Test your knowledge

In this chapter, we have reviewed a broad range of topics, from taking full advantage of the information that fastai provides about models to making your web deployments available to users outside of your local system. In this section, you will get the opportunity to exercise some of the concepts you learned about in this chapter.

Explore the value of repeatable results

In the Using callbacks to get the most out of your training cycle recipe, you made a call to the set_seed() function prior to training each of the models. In that recipe, I stated that these calls were necessary to ensure repeatable results for multiple training cycles. Test out this assertion yourself by following these steps:

1. First, make a copy of the training_with_tabular_datasets_callbacks.ipynb notebook.
2. Update your new notebook by commenting out the first call to set_seed() and rerun the whole notebook. What differences do you see in the output of fit_one_cycle() between the...