Ask Learn
Preview
Please sign in to use this experience.
Sign inThis browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
米国時間 2018/12/4 (日本時間 12/5 早朝) に開催された Microsoft Connect(); 2018 にて、これらの Cognitive 関連の発表がありました;
Azure Updates > Language Understanding Container Support: Preview Azure Updates > Custom Translation capability of Text Translator
早速プレビュー使用が可能になっている LUIS Container を利用する方法をご紹介します。
以下の Docker 環境で Cognitive Services Container を利用可能です:
今回は local PC (Windows) にインストールした Docker 環境を使用します。
F0(無料版) で大丈夫ですので、Azure Portal から LUIS サービスを作成します。
作成した LUIS サービスの Endpoint (アクセス URL) と API key を取得しておきます。これらは課金のためにContainer からアクセスされます。(分析データなどのデータは一切送られません。)
Language Understanding は 簡易自然言語処理サービス (モデリング用 と分析用のGUI およびAPI) ですので、LUIS App (NLP モデル)を作成して、それをローカルの Container にExport します。
Export してダウンロードした LUIS App は ZGIP のまま Docker からアクセスできるフォルダに保存しておきます。
Docker コマンドで LUIS container image を pull します。"latest" オプションで最新版を取得できます。
docker pull docker pull mcr.microsoft.com/azure-cognitive-services/luis:latest
LUIS container image をダウンロードしておいた LUIS app をマウントして起動します。
docker run --rm -it -p 5000:5000 --memory 4g --cpus 2 --mount type=bind,src=c:\demo\container\luis,target=/input --mount type=bind,src=C:\demo\container\luis,target=/output mcr.microsoft.com/azure-cognitive-services/luis Eula=accept Billing=https://westus.api.cognitive.microsoft.com/luis/v2.0 ApiKey=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
上記ではこのような設定になっています:
Docker image が起動したら、https://localhost:5000 をブラウザーに入力して動作を確認できます。
Swagger (https://localhost:5000/swagger) のページで API の仕様や、実際にGET/POST を行って自然言語の分析結果を取得できます。
Microsoft Azure Blog > Getting started with Cognitive Services Language Understanding container Github > Azure Samples - Container support for Azure Cognitive Services
Please sign in to use this experience.
Sign in