-
Notifications
You must be signed in to change notification settings - Fork 1.2k
doc: JumpStart Utility Doc Update #3866
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: JumpStart Utility Doc Update #3866
Conversation
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
@judyheflin Can you fix these doc8 errors:
Finally submit your changes after successfully running |
Yes will do -- this PR is still a WIP. |
doc/overview.rst
Outdated
@@ -725,7 +724,35 @@ retrieve the related image as follows. | |||
instance_type="ml.m5.xlarge", | |||
) | |||
|
|||
Deploy a Pre-Trained Model Directly to a SageMaker Endpoint | |||
Low-code Deployment | |||
=================== |
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.
How about moving Low-code Deployment
section before built-in Components
section?
That way, users would don't want to dive deeper into the underlying mechanics can stop reading after the Low-code Deployment
section and try it out.
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.
Restructured the order based on this feedback
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
|
||
.. code:: python | ||
|
||
from sagemaker import model_uris | ||
question = "What is Southern California often abbreviated as?" |
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 use a more generic payload, like hello world!
? I don't want to set the expectation that this is the kind of payload that should be given.
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.
Yep, can do -- will make both of these more generic
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.
Could you check with @khetan2 please?
Since this is a generative model, I feel like we should ask it a question.
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 payload was the example for another model, so I'm not sure it's appropriate here
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.
After thinking about this more and discussing with Ashish, FLAN-T5 XL is currently our #1 recommended model for question/answering, so this example prompt is fine here. GPT-J (the other example model) can also be used for question/answering, but is more flexible for additional use cases. We can mention that GPT-J can be used for things like code generation, summarization, and more, but the q&a example will still be applicable. In general, customers can explore our documentation, the pretrained models table, and example notebooks to find appropriate use cases for each model.
and https://github.com/aws/sagemaker-scikit-learn-container. Use | ||
the ``model_id`` and ``version`` of the corresponding model to | ||
retrieve the related image as follows. | ||
To check valid data input and output formats for inference, you can use the ``retrieve_options()`` method |
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 also mention the content type and accept type utilities?
|
||
.. code:: python | ||
|
||
question = "What is Southern California often abbreviated as?" |
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.
same question here about the payload
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
973e8ec
to
d970f91
Compare
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
AWS CodeBuild CI Report
Powered by github-codebuild-logs, available on the AWS Serverless Application Repository |
Description of changes: Added documentation for low-code JumpStart model deployment and fine-tuning. Included JumpStartModel, JumpStartEstimator, and utility APIs in the docs.
Testing done: Staged on local build successfully and passed tox checks.
Merge Checklist
Put an
x
in the boxes that apply. You can also fill these out after creating the PR. If you're unsure about any of them, don't hesitate to ask. We're here to help! This is simply a reminder of what we are going to look for before merging your pull request.General
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.