fix(core): move log failure patterns to a shared module - #157
fix(core): move log failure patterns to a shared module#157Krishna-kg732 wants to merge 2 commits into
Conversation
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
Krishna-kg732
changed the title
|
cc : @abhijeet-dhumal , @jaiakash |
|
|
||
|
|
||
| def test_extract_failure_hint_openshift_pip_error(): | ||
| def testextract_failure_hint_openshift_pip_error(): |
There was a problem hiding this comment.
| def testextract_failure_hint_openshift_pip_error(): | |
| def test_extract_failure_hint_openshift_pip_error(): |
|
|
||
|
|
||
| def test_extract_failure_hint_generic_permission_error(): | ||
| def testextract_failure_hint_generic_permission_error(): |
There was a problem hiding this comment.
| def testextract_failure_hint_generic_permission_error(): | |
| def test_extract_failure_hint_generic_permission_error(): |
|
|
||
| def test_extract_failure_hint_suggestion_text(): | ||
| cuda = _extract_failure_hint("RuntimeError: CUDA error: device-side assert triggered") | ||
| def testextract_failure_hint_suggestion_text(): |
There was a problem hiding this comment.
| def testextract_failure_hint_suggestion_text(): | |
| def test_extract_failure_hint_suggestion_text(): |
|
Thanks @Krishna-kg732, Yeah this totally makes sense ! |
|
@Krishna-kg732 are you still working on this ? This PR needs to be rebased on main! |
Signed-off-by: krishna-kg732 <krishnagupta.kg2k6@gmail.com>
Signed-off-by: krishna-kg732 <krishnagupta.kg2k6@gmail.com>
3eb40f8 to
86cdcf0
Compare
|
done 🫡 |
Description
Moves the training-log failure pattern table and its matcher out of
trainer/api/monitoring.pyinto a newcommon/failures.py.This is groundwork split out of the optimizer client work so it can be reviewed
and merged on its own.
Related Issue
Groundwork for KEP-34 (Katib optimizer client). No separate issue.
Checklist
make test-python)make verify)git commit -s)Testing
Full suite green on this branch (472 passed, 14 deselected). The moved
FAILURE_PATTERNSandextract_failure_hintkeep their existing coverage intrainer/api/monitoring_test.pyandtrainer/api/sdk_contracts_test.py,including the
HF_CACHE_WRITE_ERRORordering case.