How it works…

We hit the ground running in this recipe by training our neural network straight away in step 1. Several additional flags are available to customize training, depending on our needs. Now that we've trained our model, we need to output a list of 100,000 passwords, all of which have been generated by the model (step 2). These serve as intelligent guesses of likely passwords. By examining the output of step 2, we can see that the passwords appear as follows:

Now, we can use these as candidates for cracking passwords.

Get Machine Learning for Cybersecurity Cookbook now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.