v1.0.4

What's Changed

• Add system-message command by @caufieldjh in #435
• Add topic classifier by @caufieldjh in #437
• Add new food extraction template by @caufieldjh in #441
• Further development of the pathology extraction template by @caufieldjh in #444
• Update output options for multilingual analysis by @caufieldjh in #442
• Refactor file loading/parsing; add support for spreadsheets as input by @caufieldjh in #447

Full Changelog: v1.0.3...v1.0.4

v1.0.3

What's Changed

• Add template for MIRO checklist extraction by @caufieldjh in #432
• Add caching for mapper by @caufieldjh in #433
• Prep for v1.0.3 by @caufieldjh in #434

Full Changelog: v1.0.2...v1.0.3

v1.0.2

What's Changed

• Added CSV exporter by @timalamenciak in #426
• Handle HTTP 500 errors with PubmedClient by @caufieldjh in #428
• Move eliminate_empty function into utils by @caufieldjh in #429
• Replace textract with pymupdf by @caufieldjh in #430
• Prep for V1.0.2 by @caufieldjh in #431

New Contributors

• @timalamenciak made their first contribution in #426

Full Changelog: v1.0.1...v1.0.2

v1.0.1

What's Changed

• Add docs section on OpenAI-compatible Proxy Servers by @caufieldjh in #419
• Partial fix for #417 by @caufieldjh in #422
• Add JSON writer by @caufieldjh in #423
• Updates for v1.0.1 by @caufieldjh in #424

Full Changelog: v1.0.0...v1.0.1

v1.0.0

🎉 Highlights 🎉

• LLMs are now accessed through litellm, meaning OntoGPT may now be used with a large collection of API endpoints and local or alternative model providers. See the full list with the ontogpt list-models command.
• Local, open models may be downloaded and used through the ollama package.
• Numerous bugfixes
• Documentation updates
• Updates for the webapp

If something seems broken, please let us know! Open an issue here: https://github.com/monarch-initiative/ontogpt/issues

What's Changed

• Fixing failure to parse markdown, JSON formatting, and numbered lists by @caufieldjh in #394
• Add template for summarizing LinkML data validation reports by @caufieldjh in #400
• Improvements for the alzrd template by @caufieldjh in #399
• Further updates to the ADRD extraction template by @caufieldjh in #403
• Restructure Alz extraction template; split into two variants by @caufieldjh in #405
• Additional updates to the Alzheimers extraction template by @caufieldjh in #406
• Fix issue with early termination of multilanguage analysis runs by @caufieldjh in #408
• Add support for more models by @caufieldjh in #373
• Tests for new LLMClient by @caufieldjh in #410
• General cleanup: docs, code optimization, and project structure by @caufieldjh in #411
• Add option to truncate input_text in outputs by @caufieldjh in #413
• eliminate hardcoded gpt4-turbo from multilingual function by @leokim-l in #415
• Misc changes for 1.0.0rc2 by @caufieldjh in #416
• Prep for v1.0.0 release by @caufieldjh in #418

New Contributors

• @leokim-l made their first contribution in #415

Full Changelog: v0.3.15...v1.0.0

v1.0.0rc2

What's Changed

• Add option to truncate input_text in outputs by @caufieldjh in #413
• eliminate hardcoded gpt4-turbo from multilingual function by @leokim-l in #415
• Misc changes for 1.0.0rc2 by @caufieldjh in #416
• Enabled using alternative source for textract (#412)
• A small fix for pubmed-annotate

New Contributors

• @leokim-l made their first contribution in #415

Full Changelog: v1.0.0rc1...v1.0.0rc2

v1.0.0rc1

This is a pre-release. See below for a preview of changes.

What's Changed

• Fixing failure to parse markdown, JSON formatting, and numbered lists by @caufieldjh in #394
• Add template for summarizing LinkML data validation reports by @caufieldjh in #400
• Improvements for the alzrd template by @caufieldjh in #399
• Further updates to the ADRD extraction template by @caufieldjh in #403
• Restructure Alz extraction template; split into two variants by @caufieldjh in #405
• Additional updates to the Alzheimers extraction template by @caufieldjh in #406
• Fix issue with early termination of multilanguage analysis runs by @caufieldjh in #408
• Add support for more models by @caufieldjh in #373
• Tests for new LLMClient by @caufieldjh in #410
• General cleanup: docs, code optimization, and project structure by @caufieldjh in #411

Full Changelog: v0.3.15...v1.0.0rc1

v0.3.15

Highlights

• Four new templates: alz_treat, gene_extraction, storms, and onto_usage - see below for more details.
• The pubmed-extract and pubmed-annotate commands now allow setting the text segment size (in number of characters) with the max-text-length option. Many LLMs now allow for large contexts, so set this to 100000 or so to parse the full text of a scientific article (along with the --get-pmc option).
  • Example: ontogpt -vvv pubmed-annotate -t storms --get-pmc --model gpt-4o --limit 1 --max-text-length 100000 -o storms_test.yaml "36598999"

What's Changed

• Add template for alternate gene symbol extraction strategy by @caufieldjh in #390
• Template for Alzheimer's Disease extractions from literature; expose the max-text-length option for pubmed_annotate by @caufieldjh in #392
• Add ontology usage extraction template by @caufieldjh in #393
• Fixes #396 by @cmungall in #397
• Adding STORMS checklist by @cmungall in #395
• Bump version to 0.3.15 by @caufieldjh in #398

Full Changelog: v0.3.14...v0.3.15

v0.3.14

What's Changed

• Add custom template loader by @caufieldjh in #388
  • Custom templates in yaml may now be passed directly to ontogpt commands - no make needed.
• Updates for docs by @caufieldjh in #389

Full Changelog: v0.3.13...v0.3.14

v0.3.13

What's Changed

• Include only input-relevant named entities when producing output by @caufieldjh in #382
• More webapp updates - more models and all templates by @caufieldjh in #385
• Add notebook with two quick CLI and webapp examples by @caufieldjh in #386

Full Changelog: v0.3.12...v0.3.13