[AI] 呼叫布署为容器的Azure Document Intelligence

布署版本只能为v3.1

下载位置为: https://mcr.microsoft.com/en-us/

需要下载的image如下:
docker.io/library/nginx                                                                               
mcr.microsoft.com/azure-cognitive-services/form-recognizer/layout-3.1
*layout也有4.0版,但我还没试
mcr.microsoft.com/azure-cognitive-services/form-recognizer/read-3.1
mcr.microsoft.com/azure-cognitive-services/form-recognizer/custom-template-3.1
mcr.microsoft.com/azure-cognitive-services/form-recognizer/studio

**地端的api和studio要各一个dns

当你架好api时, 进入该网站会显示如下:

点上图红框就有教你各功能的呼叫法, 常用就是解析功能:
URL: https://自订DNS/formrecognizer/documentModels/{modelid, ex. prebuilt-read}:syncAnalyze?api-version=2023-07-31
Header: Content-Type=application/octet-stream
Body: 选binary, 直接上传档案
或传url
Header: Content-Type=application/json
Body: 选json, 输入 {"urlSource": "http://example.com/images/test.jpg"}

Azure Document Intelligence似乎没有api能直接呼叫, 範例都要使用其套件

Taiwan is a country. 台湾是我的国家