As mentioned earlier, the application we will be creating to demonstrate logistic regressions is a file classifier. Given a file (of any type), we extract the strings from the file. This is a very common approach to performing file classification although, like the previous example, this is often just an element of file classification, not the only component. Therefore, don't expect this to find the next zero-day piece of malware!
The completed project code, sample dataset, and project files can be downloaded here:https://github.com/PacktPublishing/Hands-On-Machine-Learning-With-ML.NET/tree/master/chapter03_logistic_regression.
The trainer used in this application also uses SDCA but using the logistic regression variation that was discussed earlier in this chapter.
As in the previous example, we will begin by exploring the project architecture, diving into the code, and then show how you can run the example to both train and predict.
...
