IKH

Improving Model Performance – II

Till now, you’ve been looking at the scree-plot to choose the number of components that explain a certain amount of variance before going for the dimensionality reduction using PCA. Now, there is a nice functionality which makes this process even more unsupervised. All you need to do is select the amount of variance that you want your final dataset to capture and PCA does the rest for you. Let’s take a look at the following demonstration to see how we can do the same.

As you saw above, all you needed to do was select a particular amount of variance that you want to be explained by the Principal Components of the transformed dataset. PCA automatically chooses the appropriate number of components on its own and proceeds with the transformation. This again saves us a lot of time!

Report an error