v0.1.0.post20250930

What's Changed

• typo(app.py): update arXiv badge by @tpoisonooo in #11
• Add citation by @ChenZiHong-Gavin in #12
• fix: fix parameter passing error & add logger by @ChenZiHong-Gavin in #14
• refactor(graphgen): update imports and adapt GraphGen class by @Tendo33 in #15
• feat(webui): support M_synth & M_train from different providers by @ChenZiHong-Gavin in #20
• Added the Dockerfile for the GraphGen by @Nick1200000 in #24
• add hugging face in readme by @royal-dargon in #32
• typo(webui): renaming SiliconCloud to SiliconFlow by @tpoisonooo in #34
• ci: add pre-commit by @ChenZiHong-Gavin in #36
• Add temporary COT data synthesis implementation by @ChenZiHong-Gavin in #35
• Revert "Add temporary COT data synthesis implementation" by @ChenZiHong-Gavin in #39
• feat: filter think tag when using reasoning models by @ChenZiHong-Gavin in #40
• Implemented web search. by @ChenZiHong-Gavin in #41
• feat: add cot data generation pipeline by @ChenZiHong-Gavin in #42
• feat: support alpaca, sharegpt & chatml output format by @ChenZiHong-Gavin in #43
• ci: sync-demo by @ChenZiHong-Gavin in #44
• fix(webui): sync gradio demo by @ChenZiHong-Gavin in #47
• feat: support Reader classes by @ChenZiHong-Gavin in #50
• Add Splitter classes by @ChenZiHong-Gavin in #51
• Optimize gradio by @ChenZiHong-Gavin in #53
• ci: push-to-ms by @ChenZiHong-Gavin in #54
• feat: add log RichHandler & RotatingFileHandler by @ChenZiHong-Gavin in #55
• tests: add e2e tests by @ChenZiHong-Gavin in #56
• Refactor KG builder by @ChenZiHong-Gavin in #52
• refactor: refactor generate pipeline by @ChenZiHong-Gavin in #57

New Contributors

• @Tendo33 made their first contribution in #15
• @Nick1200000 made their first contribution in #24
• @royal-dargon made their first contribution in #32

Full Changelog: 2025042...v0.1.0.post20250930

20250422

What's Changed

GraphGen is a framework for synthetic data generation guided by knowledge graphs, designed to tackle challenges for knowledge-intensive QA generation.

• By uploading your text chunks (such as knowledge in agriculture, healthcare, or marine science) and filling in the LLM API key, you can generate the training data required by LLaMA-Factory and xtuner online.
• We will automatically delete user information after completion.

New Contributors

• @tpoisonooo made their first contribution in #1
• @ChenZiHong-Gavin made their first contribution in #8

Full Changelog: https://github.com/open-sciencelab/GraphGen/commits/20250422