This example requires a running Kubeflow environment. The easiest way to setup a Kubeflow environment is by using the Deployment UI.
We have to create a bucket, this bucket will contain everything which is required for our Kubeflow pipeline.
gsutil mb -c regional -l us-east1 gs://your-bucket-name
Make sure you create the following environment variable.
export BUCKET=your-bucket-name
Please clone the following repository, it contains everything which is needed for this example.
git clone https://github.com/kubeflow/examples.git
Open a Terminal and navigate to the folder /examples/named-entity-recognition/
.