apache / incubator-hugegraph-ai

The integration of HugeGraph with artificial intelligence

hugegraph-ai

hugegraph-ai aims to explore the integration of HugeGraph with artificial intelligence (AI) and provide comprehensive support for developers to leverage HugeGraph's AI capabilities in their projects.

Modules

• hugegraph-llm:The hugegraph-llm will house the implementation and research related to large language models. It will include runnable demos and can also be used as a third-party library, reducing the cost of using graph systems and the complexity of building knowledge graphs. Graph systems can help large models address challenges like timeliness and hallucination, while large models can help graph systems with cost-related issues. Therefore, this module will explore more applications and integration solutions for graph systems and large language models.
• hugegraph-ml: The hugegraph-ml will focus on integrating HugeGraph with graph machine learning, graph neural networks, and graph embeddings libraries. It will build an efficient and versatile intermediate layer to seamlessly connect with third-party graph-related ML frameworks.
• hugegraph-python-client: The hugegraph-python-client is a Python client for HugeGraph. It is used to define graph structures and perform CRUD operations on graph data. Both the hugegraph-llm and hugegraph-ml modules will depend on this foundational library.

Contributing

• Welcome to contribute to HugeGraph, please see Guidelines for more information.
• Note: It's recommended to use GitHub Desktop to greatly simplify the PR and commit process.
• Code format: Please run ./style/code_format_and_analysis.sh to format your code before submitting a PR.
• Thank you to all the people who already contributed to HugeGraph!

License

hugegraph-ai is licensed under Apache 2.0 License.

Contact Us

• GitHub Issues: Feedback on usage issues and functional requirements (quick response)
• Feedback Email: [email protected] (subscriber only)
• WeChat public account: Apache HugeGraph, welcome to scan this QR code to follow us.

incubator-hugegraph-ai's Issues

[Feature] Extract schema from HugeGraph for kg construction

[Feature] rag based on HugeGraph

[Feature] knowledge graph construction by llm

Search before asking

• I had searched in the feature and found no similar feature requirement.

Feature Description (功能描述)

Knowledge graph construction by llm：Extracting Data from Unstructured Text Using LLM to Construct a Knowledge Graph

Are you willing to submit a PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct

[Question] A question about introducing black for code format

Problem Type (问题类型)

configs (配置项 / 文档相关)

Before submit

• I had searched in the issues and found no similar issues.

Environment (环境信息)

• Server Version: 1.x.x
• Toolchain Version: 1.x.x
• Data Size: xx vertices, xx edges >

Your Question (问题描述)

I am thinking about introducing BLACK for code check but still confused about the process.

I also noticed that there's a style script already using BLACK, refer: https://github.com/apache/incubator-hugegraph-ai/blob/main/style/code_format_and_analysis.sh

This script is used in the pylint.yml workflow, but only the pylint functionality is enabled. Should I create a new workflow for using BLACK, following the structure of pylint.yml, or should I modify pylint.yml to make it capable of performing code formatting tasks as well?

Vertex/Edge example (问题点 / 边数据举例)

No response

Schema [VertexLabel, EdgeLabel, IndexLabel] (元数据结构)

No response

[Feature] introduce pylint for code analysis

Search before asking

• I had searched in the feature and found no similar feature requirement.

Feature Description (功能描述)

introduce pylint for code analysis

Are you willing to submit a PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct

[Summary] HugeGraph-ai roadmap & Contribution Wanted

Feature Description (功能描述)

If you are interested in the following tasks, feel free to claim them. Just comment below with the feature you would like to implement and create the corresponding issue.

• hugegraph-llm
  • [Feature] Introduce FastAPI as the web framework of hugegraph-ai
  • [Feature] Support custom prompts by developer
  1. kg construction
  • #6 @lzyxx77
  • #18 @lzyxx77
  • #29
  • [Feature] add document for kg construction
  2. convert nl to gremlin
  • #10
  3. retrieval augmented generation（RAG）
  • #19 @Ling-Yuchen
  • [Feature] Explore more combinations to improve accuracy
• hugegraph-ml
  • [Feature] integration of HugeGraph with pyG
  • [Feature] integration of HugeGraph with DGL
  • [Feature] integration of HugeGraph with PGL (consider reuse)
• hugegraph-python （refer to hugegraph-restful-api-doc ）
  • #4
  • #28
  • #26
  • #14
• others
  • #21
  • [Feature] introduce black for code format
  • [Feature] improve hugegraph-ai ci
  • [Feature] improve unit test coverage
  • [Feature] improve documents
  • #16
• bugfix
  • #12
  • #13

Welcome everyone to provide relevant suggestions and participate in this plan~ 💯

☎️ Add WeChat shirning to join the contributor group
☎️ 欢迎加微信 shirning 加入开发群，备注 ai

Reference:

• apache/incubator-hugegraph#2251
• apache/incubator-hugegraph#2318

[Feature] Convert natural language to gremlin

[Feature] implement variables api, add unit test and document

[Feature] Support for customizing base_url when using openai

Search before asking

• I had searched in the feature and found no similar feature requirement.

