Understanding the similarity and difference between Adaboost M1 and Linear Support Vector Machine (SVM) classifiers
-
linear_svm.py: Linear SVM was implemented using sklearn library and used to classify given datasets dataset1 and dataset2. Decision boundaries were plotted with different values of "c" to better understand underlying mechanism in linear svms.
-
adaboost.py: Adaboost-M1 classifier based on linear SVM was implemented by myself without any library to classify data belong to two classes class A and class B. 10-times-10-fold cross validation was used to show that mean and variance of accuracy improved comparing to only linear svm.
For more information, check doc folder.