Kubeflow for Machine Learning From Lab to Production

Делаю:
24.08.2022

Setting Up Your Kubeflow Development Environment

$ cd ~
$ virtualenv kfvenv --python python3
$ source kfvenv/bin/activate

$ pip install https://storage.googleapis.com/ml-pipeline/release/latest/kfp.tar.gz --upgrade

$ mkdir -p ~/repos
$ git -C ~/repos clone --single-branch --branch google-cloud-pipeline-components-0.3.0 https://github.com/kubeflow/pipelines.git

// ???
$ git clone https://github.com/kubeflow/example-seldon

Training and Monitoring Progress

$ cd ~/tmp/Kubeflow-for-Machine-Learning-From-Lab-to-Production/ch02_seldon_examples/
$ dsl-compile --py TrainPipeline.py --output TrainPipeline.yaml

localhost:7777

Pipelines -> Upload -> TrainPipeline.yaml

Run

// Не знаю как победить
This step is in Failed state with this message: error: error when creating "/tmp/manifest.yaml": Post https://10.96.0.1:443/apis/machinelearning.seldon.io/v1alpha2/namespaces/kubeflow/seldondeployments: stream error: stream ID 123; INTERNAL_ERROR

Test Query

???

https://github.com/intro-to-ml-with-kubeflow/intro-to-ml-with-kubeflow-examples/blob/master/ch2/query-endpoint.py