chore(tests): Mock LLM in tests for PRs #87
Conversation
This file contains hidden or 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
richiejp
force-pushed
the
mock-llm
branch
2 times, most recently
from
7c3e46e to
f601b95
Compare
|
I'm OK with this, but from one end the current test suite its very nice to give a glimpse on the model quality for agentic actions, mmmh.. maybe we can split the current test in a separate workflow which is not gating PRs? |
|
Yeah can try that, I'm a bit unclear on what GitHub supports in that regard. Some other options I thought about:
|
richiejp
force-pushed
the
mock-llm
branch
3 times, most recently
from
745d35d to
828bffb
Compare
|
OK, there are now two workflows, one is not required for PRs to pass. |
|
Seem to be missing env vars on the new workflow |
richiejp
force-pushed
the
mock-llm
branch
4 times, most recently
from
1076227 to
24a3286
Compare
|
OK, fixed a number of bugs in the tests, so both mock and real LLM tests should work |
This saves time when testing on CPU which is the only sensible thing to do on GitHub CI for PRs. For releases or once the commit is merged we could use an external runner with GPU or just wait. Signed-off-by: Richard Palethorpe <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
Mock LLM in unit tests so that the CI for PRs can pass in a reasonable time
For testing releases I think real LLMs should be used and on a variety of hardware, but it requires time and other resources.
EDIT:
This saves time when testing on CPU which is the only sensible thing
to do on GitHub CI for PRs. For releases or once the commit is merged
we could use an external runner with GPU or just wait.
This reduces the time to <10mintues from about 40-50. Most of the time is now taken installing docker and downloading images. I don't think there is a way to reduce that time without moving off of GitHub's default runners.