feat(text-metrics): split qa_accuracy - #645

Closed

davidberenstein1957 wants to merge 13 commits into

feat/vlm-pr-2-infrastructurefrom

feat/vlm-pr-3a-qa-accuracy

Commits on May 8, 2026

feat(infrastructure): add VLM base classes and utilities
davidberenstein1957
committed
7054e53
fix(infra): scope metrics exports to infra-only symbols

davidberenstein1957
and
cursoragent
committed
d4509da

Commits on Jun 2, 2026

fix(evaluation): VLM infra review fixes for stacked PR #638

davidberenstein1957
and
cursoragent
committed
8d1c8e3
feat(text-metrics): split qa_accuracy into dedicated PR branch
davidberenstein1957
committed
7f0966c
feat(text-metrics): export QAAccuracyMetric in split branch

davidberenstein1957
and
cursoragent
committed
f2d5c5a
fix(metrics): qa_accuracy keyword-only aggregation

davidberenstein1957
and
cursoragent
committed
e7fca2e

Commits on Jun 4, 2026

fix(metrics): unblock CI sync and VLM metric init

davidberenstein1957
and
cursoragent
committed
03677c4
fix(ci): lint/docstrings and stack-appropriate VLM tests

davidberenstein1957
and
cursoragent
committed
1e36ec2
fix(ci): ruff on infra VLM test template

davidberenstein1957
and
cursoragent
committed
de1cc64
chore: drop local-only scripts from PR scope

davidberenstein1957
and
cursoragent
committed
2dc8924

Commits on Jul 2, 2026

fix(metrics): simplify TorchMetrics enum docstring

davidberenstein1957
and
cursoragent
committed
a4498d9
fix(metrics): document Enum boundary in TorchMetrics docstring

davidberenstein1957
and
cursoragent
committed
232ec87
fix(ci): lint, format, and benchmark registration
davidberenstein1957
committed
77a4886