v0.1.0

What's Changed

• Initialize premove-itn proof of concept by @aryamantodkar in #1
• Keep the project identity and scope consistent by @aryamantodkar in #2
• Add reusable engineering skills by @aryamantodkar in #3
• Freeze the premove-itn proof architecture by @aryamantodkar in #4
• Add deterministic Rust realization adapter by @aryamantodkar in #5
• Expand contextual model plan to all configured span kinds by @aryamantodkar in #6
• Define model labels and document dependency licensing by @aryamantodkar in #7
• Define immutable normalization contracts by @aryamantodkar in #8
• Add source-preserving tokenizer by @aryamantodkar in #9
• Freeze reviewed evaluation dataset by @aryamantodkar in #10
• Add strict Google TN dataset ingestion by @aryamantodkar in #12
• Add Google TN candidate extraction by @aryamantodkar in #13
• Add Google TN candidate validation by @aryamantodkar in #14
• Add Google TN sentence assembly by @aryamantodkar in #15
• Finish Google TN training records by @aryamantodkar in #16
• Organize Google TN dataset pipeline by @aryamantodkar in #17
• Add Google TN ingestion audit milestone by @aryamantodkar in #18
• Materialize Google TN accepted corpus shards by @aryamantodkar in #19
• Expand trusted Google TN equivalence by @aryamantodkar in #20
• Materialize the full Google TN accepted corpus by @aryamantodkar in #21
• Add deterministic dataset candidate selection by @aryamantodkar in #22
• Add Google TN candidate selection writer by @aryamantodkar in #23
• Add SGD context extraction by @aryamantodkar in #24
• Add deterministic SGD context audit by @aryamantodkar in #25
• Add SGD enrichment policy and trusted donor pools by @aryamantodkar in #26
• Separate data pipeline from runtime distribution by @aryamantodkar in #27
• Build compiled enrichment reservoir by @aryamantodkar in #28
• Select enrichment and assemble Dataset V1 by @aryamantodkar in #29
• Freeze Dataset V1 and add optimized DeBERTa Small BIO training by @aryamantodkar in #30
• Add a versioned Model V1 label contract by @aryamantodkar in #31
• Freeze the Model V1 semantic policy by @aryamantodkar in #32
• Revert BIO architecture to deterministic core by @aryamantodkar in #33
• docs: define epic branching strategy by @aryamantodkar in #34
• Fix digit-sequence edge cases by @aryamantodkar in #35
• fix: expand cardinal semantic coverage by @aryamantodkar in #36
• fix: expand date semantic coverage by @aryamantodkar in #37
• fix: expand TIME realization and representation equivalence by @aryamantodkar in #38
• fix: expand money realizer coverage by @aryamantodkar in #39
• fix: expand decimal and cross-kind edge coverage by @aryamantodkar in #40
• fix: harden electronic span validation by @aryamantodkar in #41
• fix: harden measurement realization and equivalence by @aryamantodkar in #42
• fix: harden ordinal realization and equivalence by @aryamantodkar in #43
• fix: harden phone realization and equivalence by @aryamantodkar in #44
• fix: expand punctuation realization aliases by @aryamantodkar in #45
• fix: harden whitelist matching boundaries by @aryamantodkar in #46
• fix: harden word realization edge cases by @aryamantodkar in #47
• fix: close generic realization edge cases by @aryamantodkar in #49
• feat: complete deterministic realizer edge coverage by @aryamantodkar in #50
• feat: merge contextual candidate selection MVP by @aryamantodkar in #75
• epic: finalize candidate generation and First Evaluation by @aryamantodkar in #78
• feat: publish inference-only model artifact workflow by @aryamantodkar in #80
• feat: add public contextual inference API by @aryamantodkar in #81
• Stage 3: make contextual inference installable by @aryamantodkar in #82
• Stage 4: add contextual ITN CLI by @aryamantodkar in #83
• Stage 5: polish public documentation and product surface by @aryamantodkar in #84
• Stage 6: automate CI and release gates by @aryamantodkar in #85
• Stage 7: certify v0.1.0 platform compatibility by @aryamantodkar in #86
• Release v0.1.0 to PyPI and GitHub by @aryamantodkar in #87

Full Changelog: https://github.com/premove-ai/premove-itn/commits/v0.1.0