GitHub - wmeddie/deeplearning-servicefabric: Example of running a deep neural network model on Service Fabric

wmeddie / deeplearning-servicefabric Public

Notifications You must be signed in to change notification settings
Fork 0
Star 3

Example of running a deep neural network model on Service Fabric

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
project		project
service		service
training/iris-model		training/iris-model
.gitignore		.gitignore
LICENSE		LICENSE
README		README
activator		activator
activator-launch-1.3.5.jar		activator-launch-1.3.5.jar
build.sbt		build.sbt

Repository files navigation

DeepLearning4J on Service Fabric
================================

日本語
-----
このリポジトリでDeepLearning4JのニューラルネットワークをService Fabric ClusterのStatelessService
に展開する方法の一つを紹介します。

利用手順

- モデルを作成する
    - bashのターミナルを開き
    - training/iris-modelにcdし
    - ./package.shを実行
    - できたnugetパッケージをローカルのリポジトリにプッシュnuget push target/nuget/Example.IrisModel.1.0.0.nupkg -Source local
- Service Fabric Applicationのパッケージを作成
    - bashのターミナルを開き
    - service/IrisApp/Scriptsにcd
    - ./build.shを実行
    - service/IrisApp/ApplicationPackageRootがデプロイ出来る状態です。

deploy-local.ps1を利用してローカルのService Fabricにデプロイできます。

English
-------

This repository shows how to run a DeepLearning4J neural network inside a
Service Fabric Cluster StatelessService.

Steps:

- Train a model
    - Open a bash terminal (Git Bash on Windows)
    - cd to training/iris-model
    - run ./package.sh
    - publish the resulting nuget package with nuget push target/nuget/Example.IrisModel.1.0.0.nupkg -Source local
- Build a Service Fabric Application package
    - Open a bash terminal
    - cd to service/IrisApp/Scripts/
    - run ./build.sh
    - ApplicationPackageRoot is now a deployable Service Fabric Application package

You can use deploy-local.ps1 to deploy it to your local Service Fabric instance.