Skip to content

routines for performing rapidly-parallelized linear regression on nvidia GPUs

Notifications You must be signed in to change notification settings

clayspacelab/gridfitgpu

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

gridfitgpu

routines for performing rapidly-parallelized linear regression on nvidia GPUs

Rapidly determine the best model predictions to account for each of many measurements simultaneously. Requires a compatible NVidia GPU (see MATLAB documentation), which requires Windows or Linux.

Approximately 6-20x speedup, depending on compute capabilities. Identical results to using similar functions on CPU.

Required for GPU-accelerated functions in vistasoft (https://github.com/tommysprague/vistasoft_ts/tree/sprague_gridfit_updates)

About

routines for performing rapidly-parallelized linear regression on nvidia GPUs

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • MATLAB 100.0%