开源自荐 cocoindex - 为AI实时索引数据

项目地址：

https://github.com/cocoindex-io/cocoindex

项目描述

CocoIndex是开源ETL框架， 为AI实时索引数据。 CocoIndex是全世界第一款支持自定义逻辑，并且自带增量更新（incremental update）的数据框架。CocoIndex 可以有效地帮你给AI准备数据（RAG，Semantic Search）。以最简单的形式，像乐高一样搭建你的ETL pipeline，并且提供增量更新（incremental update）。

CocoIndex框架+引擎，里面可以套任何的自定义模块，各种PDF parsing，chunking，embedding都可以套进去用。

🔥 核心feature:

• 像乐高一样搭建你的RAG Pipeline。
• 增量更新，当你源数据改变后CocoIndex引擎会减少计算和数据更新，只更新需要的差量delta。
• 数据流编程（Data flow programming），以最简单的形式定义数据流。
• 高效稳定，核心是用Rust🦀写的。给各位爱好Python🐍的小伙伴们提供了Python SDK。

Cocoindex 文档齐全，新手包友好。模块化的搭建你的RAG Pipeline，五分钟上手🚀。

示例代码

@cocoindex.flow_def(name="TextEmbedding")
def text_embedding_flow(flow_builder: cocoindex.FlowBuilder, data_scope: cocoindex.DataScope):
    # Add a data source to read files from a directory
    data_scope["documents"] = flow_builder.add_source(cocoindex.sources.LocalFile(path="markdown_files"))

    # Add a collector for data to be exported to the vector index
    doc_embeddings = data_scope.add_collector()

    # Transform data of each document
    with data_scope["documents"].row() as doc:
        # Split the document into chunks, put into `chunks` field
        doc["chunks"] = doc["content"].transform(
            cocoindex.functions.SplitRecursively(
                language="markdown", chunk_size=300, chunk_overlap=100))

        # Transform data of each chunk
        with doc["chunks"].row() as chunk:
            # Embed the chunk, put into `embedding` field
            chunk["embedding"] = chunk["text"].transform(
                cocoindex.functions.SentenceTransformerEmbed(
                    model="sentence-transformers/all-MiniLM-L6-v2"))

            # Collect the chunk into the collector.
            doc_embeddings.collect(filename=doc["filename"], location=chunk["location"],
                                   text=chunk["text"], embedding=chunk["embedding"])

    # Export collected data to a vector index.
    doc_embeddings.export(
        "doc_embeddings",
        cocoindex.storages.Postgres(),
        primary_key_fields=["filename", "location"],
        vector_index=[("embedding", cocoindex.VectorSimilarityMetric.COSINE_SIMILARITY)])

截图或演示视频

No repository file, test, or entry point is mentioned. This is a project recommendation rather than a defined code or documentation change, and the issue does not specify the editorial action or completion criteria.