This workshop will provide you a 2 hours experience into the work of data scientists.
It's a fun hands-on workshop on how to solve a customer churn problem (classification problem) for a cellular company using an XGBoost Algorithm on the SageMaker platform. You'll be completing relevant code segements by checking the API documentation online.
short intro into machine learning and the AWS AI/ML stack, working with hosted Jupyer notebooks as hosted Sagemaker notebooks, Numpy/Pandas API, data preperations, hosted training, parallel model tuning, online inference in production endpoint and batch inference jobs.
basic python. Optionally, visit here to learn a bit on the SageMaker platform: https://www.youtube.com/watch?v=uQc8Itd4UTs&list=PLhr1KZpdzukcOr_6j_zmSrvYnLUtgqsZz
You’ll experience the data scientist methodology and tools, and understand what's required to take an existing use case example, and adjust it to your use case.
Open a SageMaker notebook, and clone the repository.
Students: open the notebook xgboost_customer_churn-student.ipynb, read through it and complete the code marked with REPLACE_ME strings.
Instructor: open the notebook xgboost_customer_churn-instructor.ipynb.