Some implementation of the book of statistical learning method by lihang
- implementation of linear classification method perceptron, including the stochastic gradient descent and show the change of the hyperplane, 2-D case, w is initialized to [1, 1], b is 1
Problem:
- The result will be different if we change the order of the data points when checking whether a point is classified right or wrong
Converge:
-
Convergence can be guaranteed when the dataset is linearly separable
-
implementation of kdtree creation and find nearest neighbor code TODO:
-
find k nearest neighbor code