Uploaded September 2018 | Updated November 2026, 3 hours ago
To get the most from this series on deep neural networks for domain experts, start watching it from the beginning youtube.com/watch?v=9-QYsN_knG4&list=PLsu0TcgLDUiIKPMXu1k_rItoTV8xPe1cj
Now that you are familiar with tfruns (TensorBoard) to compare changes to your model, I want to dig a bit deeper into the Keras sequential model. In particular, this video introduces callbacks and changing default values in optimizers.
The callback will monitor the learning and terminate it early should no progress be made. This avoid lengthy training runs.
The code is available on GitHub at github.com/juanklopper/Deep-learning-using-R
To get the most from this series on deep neural networks for domain experts, start watching it from the beginning youtube.com/watch?v=9-QYsN_knG4&list=PLsu0TcgLDUiIKPMXu1k_rItoTV8xPe1cj
Now that you are familiar with tfruns (TensorBoard) to compare changes to your model, I want to dig a bit deeper into the Keras sequential model. In particular, this video introduces callbacks and changing default values in optimizers.
The callback will monitor the learning and terminate it early should no progress be made. This avoid lengthy training runs.
The code is available on GitHub at github.com/juanklopper/Deep-learning-using-R





![Deep learning for regression problems using Mathematica
READ THE FULL DESCRIPTION...
In this tutorial I show you how create a deep learning model for a regression problem using the Wolfram Language in Mathematica.
If you would like to take a course to learn more about the Wolfram Language, check these out:
https://www.udemy.com/mathematica/
https://www.udemy.com/mathematica-for-statistics/
A regression problem means that the target variable is numerical in nature. We will use data in a spreadsheet file that might exist on your internal drive.
During this tutorial you will learn how to import the data, how to split it into a training and a test set, how to bring all the feature variables into the same scale to assist the gradient descent algorithm, and how to format the data to be used in the model.
You will also learn how to create a densely connected deep neural network to predict a numerical variable as opposed to a class (as with classification problems).
Lastly you will learn how to evaluate your model given the test data. There is also a bonus on the use of the Predict[] function.
The notebook and spreadsheet file is available on GitHub at https://github.com/juanklopper/deep_learning_using_the_Wolfram_Language Deep learning for regression problems using Mathematica](https://i.ytimg.com/vi/9qISQKugyoA/mqdefault.jpg)




