v1.6.1

What's Changed

• fix: re-add og internal reprocess fn by @elboy3 in #830
• feat: Upload model after finishing training by @franz101 in #826

Full Changelog: v1.6.0...v1.6.1

v1.6.0

What's Changed

• fix(seq2seq): Include action info in data logger errors by @dcaustin33 in #825
• chore: Update PR template by @elboy3 in #829
• chore(deps): Update to pydantic v2 by @franz101 in #820

Full Changelog: v1.5.1...v1.6.0

v1.5.1

What's Changed

• fix: logged in as none and reprocess finetune fixes by @elboy3 in #824

Full Changelog: v1.5.0...v1.5.1

v1.5.0

What's Changed

• feat: remove cloud and login on import by @elboy3 in #823

Full Changelog: v1.4.2...v1.5.0

v1.4.2

What's Changed

• feat: generation peft by @jonathangomesselman in #822

Full Changelog: v1.4.1...v1.4.2

v1.4.1

What's Changed

• Fix misalignment between token offsets returned from the API and samples in the UI by @jonathangomesselman in #821

Full Changelog: v1.4.0...v1.4.1

v1.4.0

What's Changed

• feat: token refresh by @elboy3 in #818

Full Changelog: v1.3.10...v1.4.0

v1.3.10

What's Changed

• Decoder Only Generation Support by @jonathangomesselman in #813
• fix: ignore embs if not passed by @elboy3 in #819

Full Changelog: v1.3.9...v1.3.10

v1.3.9

What's Changed

• fix: better error messaging for no model outputs by @elboy3 in #816

Full Changelog: v1.3.8...v1.3.9

v1.3.8

What's Changed

• feat: s2s add high dep tokens by @bogdan-galileo in #809
• Clean up data formaters, removing need for passing Seq2SeqDataLogger by @jonathangomesselman in #810
• fix: s2s decoder only notebook by @bogdan-galileo in #811
• feat: add warnings in s2s for bad inputs by @bogdan-galileo in #812
• fix: disallow emb cols as metadata by @elboy3 in #814
• fix: bump datasets to 2.14.6 by @elboy3 in #815

Full Changelog: v1.3.6...v1.3.8