[Performance] Avoid cuda sync in postprocess of LLM decoding #9011
Codecov / codecov/project
failed
Oct 28, 2024 in 0s
52.78% (target 58.00%)
View this Pull Request on Codecov
52.78% (target 58.00%)
Details
Codecov Report
Attention: Patch coverage is 28.57143% with 5 lines in your changes missing coverage. Please review.
Project coverage is 52.78%. Comparing base (
81f5ab5) to head (53e229a).
Report is 1 commits behind head on develop.
| Files with missing lines | Patch % | Lines |
|---|---|---|
| paddlenlp/generation/utils.py | 0.00% | 4 Missing |
| paddlenlp/generation/logits_process.py | 66.66% | 1 Missing |
Additional details and impacted files
@@ Coverage Diff @@
## develop #9011 +/- ##
===========================================
- Coverage 52.90% 52.78% -0.13%
===========================================
Files 661 661
Lines 107069 106948 -121
===========================================
- Hits 56650 56453 -197
- Misses 50419 50495 +76 ☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.
Loading