-
Notifications
You must be signed in to change notification settings - Fork 12
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
Added compile fixture #37
base: main
Are you sure you want to change the base?
Conversation
219f70f
to
c1df1e3
Compare
1571c22
to
1a386ed
Compare
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.
Moved here
@pranavm-nvidia I will continue including details abt. failing test cases + MLIR dumps as I continue to enable |
Could you also include the name of the failing test? |
b3ac379
to
f424b2b
Compare
|
||
|
||
@pytest.fixture(params=["compile", "eager"]) | ||
def compile_fixture(request): |
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.
compile_fixture
should be able to specify which args are constants / parameters. For Q/DQ, scale should be Parameters.
I do not see a lot of value in testing so many low-level APIs in compile mode. What is the time impact on L0? |
f032baa
to
114ddf7
Compare
114ddf7
to
e146aa1
Compare
This MR adds a compile fixture to our integration test suite so that we can test ops both lazily and using the compiler API. Addresses #256.