Probabilistic Graphical Models (week 9)

Week 9—Learning with Incomplete Data
In PA8, you learned tree-structured Bayesian networks to represent body poses. In this assign-
ment, you will be using the network you learned for human poses from PA8 to perform action
recognition on real-world KinectTMdata. The KinectTMdata was collected by performing actions
and extracting the human pose data associated with each action. We have collected data for the
following three actions: “clap”, “high kick”, and “low kick”. Your goal in this assignment is to
build a model that is able to identify the action performed.

最后一次作业！
楼主最近忙毕设，week8还没学，估计很久之后更新了。。各位加油~

===============
第八周作业做完~虽然没有完全搞明白吧lol 发现第九周作业更晕@@. 以前都学过一遍EM了看到公式还是各种不懂- -|

===============
看了几个例子，EM基本搞明白了，programming assignments估计要纠结一段时间了= =

===============
花了一个多小时把第一问搞定了，第二问好复杂。。HMM虽然已经很熟悉了，但是操作起来还是捉襟见肘。。加油之。。

===============
终于还是把所有问解决了，第二问应该是集大成者，第三问我把第二问refactor了一下顺利解决~
不得不说这次作业超级有用，搞定之后对整个PGM都有了全新的认识！

最后一问我的程序有问题，但懒得改了……

顺便纪念一下第一次Online Course完成~