Get Error from server (NotFound): pods "Error" not found

Followed all your steps (apparently without errors). At step 4 (./4_start_ui.sh), I get the following :

`error: error executing jsonpath "{.items[0].metadata.name}": array index out of bounds: index 0, length 0`
`Error from server (NotFound): pods "Error" not found`

It seems that the kubeflow namespace does not exist:

`kubectl get pods -n kubeflow` gives:
`No resources found.`

`kubectl get pods --all-namespaces` gives:

`NAMESPACE NAME READY STATUS RESTARTS AGE
default deploy-ml-pipeline-sl2mz-glq4g 0/1 Error 0 11m
default deploy-ml-pipeline-sl2mz-lp6wn 0/1 Error 0 17m
kube-system event-exporter-v0.2.4-5f7d5d7dd4-m8r96 2/2 Running 0 18m
kube-system fluentd-gcp-scaler-7b895cbc89-q5xrh 1/1 Running 0 18m
kube-system fluentd-gcp-v3.2.0-9vlqt 2/2 Running 0 17m
kube-system fluentd-gcp-v3.2.0-hlzn4 2/2 Running 0 17m
kube-system fluentd-gcp-v3.2.0-nbflx 2/2 Running 0 17m
kube-system heapster-v1.6.0-beta.1-676c6c6c95-kcwcj 3/3 Running 0 17m
kube-system kube-dns-autoscaler-76fcd5f658-wf6r7 1/1 Running 0 18m
kube-system kube-dns-b46cc9485-5hm6m 4/4 Running 0 18m
kube-system kube-dns-b46cc9485-prklw 4/4 Running 0 18m
kube-system kube-proxy-gke-mykfp3-default-pool-e50b075b-39gj 1/1 Running 0 18m
kube-system kube-proxy-gke-mykfp3-default-pool-e50b075b-6c5q 1/1 Running 0 18m
kube-system kube-proxy-gke-mykfp3-default-pool-e50b075b-xsf7 1/1 Running 0 18m
kube-system l7-default-backend-6f8697844f-vf9nw 1/1 Running 0 18m
kube-system metrics-server-v0.3.1-5b4d6d8d98-rzsw8 2/2 Running 0 17m
kube-system prometheus-to-sd-6h84w 1/1 Running 0 18m
kube-system prometheus-to-sd-ksljn 1/1 Running 0 18m
kube-system prometheus-to-sd-tldmd 1/1 Running 0 18m`

Any help someone? What step is supposed to create the kubeflow namespace?
thanks