Skip to content

feat: Fact tracing benchmark [WIP] - #278

Merged
dilyabareeva merged 66 commits into
mainfrom
fact_tracing_rebase
Apr 27, 2026
Merged

feat: Fact tracing benchmark [WIP]#278
dilyabareeva merged 66 commits into
mainfrom
fact_tracing_rebase

test: increase coverage

4d4041f
Select commit
Loading
Failed to load commit list.
Codecov / codecov/project succeeded Apr 27, 2026 in 0s

94.95% (-1.20%) compared to c178d32

View this Pull Request on Codecov

94.95% (-1.20%) compared to c178d32

Details

Codecov Report

❌ Patch coverage is 89.55224% with 70 lines in your changes missing coverage. Please review.
✅ Project coverage is 94.95%. Comparing base (c178d32) to head (4d4041f).
⚠️ Report is 1 commits behind head on main.

Files with missing lines Patch % Lines
quanda/utils/common.py 58.92% 23 Missing ⚠️
quanda/explainers/wrappers/dattri_influence.py 81.81% 8 Missing ⚠️
quanda/benchmarks/downstream_eval/_fact_tracing.py 86.04% 6 Missing ⚠️
quanda/metrics/downstream_eval/tail_patch.py 93.68% 6 Missing ⚠️
quanda/utils/tokenization.py 80.64% 6 Missing ⚠️
quanda/benchmarks/config_parser.py 93.75% 5 Missing ⚠️
quanda/metrics/downstream_eval/recall_at_k.py 86.66% 4 Missing ⚠️
quanda/metrics/downstream_eval/mrr.py 89.28% 3 Missing ⚠️
quanda/benchmarks/downstream_eval/recall_at_k.py 88.88% 2 Missing ⚠️
quanda/benchmarks/downstream_eval/tail_patch.py 92.00% 2 Missing ⚠️
... and 4 more
Additional details and impacted files 
@@            Coverage Diff             @@
##             main     #278      +/-   ##
==========================================
- Coverage   96.14%   94.95%   -1.20%     
==========================================
  Files          64       71       +7     
  Lines        3010     3625     +615     
==========================================
+ Hits         2894     3442     +548     
- Misses        116      183      +67

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
View more details on Codecov