Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Proposal: Official Jupyter Images for Kubeflow #52

Closed
jlewi opened this issue Dec 20, 2017 · 4 comments
Closed

Proposal: Official Jupyter Images for Kubeflow #52

jlewi opened this issue Dec 20, 2017 · 4 comments
Labels
area/jupyter Issues related to Jupyter lifecycle/stale

Comments

@jlewi
Copy link
Contributor

jlewi commented Dec 20, 2017 •
edited
Loading

Opening this issue to discuss providing one or more official Kubeflow Jupyter images

Some things to consider:

  • What libraries/frameworks/tools should be in the image
    • TensorFlow?
    • GRPC client libraries?
    • xgboost?
    • scikits?
    • ksonnet? helm?
  • Do we want to produce a small number of fat images or many smaller images?
  • Can we reuse existing images? e.g. TensorFlow docker images? Jupyter images?

Related:
#50 Docker images for TF Serving
#37 Size of our Jupyter images

/cc @flx42 @yuvipanda
@flx42
Copy link

flx42 commented Dec 20, 2017

Unless we realize that we need many tools, a benefit would be that we can make the official TF images lighter by removing Jupyter: tensorflow/tensorflow#15284

@zomglings
Copy link
Contributor

A good place to start, from the point of view of what a data scientist could reasonably expect in their environment, is Kaggle's Python image.

If size is a concern, as per Issue 37, this is probably not satisfactory as I think their image weighs in at around 10GB. If the intention is to support more than just TensorFlow, I don't see this as being a problem, though.

@zomglings
Copy link
Contributor

One problem that Kaggle has not yet had to contend with is how to handle GPUs.

This is a problem that Datalab has solved by offering a standard image and a GPU image.

@jlewi jlewi added the area/jupyter Issues related to Jupyter label Feb 27, 2018
kimwnasptd pushed a commit to arrikto/kubeflow that referenced this issue Mar 5, 2019
@avdaredevil @k8s-ci-robot
Adding apverma@google.com (github handle: avdaredevil) to Kubeflow Eng (
31827fd 
kubeflow#52)

* Update org.yaml

* Added to members

* Update release-team.members.txt

* Add apverma@

* Fixed alphabetic ordering
@jlewi
Copy link
Contributor Author

jlewi commented Apr 15, 2019

/lifecycle stale

@jlewi jlewi closed this as completed Apr 15, 2019
yanniszark pushed a commit to arrikto/kubeflow that referenced this issue Feb 15, 2021
@YujiOshima @k8s-ci-robot
Fix modelsave (kubeflow#52)
56e143a 
* refactor Model API

Signed-off-by: YujiOshima <yuji.oshima0x3fd@gmail.com>

* fix ModelStore IF name

Signed-off-by: YujiOshima <yuji.oshima0x3fd@gmail.com>

* fix model save bugs

Signed-off-by: YujiOshima <yuji.oshima0x3fd@gmail.com>

* avoid nil pointer

Signed-off-by: YujiOshima <yuji.oshima0x3fd@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
area/jupyter Issues related to Jupyter lifecycle/stale
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@jlewi @flx42 @zomglings @k8s-ci-robot