Predict on the survivors of the Titanic using a neural network made with Jax

The last blog post centred around making a Jax neural network that would make predictions on one of sklearn’s toy datasets, being the breast cancer dataset, and that blog post can be found here:- https://medium.com/@tracyrenee61/predict-on-the-breast-cancer-dataset-using-a-neural-network-made-with-jax-6c47ca338bf6

In this blog post, I intend to show how predictions can be made on a dataset produced in a csv file. This particular csv file is very popular, being the Titanic dataset, and the data scientist must predict on the survivors of this famous but ill fated cruise. The Titanic dataset is a binary classification problem, so is written slightly differently from the previous programs I written where the label was of the multiclass variety.

I have created the program using Google Colab, which is a free online Jupyter Notebook hosted by Google. It is a great platform to create programs in Python, but the only real flaw in this program is the fact that it does not have an undo function. Care needs to be taken not to inadvertently overwrite or delete valuable code, however, because if that happens it could be lost forever.

Once I created the program, I imported the libraries that I would need to execute the program, being:-