Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.

Visual Studio Code Extension based on Xtext for Deep Learning DSL

This is a Deep Learning Script integration for Visual Studio Code based on Xtext grammar using the Language Server Protocol for communication.

Build Status

Join the chat at



Requires Visual Studio Code (VS Code) with version 1.4.0 or greater to be on the path as code and Java 8+ available as java.

  • Run ./gradlew startCode

This will start VS Code and after a few seconds load the demo folder of this repository.

Building in Details

  1. Make sure that java -version is executable and pointing to a Java 8+ JDK.
  2. Type code. If the command is not known, open VS Code and select View / Command Palette. Enter code and select to install code on the path.
  3. Run ./gradlew startCode to build the DSL and the VS Code extensions.

Scenario 1 (embedded server)

  1. Install the self-contained extension into VS Code using code --install-extension vscode-extension-self-contained/build/vscode/vscode-extension-self-contained-0.0.1.vsix
  2. Run a second instance of vscode on the demo folder code demo

Scenario 2 (client-only with separate server process)

  1. Run ./gradlew run or launch RunServer from Eclipse.
  2. Open vscode-extension in VS Code and F5 to launch new editor (you may need a Debug -> Start Debugging initally).
  3. Open folder demo in the new editor.

Build VS Code Extension Package manually (manually Gradle)

npm install -g vsce
cd vscode-extension
vsce package
cd ../vscode-extension-self-contained
vsce package