add document intelligence example

Author: ks6088ts

.env.template

@@ -17,3 +17,8 @@ AZURE_COSMOS_DB_CONTAINER_NAME="chat"
 AZURE_AI_SEARCH_ENDPOINT="https://<YOUR_AZURE_SEARCH_NAME>.search.windows.net/"
 AZURE_AI_SEARCH_API_KEY="<YOUR_API_KEY>"
 AZURE_AI_SEARCH_INDEX_NAME="chat"
+
+# Azure Document Intelligence: https://github.com/Azure/azure-rest-api-specs/tree/main/specification/ai/data-plane/DocumentIntelligence
+AZURE_DOCUMENT_INTELLIGENCE_ENDPOINT="https://<YOUR_AZURE_DOCUMENT_INTELLIGENCE_NAME>.cognitiveservices.azure.com/"
+AZURE_DOCUMENT_INTELLIGENCE_API_VERSION="2024-07-31-preview"
+AZURE_DOCUMENT_INTELLIGENCE_API_KEY="<YOUR_API_KEY>"

apps/9_streamlit_azure_document_intelligence/README.md

@@ -0,0 +1,37 @@
+# Call Azure Document Intelligence API with Streamlit
+
+This Streamlit app demonstrates how to call Azure Document Intelligence API with Streamlit.
+
+## Prerequisites
+
+- Python 3.10 or later
+- Azure Document Intelligence API key
+
+## Usage
+
+1. Get Azure Document Intelligence API key
+1. Copy [.env.template](../../.env.template) to `.env` in the same directory
+1. Set credentials in `.env`
+1. Run [main.py](./main.py)
+
+```shell
+# Create a virtual environment
+$ python -m venv .venv
+
+# Activate the virtual environment
+$ source .venv/bin/activate
+
+# Install dependencies
+$ pip install -r requirements.txt
+
+# Run the script
+$ python -m streamlit run apps/9_streamlit_azure_document_intelligence/main.py
+```
+
+### Example
+
+![Streamlit Chat](../../docs/images/9_streamlit_azure_document_intelligence.main.png)
+
+## References
+
+- [Azure AI Document Intelligence client library for Python - version 1.0.0b3](https://learn.microsoft.com/en-us/python/api/overview/azure/ai-documentintelligence-readme?view=azure-python-preview)

apps/9_streamlit_azure_document_intelligence/main.py

@@ -0,0 +1,108 @@
+import base64
+import json
+from os import getenv
+
+import streamlit as st
+from azure.ai.documentintelligence import DocumentIntelligenceClient
+from azure.ai.documentintelligence.models import AnalyzeResult, ContentFormat
+from azure.core.credentials import AzureKeyCredential
+from dotenv import load_dotenv
+
+load_dotenv()
+
+with st.sidebar:
+    azure_document_intelligence_endpoint = st.text_input(
+        label="AZURE_DOCUMENT_INTELLIGENCE_ENDPOINT",
+        value=getenv("AZURE_DOCUMENT_INTELLIGENCE_ENDPOINT"),
+        key="AZURE_DOCUMENT_INTELLIGENCE_ENDPOINT",
+        type="default",
+    )
+    azure_document_intelligence_api_version = st.text_input(
+        label="AZURE_DOCUMENT_INTELLIGENCE_API_VERSION",
+        value=getenv("AZURE_DOCUMENT_INTELLIGENCE_API_VERSION"),
+        key="AZURE_DOCUMENT_INTELLIGENCE_API_VERSION",
+        type="default",
+    )
+    azure_document_intelligence_api_key = st.text_input(
+        label="AZURE_DOCUMENT_INTELLIGENCE_API_KEY",
+        key="AZURE_DOCUMENT_INTELLIGENCE_API_KEY",
+        type="password",
+    )
+    "[Azure Portal](https://portal.azure.com/)"
+    "[Azure OpenAI Studio](https://oai.azure.com/resource/overview)"
+    "[View the source code](https://github.com/ks6088ts-labs/workshop-azure-openai/blob/main/apps/9_streamlit_azure_document_intelligence/main.py)"
+
+
+def is_configured():
+    return (
+        azure_document_intelligence_endpoint
+        and azure_document_intelligence_api_version
+        and azure_document_intelligence_api_key
+    )
+
+
+def get_client():
+    return DocumentIntelligenceClient(
+        credential=AzureKeyCredential(azure_document_intelligence_api_key),
+        endpoint=azure_document_intelligence_endpoint,
+        api_version=azure_document_intelligence_api_version,
+    )
+
+
+st.title("9_streamlit_azure_document_intelligence")
+
+if not is_configured():
+    st.warning("Please fill in the required fields at the sidebar.")
+
+# ---------------
+# Upload file
+# ---------------
+st.header("Upload file")
+st.info("Upload a file")
+uploaded_file = st.file_uploader(
+    "Upload an input file",
+    type=(
+        "txt",
+        "text",
+        "pdf",
+        "pptx",
+        "jpg",
+        "jpeg",
+        "png",
+    ),
+)
+if uploaded_file:
+    bytes_data = uploaded_file.read()
+    st.write(f"File name: {uploaded_file.name}")
+    st.write(f"File type: {uploaded_file.type}")
+    st.write(f"File size: {len(bytes_data)} bytes")
+
+if st.button(
+    "Analyze",
+    key="Analyze",
+    disabled=not is_configured(),
+):
+    with st.spinner("Analyzing..."):
+        try:
+            poller = get_client().begin_analyze_document(
+                model_id="prebuilt-layout",
+                analyze_request=bytes_data,
+                content_type="application/octet-stream",
+                output_content_format=ContentFormat.TEXT,
+            )
+            result: AnalyzeResult = poller.result()
+            output_encoded = base64.b64encode(
+                json.dumps(
+                    result.as_dict(),
+                    indent=2,
+                ).encode()
+            ).decode()
+            # Generate a link to download the result
+            st.markdown(
+                f'<a href="data:file/txt;base64,{output_encoded}" \
+                    download="{uploaded_file.name}.json">Download Result</a>',
+                unsafe_allow_html=True,
+            )
+            st.write(result.as_dict())
+        except Exception as e:
+            st.error(e)

docs/images/9_streamlit_azure_document_intelligence.main.png

@@ -22,6 +22,7 @@ azure-search-documents = "^11.5.1"
 azure-identity = "^1.17.1"
 ultralytics = "^8.2.77"
 mediapipe = "^0.10.14"
+azure-ai-documentintelligence = "^1.0.0b3"
 
 [tool.poetry.group.dev.dependencies]
 pre-commit = "^3.8.0"

poetry.lock

@@ -15,3 +15,5 @@ azure-identity==1.17.1
 
 # To run 99_streamlit_examples/pages/11_Pose_Estimation.py
 # mediapipe==0.10.14
+
+azure-ai-documentintelligence==1.0.0b3