Skip to content

Develop #19

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.

Sign up for GitHub

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

Jump to bottom
Open
wants to merge 177 commits into
base: main
Choose a base branch
from
Open

Develop #19

wants to merge 177 commits into from

Conversation

mattteochen
Copy link
Owner

Before submitting
  • Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
  • Did you read the contributor guideline, Pull Request section?
  • Did you make sure to update the docs?
  • Did you write any new necessary tests?

What does this PR do?

Fixes # (issue).

PR review

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

Make sure you had fun coding 🙃

mattteochen and others added 30 commits July 2, 2024 17:25
@mattteochen
Backend optimized, naive search setup
6605ac9
@mattteochen
Merge branch 'Lightning-AI:main' into looking-around
7cbecfd
@mattteochen
Single trace region placement impl
4ec2fa0
@mattteochen
Serial exaustive search
8d084a8
@mattteochen
Serial greedy search
7781d75
@mattteochen
Extended incremental strat with a fusion try after the greedy search …
99eb302 
…and comparing both with the priority approach
@mattteochen
Fixed key error after fusion_pass (symbol deleted by CSE)
2dbef84
@mattteochen
Removed from tracking
8c4fd7d
@mattteochen
Removed from tracking and updated tests
4651f26
@mattteochen
Merge branch 'main' into automatic-backends-placement
83a1bea
@mattteochen
Removed import
f09f9b3
@mattteochen
Change timing function
17df099
@mattteochen
Cleaned impl and updated debug files generation
bdca10b
@mattteochen
Moved already present fusion regions count at FusionOperator class le…
a9f0dbd 
…vel / Incremented optimzier benchmark iters
@mattteochen
Wip
3bfb690
@mattteochen
Added torch empty cache during benchmark execution / added trace prov…
6f93cba 
…enance to optimizer object
@mattteochen
Wip fuser ex focus
0945159
@mattteochen
Fusion executor placements / runtime vs memory placements options
24ad509
@mattteochen
Updated test model
d94e841
@mattteochen
Merge branch 'main' into automatic-backends-placement
77b7871
@mattteochen
Enhanced bw trace placement seach / support for more models (#3)
15f51bf
@mattteochen
Updated test
0f7ced3
@mattteochen
Fixed bad list index / removed print
579888d
@mattteochen
Disabled graphviz / modified test runner
3528e6c
@mattteochen
Using user defined executor list or default as unique executors ref i…
f67bb1c 
…n autotuner
@mattteochen
Computing bw traces taking in consideration every fw traces options (…
5f76bcf 
…from different FusionExecutors) (#4)
@mattteochen
Before transform_for_execution executors placement autotune / nvsig…
5dedbff 
…ht benchmark interface / nanoGPT model test (#5)
@mattteochen
Fixed remat visual timing / Switched towards memory strat / General…
15f914e 
… cleanup (#6)
@mattteochen
Testing different compile options for nvfuser (#7)
89bba8d
@mattteochen
Benchmarking different compile options for nvFuser (#8)
5be42dc
@mattteochen
Copy link
Owner Author

nvmath should be installed https://docs.nvidia.com/cuda/nvmath-python/latest/getting-started.html#install-from-pypi for CI tests.

@mattteochen
Enhanced logs description
cdc9157
@mattteochen
Merge branch 'main' into develop
3990619
@mattteochen
Small fixes to align main
6954062
@mattteochen
Fixed OOM errors during trace benchmarks leading to a premature end o…
87a79fd 
…f the tuning process
@mattteochen
Handled nvmath missing installation
75b856d
@mattteochen
Prev commit
f2b934b
@mattteochen
Prev commit
42401b1
@mattteochen
Added comment
02f5038
@mattteochen
Updated test runner
fadafe4
@mattteochen
Log applied executors
e4707fa
@mattteochen
Torch timer for benchmarks
b728ab6
@mattteochen
Doc
bb30805
@mattteochen
Updated Anyproxy hash / updated test runner file
191f403
@mattteochen
Formatter
7f64dbf
@mattteochen
Restored manual benchmark configuration / added env var for test runner
350c451
@mattteochen
Disabled flag
580050a
@mattteochen
Updated doc
8bbe1ec
@mattteochen
Autotuner for jit with no autograd
df0469e
@mattteochen
Added CUDA barrier for unit test
7f57584
@mattteochen
Integrated autotuner in benchmark script
82017d2
@mattteochen
Added missing flag
26044a3
@mattteochen
Fixed comments
0a17128
@mattteochen
Removed comment
afb5637
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@mattteochen