-
-
Notifications
You must be signed in to change notification settings - Fork 183
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add GPU Cholesky Primitive #1059
Merged
Merged
Changes from 1 commit
Commits
Show all changes
77 commits
Select commit
Hold shift + click to select a range
12e5d8e
revised cholesky prim
rok-cesnovar ceee137
added comments & minor stuff
rok-cesnovar 7528d50
inverse fixes and added function to /prim
rok-cesnovar 8284c14
removed files
rok-cesnovar 834ff4a
now passing
rok-cesnovar 4c87e21
Merge remote-tracking branch 'upstream/develop' into gpu_cholesky_prim
SteveBronder 27beacc
Adds to docs, cleans up some code, use auto and const where possible
SteveBronder 87a67df
Fixes docs and changes name of test against cpu for fixed matrix
SteveBronder be110d2
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot 287ae57
forgot to include algorithm, we need to fix that lint check
SteveBronder 898ebca
Merge branch 'gpu_cholesky_prim' of https://github.com/bstatcomp/math…
SteveBronder 4fdb965
include algorithm again
rok-cesnovar 36bea8b
remove auto on return type
rok-cesnovar c126ceb
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot 6c0daff
move check for square and symmetric to top of cholesky decompose prim
SteveBronder 2d00cdf
[Jenkins] auto-formatting by clang-format version 6.0.0 (tags/google/…
stan-buildbot e8f9540
Move STAN_OPENCL check to above includes for both tests
SteveBronder 6b57d85
Remove unneded comments, removes the call to the zero kernel and fill…
SteveBronder 1a4a17b
fixed the zeroing in the kernel
rok-cesnovar 36a6d4a
moved the recursion or kernel step to a separate function
rok-cesnovar 10c4f59
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot a781a96
Merge branch 'develop' into gpu_cholesky_prim
rok-cesnovar 88c26a0
using operators for * and -
rok-cesnovar cc187e6
added tests for edge cases of lower_tri_inverse
rok-cesnovar 0115c85
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot 6bea215
Merge remote-tracking branch 'upstream/develop' into gpu_cholesky_prim
SteveBronder b3ef66d
Merge branch 'gpu_cholesky_prim' of https://github.com/bstatcomp/math…
SteveBronder 1c9bf89
Adds size checks to the cholesky and removes the part in the docs abo…
SteveBronder 763aca2
merge to dev
SteveBronder 021d87d
merge to dev
SteveBronder 0422c04
adds floor and uses half the rows of m in chol for the starting block…
SteveBronder c120223
Merge pull request #8 from bstatcomp/gpu-chol-prim-floor
SteveBronder 5bc8b71
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot 774aef5
merge to develop
SteveBronder 58666aa
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot da52c49
Remove extra ifdef in prim
SteveBronder dc14da2
Remove extra ifdef in prim
SteveBronder a736014
Removes tuning parameters from gpu cholesky and recursion
SteveBronder 6649984
Adds argument in cholesky_gpu for the min block size
SteveBronder 7d68799
Merge commit 'fb2cc51188b0171d70d63d5ef8be44998f5b3814' into HEAD
yashikno f557e47
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot d71e4a6
replace | with || in if for cholesky recursion
SteveBronder d9675c5
Fix docs
SteveBronder 775f7e4
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot 0be4541
Restructure GPU cholesky to get rid of the explicit recursion functio…
SteveBronder 1e90855
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot d461d06
1. Places recursive cholesky in the internal namespace. This
SteveBronder bb34aec
Merge branch 'gpu_cholesky_prim' of https://github.com/bstatcomp/math…
SteveBronder d7d09e7
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot d3bd266
Fix ifdef in chol prim
SteveBronder 1b65f2a
Fix docs to match code
SteveBronder 558f4f7
Changes the min_block logic to be a bit more clever. Catch local size…
SteveBronder 85eb0f3
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot 907f1ec
Minor comment cleanups, remove excessive auto, use max_thread_block_…
SteveBronder dc24e9f
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot 2f8be5b
Merge remote-tracking branch 'upstream/develop' into gpu_cholesky_prim
SteveBronder 89ebee8
Merge remote-tracking branch 'upstream/develop' into gpu_cholesky_prim
SteveBronder 7a146e4
Merge remote-tracking branch 'upstream/develop' into gpu_cholesky_prim
SteveBronder fbb97a3
Update so that Cholesky pulls the required size to run the Chol kerne…
SteveBronder f594437
Make a catch for old devices with small max thread block size for cho…
SteveBronder 6c26574
[Jenkins] auto-formatting by clang-format version 5.0.2-svn328729-1~e…
stan-buildbot d2285e0
Hard coded values for Cholesky tuning are now moved to the opencl con…
SteveBronder 8b04db9
Update Opencl context
SteveBronder 6bc3eba
[Jenkins] auto-formatting by clang-format version 5.0.2-svn328729-1~e…
stan-buildbot 749e21c
update context and cholesky docs
SteveBronder 1d8a707
Merge branch 'gpu_cholesky_prim' of https://github.com/bstatcomp/math…
SteveBronder 89d1c21
removed zeroing, changed to >= in the size check
rok-cesnovar 9215ebc
Fix names for opencl cholesky and add tests for tuning parameters
SteveBronder a0b4597
Merge commit '70edefd8b0f009e3d657f9488fe4f1cedb823ba1' into HEAD
yashikno dcc8c62
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot 32dd8ca
forgot to include vector in opencl chol test
SteveBronder f611fc7
forgot to remove doc for opencl_context.tuning_opts()
SteveBronder 15fbbe1
prim was not using the return value
rok-cesnovar b704bce
added zeroing for multiply(Nx0,0xM)
rok-cesnovar 08b19f0
[Jenkins] auto-formatting by clang-format version 5.0.0-3~16.04.1 (ta…
stan-buildbot a3f8881
Reduce and broaden chol test
seantalts 96118c8
Fix runTests.py argument parsing
seantalts File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Reduce and broaden chol test
- Loading branch information
commit a3f8881dcbccf6f1da1ca6556256867f71c8ceaa
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@seantalts for the tests here should I just do two for each? Otherwise this takes a minute and not sure if all the combinations are needed
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah, we can reduce some of these. A minute isn't too bad though - the distribution tests take 12 hours, so these aren't really a bottleneck yet :P
What about:
the ones at the bottom are doing more traditional testing where you try to look for edge cases and boundaries that you know the implementation will have to deal with.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah sure! I'm at work but if you can copy/paste the above over the current test now to let it run that would be rad. Else I can do it when I get home from work
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done and pushed! I also had to fix the runTests.py script - the filtering thing wasn't working right. if you specified it twice as the help suggested, only the latest one would stick.