This is the second tutorial on support vector machines (SVMs) and the 26th in the series of machine learning with scikit-learn. In this video we learn about the kernel trick.
I have to admit, out of all the tutorials in this series, this is one of those that I spent a lot of time working on and researching about. Kernels, in my view, could not be easily understood maybe because of the mathematics behind them.
But, in my view, the mathematics of kernels are simply beautiful. However, I'm not going to go into the maths in this tutorial. Rather, I explain kernels and the kernel trick conceptually. I also use some visuals (graphics) that may further help grasping down the concepts.
Ultimately we're gonna apply a kernel - the radial basis function kernel (RBF) - with an SVM on the cancer dataset in scikit-learn. But first things first, kernels...
As a reminder:
In this series I'm going to explore the cancer dataset that comes pre-loaded with scikit-learn. The purpose is to train the classifiers on this dataset, which consists of labeled data: ~569 tumor samples, each labeled malignant or benign, and then use them on new, unlabeled data.
Previous videos in this series:
@cristi Cristi Vlad, Self-Experimenter and Author
