https://github.com/yingding/kf-examples
this repository contains examples of kubeflow pipeline
https://github.com/yingding/kf-examples
Last synced: 11 months ago
JSON representation
this repository contains examples of kubeflow pipeline
- Host: GitHub
- URL: https://github.com/yingding/kf-examples
- Owner: yingding
- License: apache-2.0
- Created: 2022-09-26T09:14:08.000Z (almost 4 years ago)
- Default Branch: main
- Last Pushed: 2024-03-21T09:00:38.000Z (over 2 years ago)
- Last Synced: 2025-04-01T20:00:10.847Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 7.65 MB
- Stars: 3
- Watchers: 2
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# kf-examples
this repository contains examples of kubeflow pipeline for the talks at:
* [Kubeflow Workshop and hands-on tutorials on 23 Feb. 2024](./workshops/README.md)
* [GDG Cloud Munich October Gathering on 5 Oct. 2022](https://gdg.community.dev/events/details/google-gdg-cloud-munich-presents-october-gathering-1/)
# Notebooks
* exploring-iris-data-notebook: explores the iris dataset used later in the pipelines
* single-model-pipeline-example: demonstrates a pipeline trains a single ANN model with tensorflow and shows the performance metrics and confusion matrix visualization for the test dataset.
* multiple-models-pipeline-example: demonstrates a pipeline trains an ANN model with tensorflow, a SVM model with scikit-learn and shows the performance metrics and confusion matrix visualization for both trained models performing on the test dataset.
* patch-notebook-env-kfp-run: helps to setup the right environment while running a kfp pipeline with kfp python sdk from notebook the first time.
# Issue
Should you encounter a 500 server issue while runing the demo pipelines with kfp python sdk for the first time from the Kubeflow notebook server.
Please double check whether the `kfp-server-api 1.6.2` is outdated.
* For KF 1.4.1, `kfp-server-api 1.7.1` are necessary.
* For KF 1.5.1, `kfp-server-api 1.8.2` are necessary.
* For KF 1.6.1, KF 1.7.0 `kfp-server-api 1.8.5` are necessary.
The major version number to the kfp-server backend shall match, so that a pipeline can be started.