moises Fernandez authored 12 years ago
  
  38ab83a8
- make faster some methods that use thrust library · ddef95dc
  Moises Fernandez authored 12 years ago
  
  ddef95dc
- make faster mix parameters after repeat a vox because f0 · 804eef94
  Moises Fernandez authored 12 years ago
  
  804eef94
- changed the way of round some math operations · 2e881906
  Moises Fernandez authored 12 years ago
  
  2e881906
- new version Levenberg: several threads per voxel · 4abbf8cb
  Moises Fernandez authored 12 years ago
  
  4abbf8cb
- changed the pow(x,2) function with x*x · a5e7f009
  Moises Fernandez authored 12 years ago
  
  a5e7f009
- new version Levenberg: several threads per voxel · d4a9f9f7
  Moises Fernandez authored 12 years ago
  
  d4a9f9f7
Feb 22, 2013
- Remove warnings · 6b4a6a7d
  Stamatios Sotiropoulos authored 12 years ago
  
  6b4a6a7d
- Changes in grad-nonlin functions · c88a7cfa
  Stamatios Sotiropoulos authored 12 years ago
  
  c88a7cfa
Jan 18, 2013
- Use ARD for dstd · 8b9738f4
  Stamatios Sotiropoulos authored 12 years ago
  
  8b9738f4