sineha-mani 384c001825 feat(evaluation): remaining python-style evaluate entry points (#18) ...

## Summary
This change adds the remaining Python-style evaluate entry points on top
of #17.

Feature | New in #18?
-- | --
EvaluateTarget (Runnable-style target) | ✅
evaluate(client, EvaluateTarget, params) | ✅
evaluateExisting() / evaluateExistingAsync() | ✅
evaluate(client, experimentId, params) convenience overload | ✅
evaluateComparative() | ✅
evaluate(client, Pair(expA, expB), params) convenience overload | ✅
ComparativeEvaluator / comparativeEvaluator() | ✅
EvaluateExistingParams | ✅
EvaluateComparativeParams | ✅
ComparativeExperimentResults | ✅

References
[LSDK-241](https://linear.app/langchain/issue/LSDK-241/add-evaluate-to-langsmith-java-sdk-example-release)

<img width="1173" height="508" alt="Screenshot 2026-06-25 at 3 01 38 PM"
src="https://github.com/user-attachments/assets/a86df925-3657-4ad2-ba21-e250b9f5b4b7"
/>

## Test plan
- [x] `./gradlew :langsmith-java-core:test --tests
"com.langchain.smith.evaluation.*"` passes
- [X] Ran locally with a script

---------

Co-authored-by: Cursor <cursoragent@cursor.com>

2026-07-01 15:27:36 -07:00

..

src

feat(evaluation): remaining python-style evaluate entry points (#18)

2026-07-01 15:27:36 -07:00

build.gradle.kts

Revert "Build SDK"

2026-06-03 21:20:47 +02:00