Kubeflow Training Operator provides Kubernetes custom resources that makes it easy to run distributed or non-distributed TensorFlow/PyTorch/Apache MXNet/XGBoost/MPI jobs on Kubernetes.
Questions tagged [kubeflow]
433 questions
0
votes
0 answers
Kubeflow pipeline doesnt create any pod; unknown status
I started working with kubeflow and created a first, little pipeline. Unfortunately it doesn't work, so when I try to create a run with my pipeline nothing happens. Neither it creates a Kubernetes pod nor does the status of the run change (it keeps…
Pascal
- 103
- 2
- 7
0
votes
1 answer
Accessing Kubeflow UI by the same domain users
Kubeflow deployed on GKE. then accessing UI is OK.
But the other users of the same domain cannot access to it. It likes below image.
How could I grant a permission to the other users?
Thanks.
Mu-ik Jeon
- 843
- 7
- 15
0
votes
2 answers
Kubeflow pipeline on AWS/On-prem currently feasible?
I’m testing kubeflow pipeline and would like to use it on AWS/On-prem but I saw the below comment on the documentation. Should I wait using it with AWS/on-prem?
Due to kubeflow/pipelines#345 and kubeflow/pipelines#337, Kubeflow Pipelines depends on…
user3368526
- 2,168
- 10
- 37
- 52
0
votes
2 answers
How to set Kubeflow to use preemptible VMs?
I'm trying to set up a new Kubeflow deployment to use preemptible VMs. Referring to this doc, I see that there is some YAML to specify preemptible nodes but I am not sure where to put it in the myriad of YAML files generated by kfctl.sh. Meanwhile,…
ontologist
- 595
- 4
- 20
0
votes
2 answers
Deploying Kubeflow pipelines using cli or API, not on UI
Kubeflow pipelines can be created by uploading on the pipelines UI.
But it's annoying to develop pipelines, repeating some changes and uploading again.
There seems to be no related documents.
Is it possible to deploy Kubeflow pipelines using cli or…
Mu-ik Jeon
- 843
- 7
- 15
0
votes
1 answer
Kubeflow on kubernetes, ksonnet apply function is not reacting properly?
I'm trying the following codelab:
https://codelabs.developers.google.com/codelabs/kubeflow-introduction/index.html
When I want to apply my container to the cluster the shell freezes and after a couple of minutes it returns 'Killed'.
This is the…
Jacob Verschaeve
- 159
- 2
- 10
0
votes
1 answer
Kubeflow: Not able to build Makefile - recipe for target 'presubmit' failed
I want to build kubeflow from source and deploy it in a container.
I used the setup instructions from over here:
https://www.kubeflow.org/docs/started/getting-started/
My resulting docker container looks like this:
FROM ubuntu:16.04
ARG…
0
votes
1 answer
Why don't setup kube-dns in PAI k8s cluster?
I tried to deploy kubeflow in the clusters and found there was no dns service available. I am not sure the reason, could someone explain it ?
Dongqing
- 674
- 1
- 5
- 19
0
votes
2 answers
how can we specify version of tensorflow serving in kubeflow?
I am trying to use tensorflow serving to serve a model. when I try to apply serve component using ksonnet, I see that workload created on kubernetes(gke) is using tensorflow…
rhg
- 173
- 2
- 16
-1
votes
1 answer
CrashLoopBackOff when launching notebook from Kubeflow DashBoard
Launching notebook from kubeflow dashboard using minikube as kubernetes server does not end due to CrashLoopBackOff state. When I run kubectl get po -n kubeflow, there is no error. So I am confused. I already posted issue in kubeflow community…
Ryo Matsuzaka
- 135
- 1
- 1
- 9
-1
votes
1 answer
Encountered error applying application application: (kubeflow.error): Code 500 with message: Apply.Run
I'm trying to deplo kubeflow on AKS using this YAML file https://raw.githubusercontent.com/kubeflow/manifests/v1.2branch/kfdef/kfctl_k8s_istio.v1.2.0.yaml.
But I'm having the error below I could not know how to fix it .
I followed exactly what was…
ELOUTMADI
- 59
- 1
- 6
-1
votes
1 answer
Tensorflow serving object detection predict using Kubeflow
I followed the steps given in this post to deploy my tensorflow model for prediction using GPUs on Google Kubernetes Engine and Kubeflow. I have exposed the service as a load balancer by modifying the YAML file in this way where I changed the type…
user9460641
-1
votes
1 answer
How to deploy jupyterHub by kubeflow
when i 'ks apply default -c jupyter' to deploy a jupyterHub, it always response following errors:
[I 2019-01-09 14:28:52.866 JupyterHub app:871] Writing cookie_secret
to /jupyterhub_cookie_secret [I 2019-01-09 14:28:52.880
…
