You should follow the git workflow to contribute, fork the repo and pull request
- This project use JDK17 & Maven
- Springboot 3.0+
- lombok plugin
- For macOS, you need install
coreutils
&wget
viabrew install
- For ZenML spec, please refer to ZenML Introduction
$ mvn clean install
$ ./build/scripts/package.sh
package.sh
generates tar.gz
files as below:
- General package without embedded openjdk17
Kyligence-ZenML-Toolkit-${project.version}.tar.gz
- OS specific package with openjdk17
Kyligence-ZenML-Toolkit-Darwin-AArch64-${project.version}.tar.gz
Kyligence-ZenML-Toolkit-Darwin-x64-${project.version}.tar.gz
Kyligence-ZenML-Toolkit-Linux-x64-${project.version}.tar.gz
The openjdk17 is downloaded during package process, you can refer to build/scripts/download-jdk.sh
for more details.
$ ./build/scripts/build_docker_image.sh
Start docker container
docker run -p 9000:9000 -dit kyligence/zenml-toolkit:${project.version}
We provide an CLI interface for user via command ./bin/zen.sh -i <arg> -o <arg>
.
The command options is defined in io.kyligence.zenml.toolkit.ZenMlToolkitCLI
We also provide a server mode and API for user to upload a file, the metrics metadata will be extracted to a ZenML file and an Excel file, compressed as a zip file to download.
The entry class is io.kyligence.zenml.toolkit.ZenMlToolkitServer
Debug io.kyligence.zenml.toolkit.ZenMlToolkitServer#main()
, in debug configuration, you need to define an Environment
variable named ZEN_HOME
, the value is a directory path, put build/conf/toolkit.properties
to $ZEN_HOME/conf/toolkit.properties
See Frontend README
We use the Apache License 2.0, you should put the license content in the beginning of source code files