00:01
For this problem, there are a few key things that we need to note.
00:03
We're told that we found that the average r -squared value on the training data was 0 .5, then 0 .99.
00:33
So, we need to be careful here, because, well, when we're doing a linear regression model, when we're doing a machine learning model, we split things up into the training data and the testing data to make sure that we can actually make meaningful predictions using our model.
00:51
So, we want to test, we want to find r -squared on the test data, not the training data.
01:07
One of the things that i need to note here, too, is that a one -hundredth order polynomial transform is basically a lot of, that is a lot of terms in our polynomial transformation.
01:32
If you're familiar with the concept of a taylor series, basically the idea is that any set of data can be approximated using a polynomial transformation, or using a polynomial function, if you have enough terms...