datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
CodeGraph
CodeGraph
An open-taxonomy, Wikidata-grounded semantic knowledge graph over 142M source files.
CodeGraph annotates 144,910,008 source files from Stack-Edu, across 14 programming languages, along four orthogonal semantic axes — application domains, algorithms (with category and asymptotic complexity), programming paradigms, and design patterns — and grounds the resulting concept vocabulary in Wikidata. The result is a typed property graph of 158.2M nodes and 1.02B edges that… See the full description on the dataset page: https://huggingface.co/datasets/halykoss/CodeGraph.mtpnet_graph_models
模型训练过程汇总[该仓库只含有graph model的训练过程]
本仓库采用扁平化的目录结构和标签系统来组织模型，具体说明如下：
仓库结构
一级目录：直接以模型名称-数据集，例如 ResNet-CIFAR-10、GraphMAE_QM9-Cora 等
二级目录：包含该模型在该数据集下的不同训练任务或变体，例如 normal、noisy、backdoor_invisible 等
训练过程目录结构：每个模型目录下包含：
scripts/：存放模型相关代码和训练脚本
epochs/：存放模型训练过程和权重文件
每个epoch的权重文件(model.pth)和embedding(.npy)
dataset/：模型需要的数据集
仓库结构展示
文件结构展示
CodeGraphIf you find our work useful or helpful for your R&D works, please feel free to cite our paper as below.
@misc{tao2025codegraphmodelcgm,
title={Code Graph Model (CGM): A Graph-Integrated Large Language Model for Repository-Level Software Engineering Tasks},
author={Hongyuan Tao and Ying Zhang and Zhenhao Tang and Hongen Peng and Xukun Zhu and Bingchang Liu and Yingguang Yang and Ziyin Zhang and Zhaogui Xu and Haipeng Zhang and Linchao Zhu and Rui Wang and Hang Yu and Jianguo Li and… See the full description on the dataset page: https://huggingface.co/datasets/codefuse-ai/CodeGraph.registryROOTER-Codegraphpaper-code-graph
codeofpaper: paper ↔ code graph
A high-precision dataset linking research papers (arXiv, OpenReview) to their
official and high-confidence community GitHub implementations, plus
the full fork lineage for those repositories. Built and maintained by the
codeofpaper project.
Snapshot date: 2026-05-19
Schema version: 1.0.0
License: CC-BY-4.0 (metadata aggregation + derived tier labels).
Underlying paper texts and repository code remain under their own licenses;
this dataset only ships… See the full description on the dataset page: https://huggingface.co/datasets/codeofpaper/paper-code-graph.graph_code_v1Solidity_Code_Graph_CodearenaHere is a 100 repositories from Codearena parsed as graph of functions with different types of relationships: Parent Contracts, High-Level Calls, Internal Calls, Library Calls, Low-Level Calls.
Dataset can be used for tasks like code clone detection (similar bugs search), code reconstruction with context pruning. Dataset is ready to be used in GNN.
You can see usage of dataset as RAG at github
Structure
auditits_info.json <-main file for easy paths navigantion
/processed… See the full description on the dataset page: https://huggingface.co/datasets/nothingisenough/Solidity_Code_Graph_Codearena.code_graph_text2cypher_datasetpaper-code-graph