adding a pytorch SageMaker example with AWS infra parameters

[sermolin]

trafficstars

What does this PR do?

PR shows how to add AWS infra parameters (subnets, security groups, kms keys, internet proxies (via env_var_dict/env parameters).

Fixes # (issue)

Who can review?

Feel free to tag members/contributors who may be interested in your PR.

[review-notebook-app[bot]]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[sermolin]

Adding a notebook that shows how to add AWS infra parameters (subnets, security groups, kms keys, internet proxies (via env_var_dict/env parameters).

[sermolin]

• Why add this notebook here: It is to be used as template for SageMaker users who are deploying similar notebooks in a secure enterprise environment with VPC, subnets, Security Groups, KMS key, and internet proxies. The notebook shows how to specify all of these parameters and how to pass them to SageMaker APIs.
• No, this notebook will not run “as is” and should be excluded from your regression testing suite. While it is possible to create a companion Infrastructure-as-a-code notebook to stand up VPC, subnets, Security Groups, and KMS keys, setting up an EC2 instance with an internet proxy server would be a bridge too far.
• Yes, I will maintain the infrastructure parameters in the notebook should the API calls change. Since the notebook is not intended to run out-of-the-box it is not necessary to keep up with the latest versions of transformers, pytorch, and python.

[sermolin]

Any update on this PR?