[MUSIC] Okay, so let's summarize what
we've learned in this module. And these notions that we've
learned here are really, really important to machine learning. Well beyond just this regression task. So this was a really important
module on how to assess performance that we're gonna see again and again. Throughout this machine
learning specialization. In particular,
we described what a loss function is and we gave different examples
of loss functions. We also talked about differentiating
between notions of training error, true and generalization error,
and test error. And we talked about how to think about
splitting our data into training and test sets. And in order to both fit our model
parameters as well as to estimate our generalization error. And we also talked about
how to think about three notions of error that
come into our predictions. And finally, we talked about this idea of choosing the model complexity as
a key component of our work flow. But how to think about doing
that by introducing this notion of a validation set. And using this to allow us to
choose model complexity and then to assess our performance
on a separate test set. [MUSIC]