Use a regression neural network made in numpy to predict on house prices

The last several neural networks that I have covered in my blogs have concerned classification models. In this blog post, however, I am going to discuss a regression model of a neural network.

The difference between a regression neural network and a classification neural network is the fact that there is no activation unit at the end of the regression neural network, so the input parameter flows out of it.

A regression neural network is a problem where a model is used to predict continuous outcomes or values, such as forecasting salary changes, house prices, or retail sales.

The program that I have written in this instance is a neural network that will make predictions on Boston house prices. I have used Jason Brownlee’s house price dataset because it is no longer available as a toy dataset in Python’s machine learning library, sklearn.

I have written the program in Python using Google Colab, which is a free online Jupyter Notebook hosted by Google. Google Colab is a great platform to use to write code in Python, but its only main drawback is the fact that it does not have an undo function, so care needs to be taken not to accidentally delete or overwrite valuable code.