Preparing the entrypoint TensorFlow and Keras training script
TensorFlow is a popular open source software library for machine learning. Keras, on the other hand, is a user-friendly high-level neural network library that helps build and train models faster.
In this recipe, we will define a custom TensorFlow and Keras neural network model and prepare the entrypoint training script. In the next recipe, we will use the TensorFlow estimator class from the SageMaker Python SDK with this script as the entrypoint argument for training and deployment. If you are planning to migrate your custom TensorFlow and Keras neural network code from your local machine and perform training and deployment with the SageMaker platform, then this recipe (and the next) is for you!
Getting ready
This recipe continues from Generating a synthetic dataset for deep learning experiments.
How to do it
The instructions in this recipe focus on preparing the entrypoint script. Let's start by creating...
