-
Notifications
You must be signed in to change notification settings - Fork 54
update sdk and demos to ray 2.5 #163
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.
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
Conversation
1c33580
to
88c5f4e
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Don't some of the examples need PyTorch installed now since it is no longer present by default in the image? Otherwise this LGTM at first glance.
Testing all the notebooks, and I'm noticing trouble with 2 of them:
I want to try now with with the regular code and see if it's the same there, or maybe I'm doing something wrong... not sure. |
@KPostOffice the notebook image still contains pytorch. It gets installed with the rest of our requirements. |
@jbusche are you using ray 2.5 in your working environment? I had the same issue if I did not use the right version of Ray |
This PR also fixes #170 |
LGTM Michael showed me how to update the Ray version in the notebook:
And then restart any running notebooks All the notebooks worked well after that. Thanks Michael! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Need to substitute an image in a few locations
Looks good Michael, Images pulling fine:
|
I walked through each of the demos using a Ray 2.5 image and updating the driver environment to use Ray 2.5. Everything seemed to work without issue.
This PR updates all the requirements to point to Ray 2.5.0 instead of 2.1.0.
It also updates the default image in the cluster config to "rayproject/ray:2.5.0-py38-cu116"
I also used the image parameter in all of the cluster config calls just to make it more explicit that the image is configurable.
closes #134