Skip to content

Changelog

0.6.0 (2026-05-11)

Features

  • add support for arize-phoenix v15.x.x (#56) (095f653)

0.5.0 (2026-05-07)

Features

  • add DatasetValidator for testset structural and content validation (#52) (cb7374e)
  • add evaluator composition factories (weighted, all-pass, any-pass) (#51) (8010f40)
  • add ResultCollector for experiment export, comparison and reporting (#53) (c775ce4)

Documentation

  • rebrand evalwire as framework-agnostic evaluation tool (#54) (5b695fd)

0.4.1 (2026-05-04)

Documentation

  • add concepts, guides, configuration, and troubleshooting pages (#48) (8b1af18)

0.4.0 (2026-04-23)

Features

  • add logo SVG and update theme colors in mkdocs configuration (#22) (67f44a9)

Bug Fixes

  • enhance package typing, logging, and schema evaluation (#28) (2a73529)
  • loosen langgraph version pin from ==1.1.6 to >=1.1,<2 (#32) (ebab441)
  • overwrite mode now deletes existing dataset before re-creating (#39) (eab726c)
  • resolve mutable defaults, narrow exception handling, and sys.modules leak (#33) (788c642)

0.3.1 (2026-03-31)

Documentation

  • update PyPI version badge to include cache parameter (29791dd)

0.3.0 (2026-03-31)

Features

  • add 7 evaluator factories with tests and updated dependencies (#13) (91b493e)
  • add verified PyPI project URLs (#17) (950f3e6)

Documentation

  • update README and docs to cover all 9 evaluator factories (#16) (5a82c1d)

0.2.2 (2026-03-30)

Bug Fixes

  • use per-thread persistent event loop to prevent 'Event loop is closed' (86f97ba)

0.2.1 (2026-03-30)

Bug Fixes

  • wrap async tasks for sync Phoenix client (#5) (9d370cc)

0.2.0 (2026-03-30)

Features

  • add pytest and pytest-mock to dev dependencies (6e55a38)
  • add runtime dependencies, extras, and CLI entry point (a97d4f6)
  • demo: auto-load .env via python-dotenv in run.py (2e996b1)
  • expose public API in package init (c8828ce)
  • implement built-in evaluators (top_k and membership) (d819763)
  • implement CLI upload and run commands (241d614)
  • implement DatasetUploader (c2f587f)
  • implement ExperimentRunner with auto-discovery (b2cba1d)
  • implement LangGraph node isolation helpers (2f7c62a)
  • implement setup_observability (3b9d67a)
  • implement TOML config loader (b877f2f)
  • initialize project structure with essential files and configurations (668cbff)
  • replace demo/requirements.txt with demo dependency-group in pyproject.toml (3c3b67e)
  • runner: implement concurrency via ThreadPoolExecutor and auto-create init.py (763c53c)

Bug Fixes

  • align uploader and runner to phoenix.Client flat API (>=13) (b768011)
  • ci: suppress unresolved-import for test_task_async.py in ty.toml (1ae7cbc)
  • error-handling: log exc_info on swallowed exceptions in uploader and runner (b82475a)
  • evaluators: guard top_k against None output when task failed (a4b96d8)
  • langgraph: annotate build_subgraph return type as CompiledStateGraph via TYPE_CHECKING (f512d54)
  • runner: switch to client.experiments.run_experiment namespaced API (5eb8e63)
  • types: resolve all ty type errors across package and tests (c0314df)
  • typing: annotate setup_observability return as TracerProvider and tighten dict types (0bc67a3)
  • uploader: switch to client.datasets.* namespaced API and fix overwrite delete step (f51d6d8)
  • uploader: use explicit list defaults for input_keys and output_keys (b71280d)
  • uploader: use real Phoenix 13.x API and type phoenix_client as Client (7b7b465)
  • use is_string_dtype to support pandas 3.x StringDtype in _load_csv (01c17ed)

Documentation

  • add lazy-import comment to _make_client and export build_subgraph/invoke_node (519a7b4)
  • add MkDocs setup with Material theme, mkdocstrings, and make targets (5bfd1a2)
  • write README and quick-start guide (f21a259)