[Bug Report] "Getting Started" notebook fails on the first few cells in Sagemaker Studio

**Link to the notebook**

[Link](https://github.com/aws/amazon-sagemaker-examples/blob/main/aws_sagemaker_studio/getting_started/xgboost_customer_churn_studio.ipynb)

**Describe the bug**
Cell [2] has various errors when you attempt to downgrade the Sagemaker SDK

Subsequently, the cell [3] does not complete successfully. It errors with a traceback related to protobuf. "TypeError: Descriptors cannot be created directly." with a suggestion to downgrade protobuf to 3.20.x or lower.

**To reproduce**
1. Launch a Python 3 notebook kernel using Data Science 3.0 image in Sagemaker Studio
2. Run cell [2], then cell [3].
3. Contemplate why the 'getting started' notebook for Sagemaker Studio, which is linked from blog posts and from the official documentation, doesn't just work out of the box.

Open aws_sagemaker_studio/getting_started/xgboost_customer_churn_studio.ipynb in a Python 3 kernel using the Data Science 3.0 image. Run cells [2] and [3] to reproduce the SDK downgrade errors and protobuf traceback; done means the Getting Started notebook's initial cells complete successfully in SageMaker Studio.

Written by the indexing model from the issue text.