-
Notifications
You must be signed in to change notification settings - Fork 5.8k
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
[Doc] Workspace template examples #32802
[Doc] Workspace template examples #32802
Conversation
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
…example_templates
Signed-off-by: Justin Yu <[email protected]>
…example_templates
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
…example_templates
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
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.
Let's actually do two things for now:
- Don't use BatchPredictor or Checkpoint APIs.
- Instead, use the Dataset ActorPool APIs and callable classes.
We are having some internal discussions on possibly deprecating / unifying BatchPredictor: #32929
doc/source/examples/02_many_model_training/many_model_training.py
Outdated
Show resolved
Hide resolved
doc/source/conf.py
Outdated
@@ -160,6 +160,7 @@ | |||
"_build", | |||
"source/workflows/api/doc/ray.workflow.*", | |||
"source/serve/api/doc/ray.serve.*", | |||
"source/templates", |
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.
maybe source/templates/*
?
Signed-off-by: Justin Yu <[email protected]>
…example_templates
"NUM_WORKERS: int = 1\n", | ||
"\n", | ||
"USE_GPU: bool = True\n", | ||
"NUM_GPUS_PER_WORKER: float = 1\n" |
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.
This cell is repeated twice.
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.
This is how I'm getting around loading a small vs. large version of the template. On product side, we can run some post init command to process the notebook:
- Filter out all notebook cells tagged as
large
. Vice versa for large scale examples. This way, the configurations get set to use the small example defaults.
jupyter nbconvert --TagRemovePreprocessor.remove_input_tags='large'
--to notebook --output batch_inference.ipynb batch_inference.ipynb
How does that sound @ericl?
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.
Can we comment these cells like # This is the large case
, etc?
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.
Yes, sounds good!
"metadata": {}, | ||
"outputs": [], | ||
"source": [ | ||
"preds = predictions.fully_executed()\n", |
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.
"preds = predictions.fully_executed()\n", | |
"preds = predictions.cache()\n", |
The fully executed call is deprecated, use cache() instead.
Signed-off-by: Justin Yu <[email protected]>
Signed-off-by: Justin Yu <[email protected]>
…example_templates
Signed-off-by: Justin Yu <[email protected]>
…example_templates
Signed-off-by: Edward Oakes <[email protected]>
Signed-off-by: chaowang <[email protected]>
Signed-off-by: elliottower <[email protected]>
Signed-off-by: Jack He <[email protected]>
Why are these changes needed?
This PR adds an initial set of examples that can be used as a template for users to get started with certain workloads.
This includes:
For each example, we include:
requirements.txt
for additional dependenciestemplates.yaml
See the contributing guide to see the steps needed to add a new template.
TODOs
To add in a follow-up PR
Related issue number
N/A
Checks
git commit -s
) in this PR.scripts/format.sh
to lint the changes in this PR.