Releases: CambioML/uniflow-llm-based-pdf-extraction-text-cleaning-data-clustering
Releases · CambioML/uniflow-llm-based-pdf-extraction-text-cleaning-data-clustering
0.0.11
What's Changed
- add classification rater by @goldmermaid in #75
- Add pre-commit and github action checks by @jojortz in #76
- add splitter config and SplitterOpsFactory to ExtractPDFFlow by @jojortz in #78
- refactor rater config and add azure openai server by @CambioML in #81
- bump up version to 0.0.11 by @goldmermaid in #82
New Contributors
Full Changelog: 0.0.10...0.0.11
0.0.10
What's Changed
- fix model_server missing import packages by @frank-suwen in #69
- rename linear op, flow, and transform to copy. move node to uniflow root directory by @goldmermaid in #70
- add Context, GuidedPrompt into uniflow init.py for easy import by @goldmermaid in #73
- Add markdown header splitter by @SayaZhang in #71
- Add markdown and ipynb support by @CluckRookie in #72
New Contributors
- @frank-suwen made their first contribution in #69
- @SayaZhang made their first contribution in #71
Full Changelog: 0.0.9...0.0.10
0.0.9
What's Changed
- Refactor to extract and transform flow with pipeline interface by @goldmermaid in #48
- Clean configs, update README, and polish notebooks @jojortz in #49, #50, #51
- Add Nougat PDF loading #54, #56
- More refactor by @goldmermaid #59
- Refactor by @boqiny #61
- Readme update, bitsandbytes, and table left align by @C-sirui #63
0.0.8
What's Changed
- Remove legacy implementation of uniflow, refactor config, add README, and fix server deadlock issue @goldmermaid in #37
- Add support for batch inference @goldmermaid in #38
- Clean model_server interface and add retry capability @jojortz in #39
- Polish examples and README @goldmermaid in #40, #41, #42, #43, #44
- add Pydantic input classes Context and GuidedPrompt and make few-shot default for all flows @jojortz in #46
v0.0.6
v0.0.5
What's Changed
- Updated flows to all run through the
Client
interface by @jojortz in #15 - Add HTML input capability for
DataGenTextFlow
by @jojortz in #22 - Replace
print
withlogging
by @jojortz in #16 #20 - Unified input for all flows to be an input file. File processing is all handled within the flow now. By @jojortz in #27
- Update the README by @jojortz and @goldmermaid in #21 #24 #25 #28