#5252nd of 60 in Data and scraping for AI
deepke
Toolkit for extracting entities, relations and attributes to build knowledge graphs
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.5k
- License
- MIT
- Last commit
- Jul 2026
- Last release
- Sep 2023
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
DeepKE is a PyTorch toolkit that trains and runs models for named entity recognition, relation extraction, attribute extraction and event extraction. It covers standard, few-shot, document-level and multimodal settings, with Hydra-configured run.py and predict.py scripts per task. A separate DeepKE-LLM environment adds LLM-based extraction, and an MCP tool service is also provided.
Who it is for: Researchers and engineers building knowledge graphs from Chinese or English text
Strengths
- Covers NER, relation, attribute and event extraction in one framework
- Supports few-shot, document-level and multimodal settings, with Chinese and English data
- Includes several published models (W2NER, LightNER, KnowPrompt, PRGC, PURE, ASP)
- Provides an online demo, docs, Colab notebook and a hosted MCP service
Weaknesses
- Pinned to Python 3.8, torch<=1.11 and transformers 4.26.0, which conflicts with newer stacks
- Last release 2023-09-21; pip package is marked not recommended
- Each task needs its own dataset download and config editing before training
- Separate environment needed for DeepKE-LLM; dataset links use a bare IP over HTTP
What it needs
- Needs PyTorch, transformers, hydra-core, wandb (optional)
- Models: BERT-style encoders, W2NER, LightNER, KnowPrompt, PRGC
Also in Data and scraping for AI
See all 60| Rank | Project | Score |
|---|---|---|
| 1 | doclingPython library that converts PDFs, Office files and more into structured text | 92 out of 100 |
| 2 | mineruParses PDFs, images and Office files into Markdown and structured JSON | 91 out of 100 |
| 3 | opendataloader-pdfPDF parser that outputs Markdown, JSON with bounding boxes, and Tagged PDF | 91 out of 100 |
| #4 | agent-reachCLI that gives coding agents read access to login-gated and Chinese web platforms | 89 out of 100 |
| #5 | markitdownPython library and CLI that converts office files, PDFs and audio to Markdown | 87 out of 100 |