Feature Description (功能描述)

Support for customizing base_url as a proxy when using openai-api

Are you willing to submit a PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct

[Bug] getEdgeByPage(): When the parameter page is not passed，the program reports an error

Bug Type (问题类型)

exception / error (异常报错)

Before submit

• I had searched in the issues and found no similar issues.

Environment (环境信息)

• Server Version: v1.x.x
• Toolchain Version: v1.x.x
• Data Size: xx vertices, xx edges

Expected & Actual behavior (期望与实际表现)

client = PyHugeClient("localhost", "8080", user="admin", pwd="admin", graph="test")
graph = client.graph()
res = graph.getEdgeByPage("knows")

the source code:

def getEdgeByPage(self, label=None, vertex_id=None, direction=None, limit=0, page=None, properties=None):
        url = f'{self._host}/graphs/{self._graph_name}/graph/edges?'

        para = ""
        if vertex_id:
            if direction:
                para = para + "&vertex_id=\"" + vertex_id + "\"&direction=" + direction
            else:
                raise NotFoundError("Direction can not be empty.")
        if label:
            para = para + "&label=" + label
        if properties:
            para = para + "&properties=" + json.dumps(properties)
        if page:
                para += '&page={}'.format(page)
        else:
            para += '&page'
        if limit > 0:
            para = para + "&limit=" + str(limit)
        url = url + para[1:]
        response = self.session.get(url, auth=self._auth, headers=self._headers, timeout=self._timeout)
        if check_if_success(response, NotFoundError("not found edges: {}".format(response.content))):
            res = []
            for item in json.loads(response.content)["edges"]:
                res.append(EdgeData(item))
            return res, json.loads(response.content)["page"]

the exception occur in the last line "return res, json.loads(response.content)["page"]"

Vertex/Edge example (问题点 / 边数据举例)

No response

Schema [VertexLabel, EdgeLabel, IndexLabel] (元数据结构)

No response

[Question] Build local development environment

Problem Type (问题类型)

others (please comment below)

Before submit

• I had searched in the issues and found no similar issues.

Environment (环境信息)

python 3.10
Pycharm
mac

Your Question (问题描述)

Use pycharm to run test.py and report an error

Traceback (most recent call last):
  File "/Users/yangjiaqi/Documents/hugegraph_concurrency/repo/incubator-hugegraph-ai/hugegraph-python/example/test.py", line 18, in <module>
    from src.client import PyHugeClient
ModuleNotFoundError: No module named 'src'

Process finished with exit code 1

Root cause:

Solution:

[Bug] hugegraph-llm/client README outdated

Bug Type (问题类型)

None

Before submit

• I had searched in the issues and found no similar issues.

Environment (环境信息)

None

Expected & Actual behavior (期望与实际表现)

As the project develops, changes to the README don't seem to be keeping up, please check and change the README.

eg:

incubator-hugegraph-ai/hugegraph-llm/README.md

Line 25 in 243f547

> Note: If you need a proxy to access OpenAI's API, please set your HTTP proxy in `build_kg_test.py`.

we can't set HTTP proxy in build_kg_test.py now.

Vertex/Edge example (问题点 / 边数据举例)

No response

Schema [VertexLabel, EdgeLabel, IndexLabel] (元数据结构)

No response

[Feature] Introduce labeler action

Search before asking

• I had searched in the feature and found no similar feature requirement.

Feature Description (功能描述)

Automatically mark which module the modified code belongs to help review the code
refer: https://github.com/actions/labeler and apache/incubator-hugegraph-toolchain#466

Are you willing to submit a PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct

[Feature] initialize hugegraph python client

Search before asking

• I had searched in the feature and found no similar feature requirement.

Feature Description (功能描述)

initialize hugegraph python client, have implement graph, graphs, schema, gremlin api for hugegraph.

Are you willing to submit a PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct

[Feature] implement version api, add unit test and document

Document refer to : https://hugegraph.apache.org/cn/docs/clients/restful-api/other/

[Feature] improve project structure

Search before asking

• I had searched in the feature and found no similar feature requirement.

Feature Description (功能描述)

improve project struct

Are you willing to submit a PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct