Random Decision Forest

鉴于目前已经在两篇文章里面看到了 random forest 概念：
1）Real Time Action Recognition Using Histograms of Depth Gradients and Random Decision Forests
2）Efficient human pose estimation from single depth images (kinect 工作原理)
决定认真的学习一下Random Forest的概念，并知道如何使用

一个比较好的有关random forest 形象的解释见下链接:
https://www.youtube.com/watch?v=loNcrMjYh64

Step1: suppose we have a matrix of training samples

Step 2: Create several random subsets and generate several random decision trees, and then we will get a random forest!

Step 3: Use each random decision tree to predict the label of the test sample, and finally choose the most reliable one

更理论性的知识以及实际使用以后再补