Neural Networks with Keras
We'll be using Keras for creating neural networks. Keras is a library like SciKit-Learn, but designed specifically for neural networks.
keras vs tf.keras
Keras used to be a standalone library, but as of September 2019, it is part of Google's TensorFlow library.
Keep that in mind if you're looking at any tutorial that was written prior to that date. Most of the API and functions will be the same, but your import statements will likely be different.
For more information, see this article on the change.
On the TensorFlow Tutorials Page, in the left sidebar, you'll see a section called "Beginner". Under that section you'll see a "ML Basics with Keras" section and a "Load and preprocess data" section.
Tutorials to Review
You'll want to review the following examples in ML Basics:
- Image Classification with Keras
- Regression with Keras
- Overfit and Underfit
- Save and Load
- Tune Hyperparameters with the Keras tuner
And at least the following under "Load and preprocess data":
You might also want to keep this Keras API reference page handy, as well as this collection of Keras guides.