Skip to main content
Open In ColabOpen on GitHub

Azure AI 文档智能

Azure AI Document Intelligence (formerly known as Azure Form Recognizer) 是一种基于机器学习的服务，可以从数字化或扫描的PDF、图像、Office和HTML文件中提取文本（包括手写体）、表格、文档结构（例如标题、段落小节等）以及键值对。

Document Intelligence 支持 PDF, JPEG/JPG, PNG, BMP, TIFF, HEIF, DOCX, XLSX, PPTX 和 HTML.

这当前使用Document Intelligence实现的加载器可以按页面引入内容并将其转换为LangChain文档。默认输出格式是markdown，可以轻松地与MarkdownHeaderTextSplitter结合用于语义分块。你也可以使用mode="single"或mode="page"来返回纯文本单页或者将文档按页面分割。

前置条件​

一个位于以下三个预览区域之一的 Azure AI 文档智能资源：East US、West US2、West Europe - 如果您尚未拥有，请遵循此文档进行创建。您将把 <endpoint> 和 <key> 作为参数传递给加载器。

%pip install --upgrade --quiet  langchain langchain-community azure-ai-documentintelligence

Example 1​

The first example uses a local file which will be sent to Azure AI Document Intelligence.

使用初始化的文档分析客户端，我们可以创建一个DocumentIntelligenceLoader 实例：<br/>

from langchain_community.document_loaders import AzureAIDocumentIntelligenceLoader

file_path = "<filepath>"
endpoint = "<endpoint>"
key = "<key>"
loader = AzureAIDocumentIntelligenceLoader(
    api_endpoint=endpoint, api_key=key, file_path=file_path, api_model="prebuilt-layout"
)

documents = loader.load()

The default output contains one LangChain 文档，内容格式为 markdown 格式:

documents

Example 2​

The input file can also be a public URL path. E.g., https://raw.githubusercontent.com/Azure-Samples/cognitive-services-REST-api-samples/master/curl/form-recognizer/rest-api/layout.png.

url_path = "<url>"
loader = AzureAIDocumentIntelligenceLoader(
    api_endpoint=endpoint, api_key=key, url_path=url_path, api_model="prebuilt-layout"
)

documents = loader.load()
documents

例 3​

您可以指定mode="page"以按页加载文档。

from langchain_community.document_loaders import AzureAIDocumentIntelligenceLoader

file_path = "<filepath>"
endpoint = "<endpoint>"
key = "<key>"
loader = AzureAIDocumentIntelligenceLoader(
    api_endpoint=endpoint,
    api_key=key,
    file_path=file_path,
    api_model="prebuilt-layout",
    mode="page",
)

documents = loader.load()

The output will be each page stored as a separate document in the list:

for document in documents:
    print(f"Page Content: {document.page_content}")
    print(f"Metadata: {document.metadata}")

示例 4​

您可以指定analysis_feature=["ocrHighResolution"]以启用附加功能。如需更多详细信息，请参阅：https://aka.ms/azsdk/python/documentintelligence/analysisfeature。

from langchain_community.document_loaders import AzureAIDocumentIntelligenceLoader

file_path = "<filepath>"
endpoint = "<endpoint>"
key = "<key>"
analysis_features = ["ocrHighResolution"]
loader = AzureAIDocumentIntelligenceLoader(
    api_endpoint=endpoint,
    api_key=key,
    file_path=file_path,
    api_model="prebuilt-layout",
    analysis_features=analysis_features,
)

documents = loader.load()

输出包含通过高分辨率附加功能识别的 LangChain 文档：

documents