Cloud Pipelines Editor is a web app that allows the users to build and run Machine Learning pipelines using drag and drop without having to set up development environment.
The Cloud Pipelines Editor VSCode extension tries to bring that functionality to VSCode. Install the extension.
Open the File/New File... menu and choose Pipeline to create a new pipeline.
Or click on any *.pipeline.component.yaml file to open it in the Pipeline Editor.
Please take a look at the short video demonstrating the visual pipeline editor.
Cloud Pipelines - Build machine learning pipelines without writing code
The early alpha version of the Cloud Pipelines Editor app shown in this video is now available at https://cloud-pipelines.net/pipeline-editor . The app is open and standalone. No registration is required.
Please check it out and report any bugs you find using GitHub Issues.
The app is under active development, so expect some breakages as I work on the app and do not rely on the app for production.
App features:
- Build pipeline using drag and drop
- Edit component arguments (double-click in the task nodes)
- Fully compatible with the Kubeflow Pipelines' components (
component.yamlfiles) You can find some components here: Ark-kun/pipeline_components or kubeflow/pipelines/components - Preloaded component library with 70+ ML-related components
- [Only works in the web app right now] Submit the pipeline to Google Cloud Vertex Pipelines for execution.
- [Only works in the web app right now] Submit the pipeline to Kubeflow Pipelines for execution (local on-prem cluster or cloud).
- [Only works in the web app right now] User component library (add private components)
- [Only works in the web app right now] Component search
- [Only works in the web app right now] Import and export pipelines
There are many features that I want to add, but I want to prioritize them based on your feedback.