Note
Form Recognizer is now Azure AI Document Intelligence!
- Code samples for each language's SDK are in the links below. The first step is to choose one (default Python).
| Python | .NET | Java | JavaScript |
|---|
- The contents of this floder default the latest version: v4.0 (2024-02-29-preview) .
- You can select v3.1 (2023-07-31-GA) to view earlier versions.
Azure AI Document Intelligence is a cloud-based Azure AI service that enables you to build intelligent document processing solutions. Massive amounts of data, spanning a wide variety of data types, are stored in forms and documents. Document Intelligence enables you to effectively manage the velocity at which data is collected and processed and is key to improved operations, informed data-driven decisions, and enlightened innovation.
- Azure subscription - Create one for free.
- Python 3.8 or later. Your Python installation should include pip. You can check if you have pip installed by running
pip --versionon the command line. Get pip by installing the latest version of Python. - Install the latest version of Visual Studio Code or your preferred IDE. For more information, see Getting Started with Python in Visual Studio Code.
- An Azure AI services or Document Intelligence resource. Once you have your Azure subscription, Create a single-service or multi-service resource. You can use the free pricing tier (F0) to try the service and upgrade to a paid tier for production later.
- Get endpoint and keys to your Document Intelligence resource.
- Open a terminal window in your local environment and install the Azure AI Document Intelligence client library for Python with pip:
pip install azure-ai-documentintelligence --pre- Clone or download this sample repository
- Open the sample folder in Visual Studio Code or your IDE of choice.
- Open a terminal window and
cdto the directory that the samples are saved in. - Set the environment variables specified in the sample file you wish to run.
- If necessary, select Data to get your document.
- Below are some sample code guidelines so that you can choose the sample according to your needs.
Note: For more samples, see Azure SDK for Python and Async Samples.
Select the link of the model name to reach the corresponding topic page for more details. Select v3.1 (2023-07-31-GA) to view earlier versions.
Read model : Extract printed and handwritten text.
Layout mode : Extract and anlayze text, tables, and document structure.
Prebuilt model : Add intelligent document processing to your apps and flows without having to train and build your own models.
sample_analyze_invoices.py - Analyze document text, selection marks, tables, and pre-trained fields and values pertaining to English invoices using a prebuilt model.
sample_analyze_identity_documents.py - Analyze document text and pre-trained fields and values pertaining to US driver licenses and international passports using a prebuilt model.
sample_analyze_receipts.py - Analyze document text and pre-trained fields and values pertaining to English sales receipts using a prebuilt model.
sample_analyze_tax_us_w2.py - Analyze document text and pre-trained fields and values pertaining to US tax W-2 forms using a prebuilt model.
Add-on capabilities : Extend the extracted results from documents with add-on capabilities.
sample_analyze_addon_barcodes.py - Extract barcode from a document using this add-on capability.
sample_analyze_addon_fonts.py - Extract font property from a document using this add-on capability.
sample_analyze_addon_formulas.py - Extract formula from a document using this add-on capability.
sample_analyze_addon_highres.py - Extract high resolution from a document using this add-on capability.
sample_analyze_addon_languages.py - Detact language from a document using this add-on capability.
sample_analyze_addon_query_fields.py - Query fields from a document using this add-on capability.
Custom model : Train your own models to extract data from structured and unstructured documents.
sample_analyze_custom_documents.py - Analyze a document with a custom built model. The document must be of the same type as the documents the custom model was built on.
sample_classify_document.py - Classify a document using a trained document classifier.
sample_compose_model.py - This is useful when you have built different models and want to aggregate a group of them into a single model that you (or a user) could use to analyze a document.
sample_copy_model_to.py - Copy a custom model from a source Form Recognizer resource to a target Form Recognizer resource.
sample_copy_model_to.py - Copy a custom model from a source Document Intelligence resource to a target Document Intelligence resource.
sample_manage_classifiers.py - Manage the classifiers on your account.
sample_manage_models.py - Manage the models on your account.
sample_convert_to_dict.py - Convert a model returned from an analyze operation to a JSON serializable dictionary.
The Layout model provides various building blocks like tables, paragraphs, section headings, etc. that can enable different semantic chunking strategies of the document. With semantic chunking in Retrieval Augmented Generation (RAG), it will be more efficient in storage and retrieval, together with the benefits of improved relevance and enhanced interpretability. The following samples show how to use the Layout model to do semantic chunking and use the chunks to do RAG.
Note：Only available for v4.0 (2024-02-29-preview) .
sample_rag_langchain.ipynb
Sample RAG notebook using Azure AI Document Intelligence as document loader, MarkdownHeaderSplitter and Azure AI Search as retriever in Langchain.
sample_figure_understanding.ipynb
Sample notebook showcasing how to crop the figures and send figure content (with its caption) to Azure Open AI GPT-4V model to understand the semantics. The figure description will be used to update the markdown output, which can be further used for semantic chunking.
sample_identify_and_merge_cross_page_tables.ipynb and sample_identify_and_merge_cross_page_tables.py
Sample postprocessing script to identify and merge cross-page tables based on business rules.
There are usually some pre/post processing steps that are needed to get the best results from the Document Intelligence models. These steps are not part of the Document Intelligence service, but are common steps that are needed to get the best results. The following samples show how to do these steps.
Note：Applies to all versions.
sample_disambiguate_similar_characters.ipynb and sample_disambiguate_similar_characters.py
Sample postprocessing script to disambiguate similar characters based on business rules.
sample_identify_cross_page_tables.ipynb and sample_identify_cross_page_tables.py
Sample postprocessing script to identify cross-page tables based on business rules.
Check out the API reference documentation to learn more about what you can do with the Azure Document Intelligence client library.