-
Notifications
You must be signed in to change notification settings - Fork 2
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 performance tests #129
base: main
Are you sure you want to change the base?
Conversation
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.
An idea to capture the cycle count measurement would be to make it a function decorator. That way we could reuse it outside of nanoDFT
as I imagine it would be useful to define micro-benchmarks. I have a preference to break things down into smaller PRs but lets discuss.
filterwarnings = | ||
error | ||
ignore::DeprecationWarning |
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.
I wasn't familiar with these options, looks useful!
This branch is adding performance tests for
nanoDFT
.So far it is possible to just execute initial 5 tests with simple purest command
purest tests/test_benchmark_performance.py
.The code injecting the
ipu_cycle_count
into the_nanoDFT()
jitted method is WIP.The branch is going to:
nanoDFT()
performance evaluation withipu_cycle_count
pytest
for performance testingipu_cycle_count
code and move it toutils