Kedro Vertex AI Plugin
0.11.0
Contents:
Introduction
What is GCP VertexAI Pipelines?
Why to integrate Kedro project with Vertex AI Pipelines?
Installation
Plugin installation
Kedro setup
Plugin installation
Install from PyPI
Install from sources
Available commands
init
ui
list-pipelines
compile
run-once
Configuration parameters
Dynamic configuration support
Using
OmegaConfigLoader
Dynamic config providers
Grouping feature
Example
Resources configuration
Quick start
Quickstart
Preprequisites
Install the toy project with Vertex AI Pipelines support
Build the docker image to be used in Vertex AI Pipelines runs
Ensure right requirements.txt
Adjusting Data Catalog to be compatible with Vertex AI
Disable telemetry or ensure consent
Build the image
Run the pipeline on Vertex AI
Google Cloud Platform support
Using
kedro
with AI Platform Notebooks
Using
kedro-kubeflow
with AI Platform Pipelines
1. Connecting to AI Pipelines from AI Platform Notebooks
2. Authentication to AI Pipelines from local environment
3. Authenticating through IAP Proxy
Using
kedro-kubeflow
with Vertex AI Pipelines (EXPERIMENTAL)
1. Preparing configuration
2. Preparing environment variables
3. Supported commands
Kedro-Mlflow integration
Authorization
Authorization with a service account email and OAuth Client ID (IAM)
OAuth2.0 based authorization
Custom authorization method
Continuous Deployment
Github Actions
Authenticating to Kubeflow Pipelines API
1. KFP behind IAP proxy on Google Cloud
2. KFP behind Dex+authservice
Kedro Vertex AI Plugin
Index
Edit on GitHub
Index
Read the Docs
v: 0.11.0
Versions
latest
stable
0.11.0
0.10.0
0.9.1
0.9.0
0.8.1
0.8.0
0.7.0
0.6.0
0.5.0
0.4.1
0.4.0
0.3.0
0.2.0
0.1.0
Downloads
On Read the Docs
Project Home
Builds