Support Vector Machines (SVMs)

Hey, there! I am back. I was studying several classification algorithms in the past few days including k-nearest neighbor, random forests, and support vector machines. It was certainly a lot of statistics going through my head, but all worthwhile. For this post, I am going to focus on support vector machines (SVMs). As I am going through this topic, I did my own analysis and added some notes. Hopefully, it will be useful for those who are also interested in this topic.

Quick Recap
  • Used the Iris Plant Database in Python and did some descriptive statistics + visualization 
  • Trained the model using SVM by using both linear and radial basis function kernels (RBF)
  • Clarified the C and gamma parameters when using RBF kernel for classification
  • Run the GridSearchCV to search for the best parameters and then apply them to the test dataset
  • Using confusion matrix and classification report to determine the goodness of the classification

1 comment: