DeepKE

Why we included this project

DeepKE brings together named-entity recognition, relation extraction, and attribute extraction in a single PyTorch toolkit aimed at knowledge graph construction. It ships with Chinese and English checkpoints you can run directly, so you can extract structured facts from text without training a model from scratch. The repo also handles harder cases like few-shot and document-level extraction, and it includes a low-resource data preparation path for building your own labeled sets. If you're populating a knowledge graph from raw text, DeepKE offers a coherent starting point with examples detailed enough to adapt to your own schema.