# ConardLi/easy-dataset

A powerful tool for creating datasets for LLM fine-tuning 、RAG and Eval

Repository: https://github.com/ConardLi/easy-dataset
Canonical: https://ross.abutalabs.com/products/easy-dataset
Homepage: https://docs.easy-dataset.com
Language: JavaScript
License: NOASSERTION
License Family: other
Topics: dataset, javascript, llm, fine-tuning, rag
Last push: 2026-05-01T15:03:32+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 80, release rhythm 78, longevity 39
- inputs: {"age_days": 547, "days_push": 124, "days_rel": 146, "gap_med": 8.5, "n_releases_24m": 31}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 14833, forks 1524 (observed 2026-08-28T04:11:07.759456+00:00)

## What it is
Easy Dataset is a self-hosted web application for building high-quality structured datasets for LLM fine-tuning, RAG, and model evaluation. It converts domain documents (PDF, Markdown, DOCX, EPUB, etc.) into question-answer datasets via intelligent segmentation, AI-assisted generation, labeling, export, and evaluation workflows.

## Use cases
- create fine-tuning datasets from domain documents
- generate QA pairs from PDFs for LLM training
- build evaluation test sets for vertical domain models
- convert datasets between fine-tuning formats
- manage and label large batches of generated questions
- evaluate RAG recall and post-fine-tune model performance
- construct COT reasoning data for fine-tuning

## When to choose
- you need to turn domain documents into structured training or eval datasets
- you want a GUI-driven pipeline covering parsing, chunking, generation, labeling, and export
- you need dataset formats for common fine-tuning frameworks and RAG evaluation

## When to avoid
- you only need a small one-off script to transform existing JSONL data
- you require a fully automated headless data pipeline without a UI
- you need a permissively licensed library to embed in closed-source products (AGPL-3.0)

## Facets
- artifact type: application
- maturity: active
- function: data-generation, etl, rag, llm-training, prompt-engineering, pdf, nlp
- domain: large-language-models, machine-learning, data-science, artificial-intelligence
- platform: self-hosted, cross-platform
- tags: fine-tuning-datasets, dataset-creation, document-parsing, text-chunking, qa-generation, model-evaluation, data-labeling, agpl, retrieval-augmented-generation, web-server, docker, nodejs

## Member repositories
- ConardLi/easy-dataset (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:07.759456+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:07:08.315287+00:00, confidence not recorded.
  - readme: https://github.com/ConardLi/easy-dataset (fetched 2026-08-28T04:11:07.759456+00:00, sha 05d9dce95684)
  - homepage: https://docs.easy-dataset.com (fetched 2026-08-29T08:05:33.413295+00:00, sha c2233544658e)
  - site_page: https://docs.easy-dataset.com/ji-chu-gong-neng/quickstart (fetched 2026-08-29T08:05:33.423294+00:00, sha 50d8514d9283)
  - site_page: https://docs.easy-dataset.com/ji-chu-gong-neng/publish-your-docs (fetched 2026-08-29T08:05:33.425168+00:00, sha 78585e972f97)
- Data as of 2026-08-30T08:39:29.467469+00:00.