Summary

In this chapter, we explained how you can train a ML model in a notebook in SageMaker Studio. We ran two examples, one using SageMaker's built-in BlazingText algorithm to train a text classification model, and another one using TensorFlow as a deep learning framework to build a network architecture to train a sentiment analysis model to predict the sentiment in movie review data. We learned how SageMaker's fully managed training feature works and how to provision the right amount of compute resources from the SageMaker SDK for your training script.

We demonstrated SageMaker Experiments' ability to manage and compare ML training runs in SageMaker Studio's UI. Besides training with TensorFlow scripts, we also explained how flexible SageMaker training is when working with various ML frameworks, such as PyTorch, MXNet, Hugging Face, and scikit-learn. Last but not least, we showed you how SageMaker's Git integration and notebook-sharing features can help...