How to Tune LSTM Hyperparameters with Keras for Time Series Forecasting

Last Updated on August 28, 2020

Configuring neural networks is difficult because there is no good theory on how to do it.

You must be systematic and explore different configurations both from a dynamical and an objective results point of a view to try to understand what is going on for a given predictive modeling problem.

In this tutorial, you will discover how you can explore how to configure an LSTM network on a time series forecasting problem.

After completing this tutorial, you will know:

• How to tune and interpret the results of the number of training epochs.
• How to tune and interpret the results of the size of training batches.
• How to tune and interpret the results of the number of neurons.

Kick-start your project with my new book Deep Learning for Time Series Forecasting, including step-by-step tutorials and the Python source code files for all examples.

Let’s get started.

• Updated Apr/2019: Updated the link to dataset.