Release 0.1.33

Features

• extended metadata for catalog sources on data (d42419d)
• multi-arity at pipeline level (692b5a4)
• multipipeline retail example (feba0b5)

Bug Fixes

• example auto-discovery for nested/distributed directories (ec696a4)
• pythonic pipelines rerun uv sync on all builds/runs. (61a8839)
• resolve CI issue with UV cache (007fe80)

Documentation

• add xdoc (080baea)
• xdocs adjustments (1563684)

Release 0.1.32

Features

• ability to dry-run for preflight analysis (fe61f51)
• add dataset-datatype constraints to disallow nested schemas on data storage types that do not support nesting/non-primitive types (a0609be)
• add new MTG atlas example project (8bc315c)
• additional comparison pipeline (2bae1f4)
• additional dual efcore-python testing pipeline (20286c8)
• allow for gradient pre-flight checks for better unit testing (ad8947d)
• allow for nodata (unit) node input and ouput (3c7874e)
• allow for sync node outputs (fe1778c)
• allow plaintext description on pipeline nodes (83fd041)
• allow schemas to have required members (31dc519)
• begin testing framework (074f37e)
• better analyzer for missing pipeline requirements (569816e)
• catalog-level definition of inspection levels (c6ef393)
• completed efcore plugin refactor (769952d)
• configuration options for metadata providers (4eb2596)
• consolidate setup into FlowthruApplication builder, migrate critical code from test-flights to library code (1fbc156)
• create centralized SerializedLabel anno for universal label serialization (a6ca4e1)
• create oracle text full/partial embedding dataset (8ea8e29)
• custom ratchet testing architecture (b9ce539)
• distributed catalog entries (a3de8af)
• dummy pipeline template (2afd606)
• efcore extension (085f8d0)
• embedding pipeline example (e0d5276)
• enum serialization annotations (a3eb127)
• flexible & extensible catalogentries (e676bc1)
• force increment change (06d3522)
• functional ml.net wrapper start (958b29a)
• further mast coverage (b276d6c)
• generic file catalog entry (e9794aa)
• improved DAG metadata output for full/active on run (c3b98e9)
• initial library setup (e1312a3)
• internal umap timing reports (7f79ba4)
• json data catalog implementation (0918226)
• kedro iris starter (8859c17)
• magic AST (94b4d99)
• magic atlas accurate umap (488b81c)
• magic atlas embedding distribution analysis (3871d2d)
• magicatlas k-means clustering analytics (e1f756d)
• magicatlas k-means clustering analytics (cebd25c)
• mast error code system, superpower install (4578f93)
• mast revamp, unit tests (7fb7611)
• mast testing, ast (55c632c)
• metadata analysis in JSON and Mermaid form (ce56eae)
• move x-validate to data science portion to spice dag (c9523aa)
• nuget publish action flow (71c9e49)
• pca clustering, pipeline (b806414)
• pipeline builder addnode overloads (a619aeb)
• pipeline slicing (a16fecc)
• pure kedro spaceflights example (cc5b8ac)
• pythonic nodes (#6) (b0289a4)
• required parameter support for schemas (ac9aec1)
• retail data example (ecfdcda)
• run all pipelines as unified dag (68da6ab)
• scryfall card processing (79aa48b)
• service-based API surface changes (925f38d)
• simplify node pipeline registration type param requirements to remove redundant information (69efa58)
• sparse matrix optimizations (ebd6988)
• specify empty EFCore tables as errors (802326c)
• specify empty EFCore tables as errors (b622aeb)
• starter umap comparison tests (8dbc537)
• strategized umap (c69eb27)
• stronger onboarding processes (#8) (8939c37)
• templating process (acba958)
• testing process for template pipeline creation (501ddae)
• tiebreaking jiggle on atlas umap, to break up single-value columns (5bb4522)
• umap first stab (50f0552)
  *...